WO2025111309A1

WO2025111309A1 - Portable accelerated fatigue testing system

Info

Publication number: WO2025111309A1
Application number: PCT/US2024/056615
Authority: WO
Inventors: Onome Scott-Emuakpor; Philip Johnson; Aaron WEARREN
Original assignee: Hyphen Innovations
Current assignee: Hyphen Innovations
Priority date: 2023-11-21
Filing date: 2024-11-20
Publication date: 2025-05-30
Anticipated expiration: 2026-05-21

Abstract

A system and method for performing an accelerated fatigue test on a material sample. The system defines a portable accelerated fatigue testing system that employs a multi-phase approach to perform pre-test parameter setup and response discovery, generate actual fatigue data during an actual test and correlates the generated data to actual strain levels of the material sample. By providing a phase lock loop, an excitation frequency can be maintained in substantial synchronization with a resonant response frequency even in situations where changes in compliance or stiffness of the material sample undergoes changes as a result of cracks or related damage being formed in the material sample during the fatigue test. In one form, the material sample is made using an additive manufacturing process such that the system and method can gain insights into the particular modes, patterns and responses of the sample associated with the additive manufacturing process.

Description

PORTABLE ACCELERATED FATIGUE TESTING SYSTEM

This application claims priority to U.S. Provisional Application Serial No. 63/601,296 that was filed on November 21, 2023 and the entirety of which is incorporated herein by reference.

TECHNICAL FIELD

The present disclosure relates generally to a system for performing material and sample testing, and more particularly to a system that is portable to provide low-cost accelerated fatigue and failure testing of such materials and samples using optical images.

BACKGROUND ART

Additive manufacturing (AM) forms components and related materials through computer-controlled deposition and joining of successive layers of material by fusion or related techniques. When using an AM-produced component, it is important to ensure that it has repeatable and reliable structural-properties to account for material property variabilities and other process-induced artifacts that may not be present in traditional component manufacturing approaches. One way to do this is to generate extensive material datasets that traditionally are costly and time-intensive.

Material testing systems include, among others, impact testers, static testers, fatigue and fracture testers, tension and compression testers, universal testers and torsion testers. Of these, fatigue testing involves determining how such a component or sample exhibits failure-prone behavior in response to an extremely large number (typically between 10⁷ and 10⁹) of cyclic tension, compression, flexure, torsion or related loads where such loads are maintained at a level within the elastic limit of the material or component being tested.

Traditional fatigue testing systems that employ either axial-based modes of operation or rotational-based modes of operation require complex equipment such as motors, load frames with heavy bases, movable crossheads and other servo-based hydraulic equipment. Moreover, given the load range of such equipment (often capable of delivering up to 20,000 pounds of load) and need for precise alignment of a test sample, the chances of causing significant harm or damage to an operator, the specimen or both is significant if extreme caution is not exercised. The relatively low frequency (for example, no more than about 50 Hz) of servo-hydraulic loading that is inherent in linear hydraulic-based excitation precludes any form of accelerated fatigue testing, while motor- based rotational vibration excitation provides no more than 100 Hz to 200 Hz of excitation and are also limited by concerns over instrumentation reliability and loading control. Given the previously-noted number of cycles needed for typical fatigue testing, significant investments in time and energy use under these conventional approaches are required. This problem is more acute for modern manufacturing processes where variability in the underlying components and their material properties make it necessary to generate voluminous amounts of data so that designers of such components better understand how systems that incorporate such components perform.

DISCLOSURE OF INVENTION

With the foregoing in mind, the authors of the present disclosure have developed a system that is capable of accurately predicting fatigue properties in a short amount of time. Significantly, not only does the system determine if a particular component or material is susceptible to failure by fatigue, but it is also able to pinpoint the location where such failure may occur with enhanced accuracy. Moreover, the portability of the system, coupled with using computer-based statistical inference models, makes it more compatible with accelerated testing procedures where potential failure modes related to product aging, use rate or stress levels can be generated more quickly than under traditional fatigue testing approaches.

According to an aspect of the present disclosure, a portable accelerated fatigue testing system is disclosed. The system includes a test signal source; an amplifier in signal communication with the test signal source; a support configured to secure a material sample, an excitation source coupled to the support and in signal communication with the amplifier, a sensor in signal communication with the support and a computer system that includes a processor, digital conversion hardware in signal communication with the sensor a display and non-transitory computer readable storage comprising computer executable instructions. Thus, the processor carries out the computer executable instructions to output first and second graphical user interfaces to the display. The first graphical user interface receives a user-generated input command that characterizes at least one parameter of a control signal. In addition, the processor carries out the computer executable instructions to control the test signal source to output the control signal according to a vibratory profile that is based upon the user-generated input command that corresponds to a particular accelerated fatigue test regimen. The amplifier generates an excitation drive signal responsive to the control signal from the test signal source, while the excitation source imparts a vibration profile comprising at least the excitation drive signal from the amplifier into the support responsive to the excitation drive signal from the amplifier and the support imparts vibratory energy corresponding to the vibration profile into the material sample. In addition, the processor carries out the computer executable instructions to obtain digital sensor data from the digital conversion hardware corresponding to a vibration pattern that is generated in the material sample in response to receipt of the vibratory energy, measure a deflection of at least a portion of the material sample using the obtained digital sensor data, correlate the deflection to a predetermined fatigue limit of the material sample and output to the second graphical user interface an expected fatigue characteristic of the material sample.

According to another aspect of the present disclosure, a method of performing a fatigue test on a material sample is disclosed. The process includes configuring a portable accelerated fatigue testing system to generate a vibratory profile, then inducing, during a first phase of the fatigue test, a vibratory response in the material sample using the vibratory profile, after which the vibratory response is converted into a first output that corresponds to a peak response of a free edge of the material sample. Subsequently, during a second phase of the fatigue test, a test command is generated for an actual test of the material sample, fatigue test limits and actual fatigue data. In addition, during a third phase of the fatigue test, an image of the free edge of the material sample is acquired while the material sample is in an unexcited state, as is a calibrated image corresponding to the free edge of the material sample. Moreover, an input command corresponding to sampling parameters to be used is acquired along with free edge displacement of the material sample. In addition, an output is generated that corresponds to deformations at the free end that are based on a difference between the unexcited state and an excited state that is produced during the actual test of the material sample. BRIEF DESCRIPTION OF DRAWINGS

The following detailed description of specific embodiments of the present disclosure can be best understood when read in conjunction with the following drawings, where like structure is indicated with like reference numerals and in which:

FIG. 1 depicts a simplified view of a portable accelerated fatigue testing system along with a notional visual display of a vibratory response of a material sample according to one or more embodiments shown or described herein, while a notional display screen shows a comparison between an undeflected and deflected state of the material sample;

FIG. 2A depicts the fixed cooperation between the material sample and the exciter in one form factor;

FIG. 2B depicts a detailed view of the fixed cooperation between the material sample and the exciter of FIG. 2A;

FIG. 3 A depicts a flowchart showing how the portable accelerated fatigue testing system acquires material sample response data;

FIG. 3B depicts a flowchart showing how the portable accelerated fatigue testing system uses the material sample response data that was generated in FIG. 3 A and applies that to generating actual fatigue test data;

FIG. 4 depicts a flowchart showing how the portable accelerated fatigue testing system correlates the acquired data from FIG. 3B and converts it into a strain level of the material sample under an actual test condition;

FIGS. 5A through 5C depict images of the material sample under various test and analysis conditions;

FIG. 6 depicts a fatigue test conducted with the portable accelerated fatigue testing system of FIG. 1 ;

FIGS. 7A through 7D depict different clamping conditions used in finite element analysis displacement and strain predictions; FIGS. 8 A through 8C depict the results of finite element analysis displacement and strain predictions performed on three different plate materials each under two different clamping conditions;

FIG. 9A depicts a graphical representation of the displacement predictions of FIGS. 8 A through 8C; and

FIG. 9B depicts a graphical representation of the strain predictions of FIGS. 8 A through 8C.

It will be appreciated that for the sake of clarity, elements depicted in the drawings are not necessarily to scale, and that certain elements may be omitted from some of the drawings. It will further be appreciated that certain reference numerals may be repeated in different figures to indicate corresponding or analogous elements.

MODES FOR CARRYING OUT THE INVENTION

The authors of the present disclosure have discovered that one technological difficulty to overcome relates to how to perform accelerated testing of material samples (also referred to herein as a testing sample, testing coupon, specimen or the like) in general and AM-produced material samples in particular, as well as for building a reduced-order predictive model (such as those associated with machine learning (ML) and artificial intelligence (Al)) for material properties. The authors of the present disclosure have further discovered that one technological difficulty to overcome relates to how to perform small-scale testing of material samples that replicates the accuracy of larger machines that typically employ complex electromechanical, servo-electric, linear motion or servo-hydraulic modes of excitation. The Applicants have further discovered that while shortening the amount of time it takes to understand the types and causes of failure of a material sample may be achieved by accelerated life testing, subjecting such sample to extremes in temperature, voltage, vibration, pressure or the like, as well as varying the rate in which changes to one or more of these inputs, is difficult to reliably and repeatably achieve using low-cost, portable testing equipment. They have further discovered that one specific form of accelerated life testing (fatigue testing, where cyclic loads are applied to the material sample) is particularly difficult to attain with small-scale testing machinery. By providing a compact, portable platform with which to acquire one or more of displacement, velocity and acceleration data in response to controllable levels of vibratory force, frequency and duration, as well as to adjust input excitation of such vibrations in response to dynamic changes in the vibration response, the authors of the present disclosure are able to evaluate a material sample. In one form, time domain vibratory data is acquired and converted into frequency domain data, such as through a Fast Fourier Transform (FFT) or other Fourier series-based approaches, depending on the level of exactitude needed to analyze the response frequencies of the material sample. In this way, both the severity of the vibration at a point in time as well as problematic response frequencies or modes associated with such vibration can be identified. Such information may then be provided to component or system designers to ensure that potentially dangerous operating conditions are avoided.

One interesting method related to vibration-based techniques is referred to the extraction of the modal parameters from vibration data, where particular emphasis is placed on acquiring the natural frequencies and dampening ratios of the most important modes of vibratory’ response. In the present context, such natural frequencies — which are characteristics of the material, the geometric properties and other parameters of the material sample — are also known as eigenfrequencies in that the frequency with which it vibrates takes place absent any driving force. In a related way, the material sample will vibrate (for example, sinusoidally) in a normal mode at its natural frequency. Furthermore, if the force that is being generated by the portable accelerated fatigue testing system 100 takes place at the natural frequency (at which the amplitude of the motion of the material sample is greatest), such frequency is referred to as its resonant frequency. In one form of modal parameter extraction, the material sample is configured as having a thin through-the-thickness dimension along with generally flat height and width dimensions having varying degrees of aspect ratio as a way to simulate both thin, elongate blade-like members and wide-chord plate-like members. Such material samples may be sized and shaped to generally coincide with fan blades, turbine blade or other relatively flap-shaped component that is subjected to repeated, time-varying loads.

In one form, these modal parameters (and especially the natural frequencies) can be determined through algorithms that are configured to perform a modal analysis, such as a rainflow processing algorithm (such as that available through MathWorks’ MATLAB library). In another form, the results obtained from a fatigue test may be compared to analytic predictions of the modal behavior, such as those from a finite element model (FEM) or related topological design that uses finite element analysis (FEA) or related simulations to characterize the shape and related geometric properties of the material sample for certain mode shapes as a precursor to a larger fatigue analysis where frequency ranges and generated stresses are produced. As such, the modal analysis or analytic prediction may be used to generate a predetermined fatigue limit that corresponds to a maximum stress level that can be endured over an infinite number of loading cycles without expecting a fatigue failure. Within the present context, fatigue testing is that which subjects a material sample to repeated loading and unloading to evaluate how it will perform over time, whereas fatigue analysis (such as through statistical-based means, such as various fracture mechanics or stress or strain life models) is performed on the data that is acquired during a fatigue test to determine the endurance of the material sample relative to the range of stresses applied to such sample. Thus, fatigue tests may be used to generate fatigue life data, crack growth data or the like, as well as to identify where within the material sample fatigue-related problems may occur, while a corresponding fatigue analysis determines when the material sample performs satisfactorily when subjected to a particular type of cyclic loading. With particular regard to the fatigue life data, it can be used to predict various stresses such as mean stress, maximum stress, stress ranges, minimum expected stress, stress amplitudes or the like in order to conduct a fatigue analysis. As such, fatigue testing is a significant component of a fatigue analysis as it involves subjecting the material sample to cyclic loading and measuring the resulting damage attributable to the fatigue such that the resulting data may be correlated to known responses.

In one form, a combination of the test output and the FEM may be used to create a particular mode shape and it associated vibration pattern that is produced in response to a particular vibration profile that has been determined to correlate to such mode shape. Within the present disclosure, it will be understood that when an object (such as the material sample discussed at length herein) vibrates in one or more natural modes (such as first bending, second bending, first torsion, chordwise or the like) of vibration, certain readily-identifiable vibration patterns (specifically, standing wave patterns) may be formed. In this way, the use of the term “profile” is generally meant to correspond to inputs used by the portable accelerated fatigue testing system, while the terms “pattern”, “mode” and “response” are generally meant to correspond to outputs produced in the material sample. For example, the vibratory profile corresponds to the parameters that are arranged in a particular way and used by the portable accelerated fatigue testing system as part of a particular accelerated fatigue test regimen to ensure that one or more vibration-based modes, patterns or responses are produced in the material sample.

It will likewise be appreciated that while much of the subsequent discussion within this disclosure pertains to thin plate-like material samples, the portable accelerated fatigue testing system disclosed herein may be used on other samples as well, such as rods, tubes or pipes that exhibit shaft-like attributes, in addition to beams, cylinders, spheres and other shapes that may generally correspond to an actual structural component. These, as well as using the portable accelerated fatigue testing system to analyze means of manufacture or fabrication (most notably, by AM) of these shapes of material samples are within the scope of the present disclosure.

It will be appreciated that the dominant modes of vibrational response are different for high-aspect ratio samples than it is for low-aspect ratio ones. For example, elongate, relatively low-width samples may exhibit dominant first bending, second bending and torsional (that is to say, twisting) modal responses, while more plate-like samples with wider chords and lower aspect ratios may exhibit significant edgewise tip bending. Moreover, these may take place at very different vibratory frequencies, where for example first and second bending tend to occur at relatively low frequencies, while edgewise bending may occur at higher frequencies. Factors that can be correlated to vibration-induced fatigue of the sample include (among other things) the level of vibration (that is to say, amount of displacement) and the rate of such vibration, as well as the degree of dampening within the sample. As discussed elsewhere, it will be appreciated that the frequency response of the material sample may change over the course of the test, such as through various dampening effects, as well as in situations where microstructural or other changes take place within the material sample.

Regardless of the mechanism of modal response, one significant factor to take into consideration is any potential change in a resonant response frequency as the sample undergoes internal stiffness or compliance changes, such as due to a microstructural rearrangement, microcrack initiation, localized slip band formation, macroscale sample deformation or the like that can give rise to stress concentrations, stress relaxation or the like. Significantly, changes in the response frequency may be used to indicate that damage onset or other forms of material sample degradation is progressing. By way of example, response frequency changes in the form of a response phase may be used to identify and track the progression of a microcrack or other form of damage or change in structure. In one form, response frequency changes can be correlated to resonant frequency changes, such as when a degree of internal dampening is known. This and other forms of material sample degradation in turn may be used to provide a more particular understanding of life expectancy of the material sample. From a component design perspective, of particular import is understanding how these stiffness or compliance changes may be correlated to changes to the natural response frequencies and mode shapes of a plate, shaft or associated component being simulated, particularly as it relates to placing the material sample closer to a resonant condition and the ensuing more rapid accumulation of fatigue damage. This enhanced understanding may be used in conjunction with or in place of FEM or other response prediction models as a way to give a designer of a corresponding component a better sense of operational conditions that avoid resonant frequencies. By way of example, for a component that is expected to be exposed to significant rotational-based vibration (such as those experienced by shaft, rotors and blades of a gas turbine engine), the results of the testing may be conveyed to a user in visual form, such as through a Campbell Diagram or the like. Likewise, identification of particular vibrational modes and their associated patterns may be presented in the form of holographic images, photographic representations or the like.

By way of brief introduction, according to aspects herein, a fatigue test of a material sample may be determined by a portable accelerated fatigue testing system. In one aspect, the system uses a multi-phase fatigue test where a one or more particular modes of vibratory response and corresponding vibration patterns are excited within the material sample, a vibratory profile is used along with fatigue test limits and an actual test is performed to generate fatigue data and images of the material sample in both an undeformed (unexcited) state and deformed (excited). This is used in conjunction with the vibratory profile to gain fatigue- related insights into the material sample. In some implementations, the material sample is fabricated using AM so that the fatigue-related impact of artifacts within the sample that are unique to AM can be better understood.

In some implementations, the particular vibration pattern being excited is a chordwise bending mode. Likewise, the actual test on the material sample is performed until the material sample fails, such as through the detection of one or more cracks. To increase the speed with which the actual test is performed, higher-frequency modes of response (such as the chordwise bending mode) are excited, with frequencies between roughly 3000 Hz and 5000 Hz.

Referring first to FIG. 1, the portable accelerated fatigue testing system 100 is shown for performing tests on the material sample 1. It will be appreciated that although high-cycle fatigue (HCF) testing is discussed in detail in the present disclosure, there are numerous other forms (including creep, low-cycle fatigue, thermo-mechanical fatigue, crack propagation and growth, fracture toughness, high strain rate, stress-relaxation and others) that are also within the scope of the portable accelerated fatigue testing system 100.

The portable accelerated fatigue testing system 100 includes an excitation (that is to say, servo-hydraulic loading) source 110, at least one sensor 120, an amplifier 130, digital control hardware (also referred to as a test signal source, in either event, to act as a signal generator) 140, network circuitry 150 and a computer system 160. In one form, some or all of these components are disposed or otherwise situated on a platform 170 that defines dimensions (including size and weight) that permit ease of movement, setup and transport. In one form, the platform 170 may include wheels 171 for ease of movement. Significantly, the portable accelerated fatigue testing system 100 does not include a load frame or other large, heavy, generally immobile equipment that is commonly associated with commercially-available fatigue testing systems as previously discussed. In an optional form, the portable accelerated fatigue testing system 100 includes a controllable heating device (also referred to as a heating source) 180 that can be placed in thermal communication with the material sample 1, thereby permitting the acquisition of temperature- related properties of the material sample 1 , including thermal conductivity, thermal expansion and specific heat.

In operation, the portable accelerated fatigue testing system 100 may be operated to directly measure a response that may occur to the material sample 1. By way of example, such response may be in the form of vibration amplitude or related displacement that can be visualized, such as on the computer system 160. Likewise, the portable accelerated fatigue testing system 100 may be operated to indirectly detect conditions where the material sample 1 may be prone to fatigue-related damage by identifying vibratory size, frequency or other para eters of interest and correlating them to certain bending, torsion or other modes of response, where such correlation may be performed by either a priori (that is to say, using a rules-based algorithm) or ad hoc (such as through ML) means either of winch may also be implemented on the computer system 160. In one form, the portable accelerated fatigue testing system 100 may be configured to operate autonomously or in response to a user input in order to perform an intended fatigue test or analysis. In this way, an automated methodology is created that allows continuously collected data to identify key fatigue metrics, including the identification of damage-producing operational conditions where a combination of resonant frequencies and sufficient vibrational energy may lead to fatigue-based failure.

Although the computer system 160 is presently depicted as a laptop computer, it will be appreciated that other suitable form factors may also be used and which are within the scope of the present disclosure. Other exemplary forms may include those based on tablets or related handheld devices, desktop computers, minicomputers, mainframes, personal computers (PCs), as well as distributed architectures, including network or cloud-based variants of the foregoing, whether based or servers, thin clients, thick clients or serverless. For example, in one form, the computer system 160 may be configured as a stationary on-site or off-site stationary device, while in another form to be miniaturized such that it may take on a wearable form factor, such as being affixed to the individual, such as through wrist-mounted, arm-mounted, torso-mounted, leg or ankle-mounted configurations, as well as others. In still another form factor, the computer system 160 may be configured to be in the cloud. Signal communication within or by the computer system 160 may be through wired or wireless means where the latter may include known long-range or short-range approaches such as mobile telephony, WiFi, Bluetooth, nearfield communications or the like. Relatedly, as shown, the computer system 160 depicts an autonomous (that is to say, stand-alone) unit; as will be appreciated by those skilled in the art, in one form it may be the part of a larger network such as those encountered in cloud computing, where various computation, software, data access and storage services may reside in disparate physical locations. Thus, in one form (not shown), all components of the computer system 160 need not be located on-board the platform 170 or other relevant support of the portable accelerated fatigue testing system 100, such as those configurations associated with cloud computing. Such a dissociation of the computational resources does not detract from the computer system 160 (including when it is acting as a controller) being within the scope of the present disclosure. Although not shown, it will be appreciated that one or more of the components that make up the portable accelerated fatigue testing system 100, such as the sensor 120, amplifier 130 and digital control hardware 140, may be embodied in a singular package (such as the computer system 160). Whether the control and operation of the portable accelerated fatigue testing system 100 arises out of a distributed or integrated computing environment — both of which are within the scope of the present disclosure — will be apparent from the context. In one form, these and other components that are used for the control and operation of the portable accelerated fatigue testing system 100 may be commercial off-the-shelf (COTS) components that are software configurable, such as dues to set, detected or otherwise received values from laser vibrometers, accelerometers or the like, as well as from phase angles between a measured response (such as a laser response) and a control signal.

The computer system 160 may function as a control circuit that may be used for implementing the various processes described herein, including where the computer system 160 forms a part of such control circuit. In one form, the computer system 160 includes one or more microprocessors (pP) for executing instructions that make up a computer program and corresponding (hardware) memory 162 (for example, random access memory and/or read only memory) that are connected to a system bus 163. Information can be passed between the system bus 163 (via a suitable bridge 164) and a local bus 165 that is used to communicate with various input/output (I/O) devices 166 or peripherals. For instance, the local bus 165 is used to interface peripherals with the one or more microprocessors 161. In one form, these peripherals (which collectively or individually may be referred to as computer-program products) may include storage devices (such as hard disk drives, removable media storage devices such as flash drives, DVD- ROM drives, CD-ROM drives, floppy drives or the like (all of which may resemble memory 162 in terms of the ability to store machine code or data). The present list of peripherals is mentioned by way of illustration and is not intended to be limiting, and as such other peripheral devices may be suitably integrated into the computer system 160. In configurations where large amounts of image data is being collected, additional specialty-purpose processors (such as graphics processing units (GPUs) or other accelerators) or their equivalent may also form part of the computer system 160. In particular, these GPUs or other accelerator processors could be used if the large amount of image data is subjected to certain ML models such as a convolutional neural network (CNN) as part of establishing an ML inference. In one form, the computer system 160 allows the portable accelerated fatigue testing system 100 to have fully automated operation. For example, robotic- based actuators (not shown) that are responsive to control signals being issued by the computer system 160 may be made to impart the instruction signal to the excitation source 110. As will be discussed in conjunction with FIGS. 3A, 3B and 4, the control signal is one that corresponds to a particular accelerated fatigue test regimen.

In one form, the I/O devices 166 may be integrally packaged or as separate components and include input devices such as a keyboard 166A, scanner, mouse or the like, as well as output devices such as a display (also referred to herein as a display screen or a monitor) 166B, printer, user interface and network adapter. Although the display 166B is shown as part of the computer system 160, it will be appreciated that it — as well as additional or alternative screens — may be used to depict results of the fatigue test, whether as part of another system, and that all variants of how to display test results are within the scope of the present disclosure. With particular regard to the keyboard 166A and display 166B it will be appreciated that they are typically available on a computer, laptop, mobile telephone, smart watch, graphical user interface (GUI) or other related data-reporting device. With even greater regard to the display 166B and related output devices, they are configured to cooperate with the computer system 160 such that a program stored in memory 162 and executed by the one or more microprocessors 161 as machine code can convey to a user an indication of the level of deflection or other indicia corresponding to the fatigue test or a corresponding fatigue analysis. In one form, the user interface is configured as a remote user interface such that is separate structure physically decoupled from the rest of the computer system 160. In such case, signal communication between the remote user interface and the computer system 160 is exclusively through a wired or wireless signal communication protocol. In another form, the user interface is configured as structure that is part of the computer system 160 by being either integrally formed as a part thereof or in signal communication therewith. As with the remote user interface, such signal communication may be through either a wired or wireless communication protocol. Distinctions as to whether the user interface is remote from or a part of the computer system 160 may be based on numerous factors that will be apparent from the end use application; such factors may include the degree of physical or structural coupling, the use of common operating system software, the use of common power source or the like. Within the present disclosure, in one form the term GUI may include the setup of icons, links or other input and output that takes place on display 166B in order to give the display 166B its enhanced I/O functionality. In another form, the term may refer to an embodiment of the display 166B that is configured to actively permit the direct exchange of user input to and user output from its screen. Relatedly, a first GUI is one that receives a user-generated input command that characterizes at least one parameter of a control signal that corresponds to a particular accelerated fatigue test regimen, while a second GUI is one that is depicted on the display 166B, showing — for example — an expected fatigue characteristic of the material sample 1. In one form, the expected fatigue characteristic of the material sample 1 comprises an expected fatigue life. Thus, the first GUI that is being output to the display 166B corresponds to the control signal that is generated during a first phase (which will be discussed in more detail in conjunction with FIG. 3 A) of a particular fatigue test (which may include one or more particular vibratory profiles) to perform pre-test parameter setup and response discovery in order to determine a particular vibration pattern of the material sample 1. Relatedly, the second GUI that is being output to the display 166B corresponds to results of a second phase (which will be discussed in more detail in conjunction with FIG. 3B) of the fatigue test to depict actual fatigue data that is unique to the particular material sample 1. It will be appreciated that these and other usages will be apparent from the context.

The one or more microprocessors 161 control operation of the exemplary computer system 160. Moreover, one or more of the microprocessors 161 execute computer readable code (for example, stored in the memory 162, or the various forms of storage previously discussed, that instructs the one or more microprocessors 161 to implement the computer- implemented processes herein. Within the present context, the microprocessor 161 — whether in singular or multiple, distributed form — is understood to encompass variations, such as a microcontroller (including system-on-chip (SoC) varieties) that itself forms an integrated circuit that also includes memory and one or more peripherals such that reliance on separate memory (such as memory 162), I/O devices 166 or other components discussed herein is reduced or eliminated. It will be appreciated that if a configuration of the computer system 160 is particularly adapted to utilize one particular form of the microprocessor 161, such will be apparent from the context. Thus, the exemplary computer system 160 or components thereof can implement one or more of processes on the memory 162 as well as one or more computer-readable storage devices as set out in greater detail herein. It will be understood that other computer configurations may also implement the processes and/or computer-implemented processes stored on the one or more computer-readable storage devices as set out in greater detail herein. Computer-program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages. The program code may be executed either entirely or partly on the computer system 160. In the latter scenario, a remote computer may be connected to the computer system 160 through any type of network connection, for example, using the network adapter 169 of the computer system 160, or by conveying signals using wired or wireless versions of the network circuitry 150.

Within the present disclosure, the memory 162 is understood to form a part of a computer-readable medium such that in implementing computer aspects of the present disclosure, any combination of computer- readable medium may be utilized. The memory 162 (as well as other forms of computer-readable storage devices) is a tangible device or related piece of hardware that can retain and store a program (including operating instructions or other forms of computer readable program code) for use by or in connection with an instruction execution system, apparatus or device including, for example, a computer or other processing device set out more fully herein. Notably, a computer-readable storage medium does not encompass a computer-readable signal medium. Thus, a computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves through a transmission media. Thus, unlike a transitory propagating signal per se, which will naturally cease to exist, the contents of the computer-readable storage device or computer- readable hardware that define the claimed subject matter persists until acted upon by an external action. For instance, program code loaded into random access memory (RAM) 162A is deemed non-transitory in that the content will persist until acted upon, for example, by removing power, by overwriting, deleting, modifying or the like. Moreover, since hardware comprises physical element(s) or component(s) of the corresponding computer system 160, hardware does not encompass software per se. The various components that make up the computer system 160 may be configured as a controller that may contain program code, machine codes, native instruction sets, computer readable instructions or related data structures such that upon loading into the memory 162, the program code is particularly configured to execute one or more steps in a manner consistent with the methods disclosed herein. Within the present disclosure, the program code will be understood to include the organized collection of instructions and computer data that make up particular application software and system software the latter of which may include operating system software and basic input/ output that relates to the operation of the computer system 160, regardless of its form factor. This and other software (such as system software or application-specific software) provides programmed instructions that may be implemented on the one or more of the microprocessors 161 to allow it (or them) to interact with the computer system 160 or other computer-based equipment in order to perform one or more of the data acquisition, processing, communicating, analysis and related functions disclosed herein. For example, source code may be converted into executable form as machine code for use by the one or more microprocessors 161; such machine code is predefined (or configured) to perform a specific task in that it is taken from a machine language instruction set known as the native instruction set that may be part of a shared library or related non-volatile memory (including memory 162 and any removable media versions of the computer-readable storage devices) that is specific to the implementation of the one or more microprocessors 161 and its (or their) particular Instruction Set Architecture (ISA). As such, software instructions such as those embodied in the corresponding portion of the machine code configure the one or more microprocessors 161 to provide the program structure and associated functionality as discussed herein. The computer-implemented processes herein may be in the form of a machine-executable process executed on the computer system 160, or on any other structure described more fully herein. As with some of the previously-discussed components, the program code and related software used for the control and operation of the portable accelerated fatigue testing system 100 may be COTS.

In one form, a data-containing portion of the memory 162 — also associated with volatile working memory — is referred to as RAM 162A, while an instruction-containing portion of the memory — also associated with permanent or non-volatile memory — is referred to as read only memory (ROM) 162B. Thus, it will be appreciated by those skilled in the art that computerexecutable instructions can be placed within an appropriate location (such as the aforementioned memory 162) within the computer system 160 in order to achieve the objectives set forth in the present disclosure. In one form, the computer system 160 may additionally include additional chipsets (not shown) for peripheral functions. In addition to the control logic, program code or related instructions or algorithms, memory 162 may be configured to store object detection logic, object recognition logic, as well as auditory or visual indicia-generation logic, all as described in more detail elsewhere in this disclosure. In one form, software used to perform the analyses discussed herein may include actuator control software. As previously noted, this and other forms of software may be implemented as machine code that is stored in memory 162 and operated upon by the microprocessor or microprocessors 161.

Upon having program code means loaded into memory 162 in general (and in one form into ROM 162B in particular), the computer system 160 becomes a specific-purpose machine configured to perform one or more operations in a manner as described herein. As such, the computer system 160 becomes a particularly-adapted computer or computer-related data processing device that employs the salient features of such an architecture in order to perform at least some of the data acquisition, processing, communicating, analysis and related functions discussed herein.

Referring next to FIGS. 2 A and 2B, views of a test setup are shown. The excitation source 110 is in one form an electrodynamic shaker (also referred to herein as an electrodynamic exciter) in general and a permanent magnet-based shaker in particular. Unlike servo-hydraulic shakers that may be limited in their frequency response (for example, to no more than 200 to 300 Hz for rotational-based excitations and even lower (for example, no more than about 50 Hz) for linear-based excitations), when the excitation source 110 is configured as a permanent magnet shaker, it may produce vibrations of up to 3000 Hz, and in some cases up to 5000 Hz when driven by a suitable increase in voltage from the amplifier 130 upon an instruction signal that is received from the computer system 160 through the digital control hardware 140 over the network 150. This increase in excitation frequency, coupled with sufficient input force, significantly contributes to the accelerated fatigue testing described herein, as testing may be completed in as little as one two-hundredth of the time it takes to perform a fatigue test using conventional means. Moreover, this allows a significant decrease in size and weight of the excitation source 110, enabling the portable accelerated fatigue testing system 100 to assume a tabletop profile. This has additional benefits in that complex, high-powered electrical systems (such as three-phase, 480 volt infrastructure) and motors or other actuators powered by such systems may be replaced with simpler vibration drivers using one-phase or mains power. In addition to reducing the size, weight and complexity of a fatigue testing system, it reduces the need for specialized, highly -trained users. In one form, the excitation source 110 excite along a single Cartesian axis, while in another it may function as a multi-axis excitation source. Moreover, in one form the excitation source 110 may be an exciter based on a transducer, a speaker, a piezoelectric exciter or an electrodynamic exciter.

The material sample 1 is secured to the excitation source 110 through a support or fixture in the form of clamping blocks 112. It will be appreciated that although two clamping blocks 112 are presently shown creating a sandwich-like clamping of the material sample 1, greater or fewer numbers of clamping blocks 112 are also within the scope of the present disclosure. Likewise, it will be appreciated that in configurations where there is an upper and lower clamping block 112 (as can be seen in FIG. 2B), the lower clamping block 112 may either be formed separately or as part of the excitation source 110. Vertically-alignable through holes may be formed in the material sample 1 and clamping blocks 112 so that one or more threaded bolts 114 or related fastening mechanism may secure them to the excitation source 110 that in one form converts the control signal that is input from the amplifier 130 and computer system 160 into vibratory motion such as being responsive to an electromagnetic field that is either provided by a separate shaker or in one that is integral with the excitation source 110. In one form, movement of the excitation source 110 that is imparted to the material sample 1 may resemble that of a haptic motor. It will be understood that the type of excitation source 110 will be dictated by various factors, including the type of fatigue test being run, or by the nature of the component or material sample 1 being tested, and that all such versions are within the scope of the present disclosure. One or more accelerometers 116 are affixed to the excitation source 110 to record the vibrations during the fatigue test. Although not shown, it will be appreciated that — depending on the nature of the test and the need to keep the accelerometer 116 from influencing any vibratory response in the material sample 1, the accelerometer 116 may be mounted on one of the clamping blocks 112, bolt 114 or directly on the material sample 1. By using the accelerometer 116, control of a base excitation is easier to maintain, such as through constant acceleration. This input from the accelerometer 116 in turn will allow the fatigue test to be operated without having to impart a constant voltage into the excitation source 110 that would otherwise cause changes in acceleration as a range of possible response frequencies is swept. This in turn makes it easier to perform a so- called “apples-to-apples” comparison across different types of tests or different materials with a common test.

As shown with particularity in FIG. 2A, the material sample 1 is in the form of a cantilevered beam specimen, while others (such as those subjected to tension, compression, torsion, 3 -point bending, 4-point bending or the like, none of which are shown) may also be made to respond to the excitation source 110 through a suitably-configured version of the clamping blocks 112. It is within the scope of the present disclosure that actual or simulated components may also be tested in lieu of simple geometric material sample 1 shapes.

The authors of the present disclosure have determined that rather than applying fatigue cycles by exciting a conventionally-shaped material sample, the excitation frequency can be reduced by 30% or more by creating a sample with accelerated fatigue-testing features. In one form, the material sample 1 may have — in addition to apertures sized to accept one or more bolts 114 therethrough — an aperture 2 defined therein for increasing compliance while reducing mass relative to conventional tests, which both increases the ease to generate failure strain amplitudes more quickly (thereby reducing the number of testing cycles required) while increasing the loading frequency.

Additional consideration beyond the size alone may also be used to promote the desired vibratory responses, vibration patterns or the like. For example, the material sample 1 may be optimized by having a test section with an appropriate thickness in order to generate enough flexure that in turn can lead to high strain and subsequent fatigue failure. In particular, if the test section of the material sample 1 is too thin, a more powerful shaker or related excitation source 110 would be needed in order to perform the test. Because higher-powered shakers tend to have either very low operating frequencies (which would lead to prohibitively long tests) or require lengthy lead time (often 24 months or more) for development (which would lead to prohibitively expensive). With this in mind, the thickness and aperture design of the material sample 1 are specifically configured to achieve high flexure for high strain amplitude under low energy input excitation at very high testing frequencies. In one form, the aperture 2 may be sized or shaped in order to promote a particular form of resonant responses. By way of example, if the material sample 1 is configured as the aforementioned wide- chord plate-like component (which in turn may be used to generally mimic the fan blade of a gas turbine engine), one or both of the size and shape of the aperture 2 — as well as the specimen thickness — may be made to make it easier to preferentially excite the aforementioned edgewise tip bending (also referred to as a “lyre mode” or chordwise bending mode) of vibratory response where predictions of such mode may be algorithmically determined in advance, such as through the aforementioned FEM. As can be seen, the material sample 1 depicted herein is meant to generally mimic rotating blades used in gas turbine engines where fatigue failure often occurs under either combined bending and twisting movement or high order bending modes (such as the chordwise bending mode) both of which tend to produce short wavelength, high frequency stress states. As noted elsewhere, these high frequencies require the excitation source 110 to produce vibrations of up to approximately 3000 Hz, and in some cases up to approximately 5000 Hz. In addition, the significance of trying to excite the chordwise bending mode is to ensure that fatigue failure occurs away from the region of the material sample 1 that is being clamped. Moreover, unlike data that is collected in response to conventional axial tests (also referred to as uniaxial tests), the insights gained from the determination of multiaxial stresses over a wide range of loading frequencies such as those encountered in wide-chord rotating turbomachinery more closely correlates to one or more of higher-order bending modes, combined bending and twisting modes or the like all of which tend to generate short-wavelength stresses at much higher excitation frequencies. As such, a user-generated input command may be entered into the portable accelerated fatigue testing system 100 in order to output a control signal according to a vibratory profile that will preferentially generate the multi-axis chordwise bending mode (or any other mode of interest), depending on which particular accelerated fatigue test regimen is desired. In one form, vibratory energy that is imparted into the material sample 1 from the excitation source 110 and through the clamping blocks 112 that corresponds to a vibratory profile (for example, sinusoidal, random, swept or the like) is performed during a fatigue test. Within the present disclosure, the vibratory profile is used as part of an HCF analysis to help understand how the material sample 1 responds to cyclic loads that are imparted to it through the excitation source 110.

In one form, the configuration and the attachment approach of the material sample 1 is part of a particular specimen design, where through proper design of the material sample 1 as an integral part of the portable accelerated fatigue testing system 100 rather than as a mere workpiece that merely conform to a generalized set of testing system requirements, conventional concerns over sample alignment, user error, user harm and other difficulties associated with large material testing systems can be avoided. Moreover, in one form the material sample 1 is one monolithic (that is to say, unitary) piece. As such, it does not require a specialized fixturing approach. Because it supports higher frequencies than conventional bending specimens, the amount of material required to form the material sample 1 is roughly the same as that for standard rotating bending and load controlled axial fatigue tests.

Results demonstrated by the authors of the present disclosure reveal significant reductions in fatigue test time. For example, when using a conventionally-shaped sample that is approximately 4.5 inches along its chordwise dimension and made from various different aluminum, titanium, nickel and steel alloys, the excitation at about 1600 Hz was achieved using an electrodynamic shaker base. In another sample that was created by the authors of the present disclosure, a hybrid plate that is approximately 3 inches along its chordwise dimension was excited at 1100 Hz. This hybrid sample represented a roughly 90% reduction in sample volume compared to the conventional sample. Contrarily, using material sample 1 with an approximately 2.2 inch width along its chordwise dimension and using the shaped aperture 2 achieved excitation at about 4300 Hz and resulted in a greater than 99% reduction in test time compared to using a rotating bending fatigue test (ISO 1143) operating at approximately 33.3 Hz, all without the use of strain gages, precise axial alignment, heavy fixtures (the equipment weight to perform the test weighed less than 50 pounds), crush zones or high-powered voltage source. In one particular example using an aluminum alloy 6061-T6, the accelerated fatigue test of the present disclosure performed in 5 hours what it took an ISO 1143 rotating bending fatigue test using a conventional sample of the same material 99 days to perform. Even considering any additional preparation work on the material sample 1 , the authors of the present disclosure realized a 92% reduction in test time.

Referring again to FIG. 1 in conjunction with FIGS. 5A through 5C, because it is important to know the displacement that takes place within the material sample 1 in order to measure strain, some form of displacement measuring must be made using sensor 120. While various forms of sensors 120 are contemplated to be used in conjunction with the remainder of the portable accelerated fatigue testing system 100 (and therefore deemed to be within the scope of the present disclosure), the sensor 120 chosen for use by the authors of the present disclosure is an optical sensor 120 that provides non-contact measurement of the deflection, movement or related surface topography of the material sample 1 in response to vibrational or related cyclic energy that is imparted to the material sample 1 from the excitation source 110. In such form, the optical sensor 120 provides a means for collecting such deflection or movement data without the need to use extensometers, strain gauges, accelerometers or other contact-based means of detection that could have an impact on the dynamic response of the material sample 1. It will be appreciated that as an option the portable accelerated fatigue testing system 100 may supplant or supplement the information being acquired by the optical sensor 120 with contact-based measuring means or other external sensors, and that all such forms, as well as variations thereof, are within the scope of the present disclosure. In one form, the sensors 120 form part of a laser measurement system.

In this way, a correlation between the fatigue data produced by a fatigue test performed by the portable accelerated fatigue testing system 100 may be shown. Referring with particularity to FIGS. 5A through 5C, as can be seen on the display 166B, an edge-on view of the material sample 1 can be seen, including in one form (in FIG. 5C) with a superposition of the material sample 1 in both an unexcited state and an excited (that is to say, deformed) state 1 ’. In configurations where the optical sensor 120 is the aforementioned active device, various interference signals may be used, including those acquired through holographic interferometry (wherein the optical image comprises a holographic image) or other known means. It will be appreciated that other forms of visual display of the results of the fatigue test may be depicted on the display 166B, depending on the needs of the user. For example, HCF data may be presented as a stress-to-failure (S-N) curve where a control signal using a sinusoidal (or other time-varying or other swept) vibratory profile and associated vibratory load S being applied by the portable accelerated fatigue testing system 100 may be used to generate the number of cycles to failure N. In any event, in one form, the fatigue test may be performed based on such visual identification of abnormal variations in the shapes or natural frequencies of the material sample 1 that appear on the display 166B in response to excitation from the portable accelerated fatigue testing system 100. As previously noted, such an analysis may be based on algorithms or ML approaches.

In one form, the optical sensor 120 is merely a passive receiver of visible- range light emanating from the material sample 1, while in another it can be an active device that provides its own source of illuminating light to the material sample 1. In this latter embodiment, perturbations of a signal that is initially transmitted by the optical sensor 120, reflected off of the material sample 1 and received by the optical sensor 120 may be correlated to such deflections or movement of the material sample 1, such as through holographic or related interferometric (fringe pattern) comparison of the signals to produce a corresponding interferometric or holographic image. In one form of this active embodiment, the light source used to illuminate the material sample 1 is coherent rather than diffuse. As such, the displacement of the vibrating material sample 1 may be detected by the optical sensor 120 that is in the form of a laser vibrometer. For example, the coherent light may be laser light, such as through a gas-based laser (such as a helium-neon laser or argon laser), a liquid-based laser (such as a dye laser), a solid-state laser (such as a ruby laser or a neodymium-doped yttrium aluminum garnet laser) or a semiconductor laser (such as those based on gallium nitride, indium-gallium-arsenide or the like). It will be appreciated that the choice of source of the coherent light will be dictated by the nature of the test being performed, where for example static testing has different illumination requirements than those involved in dynamic testing. When configured as a laser vibrometer, the optical sensor 120 can measure a single point at a time, while measurement of several points on the material sample 1 may be used to calculate a strain.

Although not shown, deflection, vibration or related information being acquired by the optical sensor 120 may be supplemented from data from other sensors (such as the aforementioned strain gauges), as well as through other means, such as finite- element or other numerical-based approaches. Moreover, it will be appreciated that in embodiments where the optical sensor 120 is an active (that is to say, with its own light source) device, it may be integrated into a singular, unitary package or be configured as a distributed construction of one or more of its components. As such, the transmitted light signal (whether coherent or diffuse) emanating from the optical sensor 120 may either substantially coincide with its receiving portion or originate at some remote location. An example of this latter configuration may resemble the well-known Michelson configuration where partially-silvered mirrors (not shown) may be used. It will be appreciated that either of these embodiments, as well as variants thereof, are within the scope of the present disclosure.

Significantly, the optical sensor 120 may provide the input data necessary to provide a visual depiction of the vibration-induced displacement in the material sample 1 , such as that shown on display 166B. In one form, the information being acquired by the optical sensor 120 is timedomain data such that it is subsequently converted into frequency domain data (via FFT, for example) by the computer system 160 or related component. In another form, the optical sensor 120 may possess its own computational ability to convert such time domain data into frequency domain data, also through FFT. In a related manner, the optical sensor 120 may provide input optical image data that through a series of feedforward and backpropagation steps (that is to say, convolutions) extracts the region or regions of vibrational interest and converts them into feature maps necessary to provide an ML-based vibration pattern detection analysis. One example of the aforementioned ad hoc approach may include one or more classification models that are based on the aforementioned CNN that through the use of shared weighting recognizes the internal convolutions inside so that a composite of images can be stitched together at their edges as part of a geometric relationship. Such an approach is particularly configured for converting image data input and performing related computer vision tasks. Unlike other ML (or even a priori, for that matter) approaches, CNNs may find the important vibration pattern features without the need of going through a feature extraction or identification step, where such features are used as input to correlate measured response quantities, such a vibration amplitude or frequency, of the material sample 1. In this way, computational resources of the computer system 160 may be used more efficiently. As with the vibration-induced displacement in the material sample 1, the output of such a classification may be depicted on the display 166B. As previously discussed, the optical sensor 120 may form an integral part of the computer system 160 rather than being a separate component.

In one form, the amplifier 130 is a servo amplifier to ensure precise feedback-based control. In such case (and as will be discussed in conjunction with FIGS. 3A, 3B and 4), it can be made to implement a parametric-based accelerated fatigue testing model that in one form may be implemented as machine code using suitable parameters in order to produce a desired inference, including fatigue analysis, based on a fatigue test. In operation, the amplifier 130 is signally displaced between the computer system 160 and the excitation source 110 in order to boost the voltage that is transmitted from the computer system 160 in order to excite a particular vibration pattern, thereby effecting a more potent driving force into the support 112 and material sample 1. As previously discussed, the amplifier 130 may form an integral part of the computer system 160 rather than being a separate component. In one form, the digital control hardware 140 includes a digital signal generator to convert to an analog test signal using conversion circuitry (for example, digital-to-analog or analog-to-digital) in order to produce one or more varying input signals in accordance with a particular accelerated fatigue testing model. In this way, it can function as an input-output coordinator for one or more of the amplifier 130 and one or more sensors, including optical sensor 120. The produced signal becomes the basis for the control signal (which undergoes digital-to- analog conversion within the computer system 160). The control signal is conveyed over the network circuitry 150 to ensure that a desired amount of load force or power (such as through input voltage), waveform, frequency ranges and related parameters consistent with a particular accelerated fatigue test regimen is provided in the form of the instruction signal to the excitation source 110. In one form, the varying input signal may be sinusoidal or a related form of variable cyclic loads where such variability may be through amplitude, frequency or the like. As previously discussed, the digital control hardware 140 may form an integral part of the computer system 160 rather than being a separate component. Within the present disclosure, a particular accelerated fatigue test regimen is that which includes a set of operational parameters to ensure that energy being imparted to the material sample 1 by the portable accelerated fatigue testing system 100 contains one or more of a desired frequency, voltage level and duration to promote a particular vibratory mode of response, such as a chordwise bending as discussed frequently herein, or other desired mode.

The cooperation of the optical sensor 120, the amplifier 130 and the digital control hardware 140 provides servo-based control of the fatigue test by the portable accelerated fatigue testing system 100. The data being acquired by the optical sensor 120 is used to create stress-life or strain- life curves.

In one form, the network circuitry 150 is formed from wired connections, although it will be appreciated that wireless forms of connectivity, using a suitable wireless signal communication protocol, may also form part or a substantial entirely of the network circuitry 150.

Referring next to FIGS. 3A, 3B and 4, the fatigue test may be performed in a series of steps that may be generally grouped into three discrete phases 300, 400 (which are both shown respectively in FIGS. 3A and 3B) and 500 (which is shown in FIG. 4 along with representative response images in previously-discussed FIGS. 5A through 5C). Together, these phases make up an accelerated fatigue testing regimen that may be implemented in a set of operator-initiated instructions that can be carried out in an automated way by the computer system 160 for the portable accelerated fatigue testing system 100. Each of these phases will be discussed sequentially as follows.

Referring with particularity to FIG. 3A, in the first phase 300, the portable accelerated fatigue testing system 100 performs pre-test parameter setup and response discovery, such as to determine at what combination of input excitation level and input frequency or frequencies will cause the material sample 1 to likely undergo significant vibratory response. In a first step 310, a controller (such as that which may be implemented in the computer system 160) is initialized for subsequent data acquisition activities. In one form, input commands are entered in order to instruct the portable accelerated fatigue testing system 100 how to conduct this pre-test parameter setup and response discovery. These parameters include those to an input voltage, the accelerometer 116, one or more external sensors (such as optical sensor 120, as well as others (not shown)) to receive vibratory response data, input waveform shape (for example, sinusoidal) and a frequency range to be set to or otherwise swept. One or more of these parameters may be used to define one or more vibration patterns that in turn forms a portion of the control signal that is input to the excitation source 110 through the network circuitry 150 from the amplifier 130, digital control hardware 140 and computer system 160 and which corresponds to an intended (that is to say, particular) accelerated fatigue test regimen.

In one form, knowledge of the swept range may be acquired from a previously- conducted FEM analysis or other predictive analytic approach. By way of example, if a user has advance knowledge that based on the size, shape and constituent material of the material sample 1, a particular modal response (such as edgewise bending) will likely take place, an initial input (that is to say, driving) frequency may be set to a value that generally coincides with a known resonant response of the corresponding mode. In such case, the amplifier 130 provides a sufficient voltage to initiate a sinusoidal input and corresponding sample movement necessary to excite the expected mode. In another form, if a precise resonant response frequency is not known, the input frequency coming from the digital control hardware 140 can be oscillated or otherwise swept over a user-defined range. As noted elsewhere, the digital control hardware 140 may include an analog- to-digital converter (not shown) to transform continuous signals into discrete ones, while corresponding filtering or related smoothing may also be implemented. Natural frequencies that correspond to certain modes of vibration are swept to identify one or more relevant frequencies where resonant conditions and the associated vibration patterns may be sensed. By conducting slower sweeps (such as a sinusoidal sweep), more precise displacements along the free edge of the material sample 1 may be detected. Furthermore, multiple displacements along the free-edge of the material sample 1 may be acquired from one sweep. From there, phase identification and phase tracking control may be applied to the material sample 1 until a suitable condition (for example, specimen breakage) is achieved.

In a second step 320, at least two channels are receiving data: the accelerometer 116 and the external sensor (such as optical sensor 120). The data being acquired (such as through a frequency range sweep) arrives as time domain data, as is the voltage that is used to drive the excitation source 110. At this stage, the test is being performed in a simplified manner. This has the effect of creating data sets that pair to the corresponding inputs.

In a third step 330, frequency response curves may be generated. As previously noted, this involves converting the time domain data into frequency domain, such as through FFT or the like so that the spectral content (that is, the distribution of the response over frequency) of the material sample 1 may be better understood. By converting time data to frequency data, phase tracking (which is associated with the following step 340) is enabled, as it is easier to capture the phase crossover at zero that coincides with peak amplitude.

In a fourth step 340, peak responses (that is to say, where high levels of vibration amplitude (displacement) are detected. It will be understood that another rationale for acquiring accelerometer 116 data is to understand a base prior to generating resonant response data. This in turn will improve the ability of the portable accelerated fatigue testing system 100 to correspond frequencies and responses to a particular strain amplitude. In other words, by running a relatively tight frequency sweep to accurately measure a phase difference between an actual response and the data coming from the accelerometer 116, data errors are reduced, thereby allowing the system to gain a more precise knowledge of the frequencies that correspond to the excitation of certain vibratory modes. This tracking acts as a phase lock loop in order to maintain resonance within the material sample 1 in order to keep it vibrating without having to drive the shaker (excitation source) 110 too hard, especially as it relates to maintaining the input amplitude. This feedback-based approach allows the portable accelerated fatigue testing system 100 to adjust input frequencies to maintain the phase relationship. It will be appreciated that the closeness with which this relationship may be maintained can be through various approaches, such as upon a triggering response event, drifting away from a defined frequency band or the like. Once the parameter setup and response discovery of FIG. 3 A has been completed, the fatigue test can proceed to the activities set forth in FIG. 3B. In one optional form, the shaker may be operated for a long enough period to ensure actual breakage among certain material samples; this in turn allows for some so-called real-world correlation between as-tested samples and their predicted fatigue properties.

Referring with particularity to FIG. 3B, in the second phase 400, fatigue data generation under an actual fatigue test takes place. More particularly, once a particular vibratory profile is used to generate one or both of the strain amplitude and the stress amplitude of the material sample 1 in response to the pre-test parameter setup and response discovery that takes place in the first phase 300, the second phase 400 is used to perform an actual fatigue test on the material sample 1. As discussed herein, at least one of the first and second phases 300, 400 are controlled by varying at least one of numerous parameters, such as acceleration, input power and excitation frequency.

In a first step 410, the actual fatigue test is controlled through strain amplitude that in turn is determined by the strain displacement data being received from the optical sensor 120. Thus, from the acquired displacement data, the strain can be determined that in turn is fed to the computer system 160 in real time or near-real time to control the actual test that, within the context of the present disclosure, corresponds to subjecting the material sample 1 to alternating vibrations from the excitation source 110 until the material sample 1 either (a) fails due to HCF or terminates upon attainment of either (b) a certain number of vibratory cycles or (c) a user-initiated command. This strain correlates to the vibrational response, even though acceleration and frequency may change slightly. As previously noted, using a phase lock loop helps to adjust frequency as necessary (that is, to ensure that the input frequency and the frequency of the vibratory response remain substantially synchronized) in order to ensure that peak input energy is most effectively being converted into vibratory response within the material sample 1. Thus, if fatigue in the material sample 1 causes the best frequency (resonant frequency) to vibrate the sample, then the frequency driving the material sample 1 may be dynamically changed to ensure that maximum vibration-exciting energy is continuing to be imparted into the material sample 1. The test is further controlled through a schedule as a way to maintain timing while determining both strain amplitude and the number of cycles to failure of the material sample 1. In one form, the schedule may be, as previously noted, 10⁷ or 10⁹ cycles and correlated to the response frequency and duration of the test. In another form, such as where fatigue trends may be of interest rather than cycles to failure, it will be understood that a lower numbers of cycles, such as 10⁴ or 10⁵ may be run, depending on the need.

In a second step 420, limits of the fatigue test may be set. In one form, continuous vibration may be imparted to the material sample 1 at one or more resonant frequencies. In this way, any changes in such frequency or frequencies of the material sample 1 due to events taking place during the test (such as the internal stiffness or compliance changes that may arise out of the aforementioned microstructural rearrangement, microcrack initiation, sample deformation or the like) can be analyzed by the computer system 160. Thus, for example, when changes to the material sample 1 occur as a result of the fatigue test (such as growth of a crack or some other internal or surface change), feedback into the computer system 160 allows the portable accelerated fatigue testing system 100 to adjust frequencies, voltages or other test parameters to account for the decrease in time of expected material sample 1 failure. In a related way, when the portable accelerated fatigue testing system 100 and the results of the test are used as part of a larger fatigue analysis, the computer system 160 may make predictions regarding remaining life of the material sample 1. As such, the portable accelerated fatigue testing system 100 acts as a dynamic system to rapidly change its operation in response to changing test circumstances.

By way of example, the portable accelerated fatigue testing system 100 may detect if a resonant response frequency changes within the material sample 1 during a portion of the actual fatigue test. If so, then the input command frequency can be changed. Furthermore, a phase lock loop as previously discussed or related feedback mechanism may be used to ensure that the input frequency of the input command and the resonant response frequency of the material sample substantially retain synchronization (that is to say, are sufficiently close to one another such that the excitation source 110 substantially continues to excite the material sample 1 at this changed resonant response frequency). In this way, the fatigue test may proceed in an automated and accelerated manner as the number of accumulated vibrations remains as close to the driving frequency as possible.

In one form, the impact on stiffness or compliance changes within the material sample 1 as a result of damage incurred during the fatigue test may be better understood as the response phase that correlates changes away from the frequency or a particular resonant condition as the material sample 1 weakens or undergoes other such structural changes. In one form, increasing the vibration amplitude being input, such as through changes in amplifier 130 settings that in turn increase the excitation (that is to say, driving) voltage, is a way to understand these response changes in amore controllable, parametric manner. Thus, in situations where the response frequency drops (such as when cracking causes a lower in stiffness of the material sample 1), the driving force (that is to say, voltage) may need to be increased in order to maintain the responsive vibratory displacement amplitudes relatively constant. From here, measurements are taken in order to ascertain differences between input and response. These and other quantities (such as acceleration as being acquired by the accelerometer 116) may be correlated to the input power (such as through excitation voltage or the like) being used. In an alternate approach, the driving force may be lowered while keeping the excitation frequency relatively constant as a way to maintain the vibration amplitude constant, even as shifts in the response frequency occur. The computer system 160 performs one or both of these approaches repeatedly until an algorithmically- determined fatigue limit is reached using parameters such as acceleration, input power, response phase or the like. In one form, the stress for the number of total cycles corresponding to complete failure occurs is the fatigue limit strength.

During this second step 420, insight corresponding to when the material sample 1 is broken may be obtained. One way to gain such insight is to set a frequency shift limit. For example, in some tests, relatively large frequency shifts may take place in a short period of time. This sort of anomalous response can provide a user with indicia of failure of the material sample 1. Another way to gain such insight is through acceleration control, such as when crack within the material sample 1 start to interact with the portable accelerated fatigue testing system 100. In one form, this can change overall system dampening, thereby forcing acceleration to change (for example, increase) in order to keep up. Thus, while most acceleration changes are relatively small, they can change much more dramatically during failure; acquiring data related to such drastic changes provide a user with the needed indicia. Yet another way to gain such insight is through strain or displacement control. In one form, these values tend to drop off precipitously at the onset of failure. It will be understood that the underlying geometry and material makeup of the material sample 1 may significantly alter the response profile, as well as the corresponding limits being set on the fatigue. Parameter inputs may be adjusted accordingly. It will likewise be appreciated that these parameter inputs are different than those that occur during step 310, as during that earlier step certain properties (such as overall dampening levels for the particular material sample 1) are not known. Thus, while the exploratory activity under step 310 is ascertaining how acceleration corresponds to strain, as well as where certain frequencies respond, how the frequencies shift, the user at such exploratory period would not be able to set appropriate limits for the test. In other words, the limits can only be set once the phase tracking process of the first phase 300 is undertaken. Analysis of how long a particular test should be performed may be done automatically or with user input, the latter providing a human-in-the-loop (HIL) configuration. Moreover, fatigue testing on the material sample 1 may be run until either actual failure or a suitable indicia of failure has occurred. In the first case, the damage is open and notorious, while the computer system 160 allows shifts in response within the material sample 1 to be ascertained. In the second case, once satisfactory evidence of the failure is received, such as visually, algorithmically or by other means, the test may, at the option of the user or the computer system 160, be terminated or scaled back. In one form, the identification of when suitable indicia is in evidence may be based on a comparison to a limit or related predetermined criteria such as that available in the computer system 160.

In a third step 430, the fatigue data is received into the computer system 160. By reviewing trends within the data, a user may make an independent determination of when the failure occurred during such test. In another form, a trained ML model may be used to perform a diagnostic analysis of the data in order to make a comparable determination. In either event, this provides input to allow the determination of failure criteria. In this way, when a subsequent fatigue test is performed, that criteria may be used to gain further insights into similarly- configured material samples 1. In one form, such failure criteria may include a comparison of the number of cycles or steps to a detected shift (for example, 20%) of a response frequency or a detected shift (for example, 10%) of accelerometer 116 response, as well as other criteria that the user deems to be sufficiently probative of the material sample 1. As noted, these cycles or steps may correspond to time, frequency or other measure that corresponds to a particular snapshot that are accumulated over the duration of the fatigue test. At least a portion of a fatigue analysis may be performed manually, as well as internally within the portable accelerated fatigue testing system 100 or exported to another location (such as the cloud, on-site or off-site servers, third-party dedicated software (such as that available through MATLAB, Excel or a related numeric computing environment). As previously discussed, in lieu of an a priori-base analysis, an ML one may be performed, including those based on CNN or other algorithms from which a trained model and ensuing inference engine may operate on the received response data in order to provide a meaningful predictive analysis or related output. Decisions on whether to have the analysis be performed manually, within the portable accelerated fatigue testing system 100 or exported elsewhere may be based on numerous factors, including the scope of the chosen limits on the failure criteria or other form of relevant fatigue data, costs, need for real time versus later-in-time results or the like.

Referring with particularity to FIG. 4 in conjunction with previously-discussed FIGS. 5A through 5C, the third phase 500 shows how the fatigue test also correlates the acquired data from FIG. 3B to actual strain levels of the material sample 1. In one form, this correlation is used to provide visual indicia to a user for subsequent decision-making. In one form, this correlation is used to provide recordable data for real time or subsequent use or analysis. It will be appreciated that the portable accelerated fatigue testing system 100 may be operated to provide numerous different degrees of human intervention, all the way from substantially automated to substantially manual, as well as varying degrees of HIL, and that all such variants are within the scope of the present disclosure. In a first step 510, the optical data from the optical sensor 120 is received. As shown with particularity in FIG. 5A, a video or still image of the material sample 1 is recorded in memory 162 (not presently shown), as well as on a display such as display 166B. Additional indicia, such as descriptors of the thickness, free edge displacement F(x) or other attributes of the material sample 1 , may also be shown on the display 166B. When the free edge displacement F(x) is zero, the corresponding up-or-down position of the free edge of the material sample 1 along the horizontal axis (when the material sample 1 is configured as a flat, substantially two-dimensional structure) is correspondingly zero. As shown, the display provides a two-dimensional representation of the material sample 1 in its undeformed shape. In one form, a screen grid Gi may be provided in order to present scale or other quantified values. As can be seen, the free edge displacement F(x) is equal to zero when the material sample 1 in its undeformed shape.

In a second step 520, the image units are calibrated. The screen grid Gi is based on the free-edge thickness of the material sample 1. The thickness T (presently shown extending a vertical Cartesian axis direction) is known, such as through manual means, known databases or scanned from measurements such as those available through the optical sensor 120 or other means. In one form, the screen grid Gi may be used to provide quantitative indicia.

In a third step 530, the input commands from step 310 are established. As shown with particularity in FIG. 5B, a discretized grid G2 may be superimposed on the displayed free-edge of the material sample 1 to correspond to the number of measurement points being taken. In this way, a curve fir of free-edge displacement may be made. A refresh rate related to the frequency of the image is recorded and various types of data (such as refresh rate or the like) is measured. By way of example, the refresh rate may be set to 3 seconds or some other suitable duration.

In a fourth step 540, the real time deformed image results are received into the portable accelerated fatigue testing system 100. At this time, the fatigue test is being performed, such as through the imposition of the previously-described vibration patterns onto the material sample 1. A frame rate of the optical sensor 120 (or some other camera, if needed) is set to measure the free edge displacement F(x) repeatedly, such as every few seconds. As shown with particularity in FIG. 5C, the free edge displacement F(x) is the displacement that during the fatigue test is that which corresponds to each of the locations on discretized grid G2 of FIG. 5B. It will be appreciated that in certain modes of vibration such as first and second flexural bending, most (if not all) of the vibration will be up and down along the vertical axis, while in other modes of vibration such as edgewise bending, much of the vibration will be side-to-side along the horizontal axis, as well as having tip components that exhibit telltale vibration patterns up and down along the vertical axis. It will be further appreciated that certain patterns associated with the modes are known to those skilled in the art and as such will not be discussed further.

In a fifth step 550, a fit is made between the free edge displacement F(x) and a strain s. Referring with particularity to FIG. 5C, in one form, a function corresponding to the free edge displacement F(x) can be defined by curve fitting the displacements at the locations of the discretized grid G2 of FIG. 5B. When the free edge displacement F(x) (which is a function of where along the horizontal axis of the free edge of the material sample 1 a particular measurement is taken) is known for a sufficiently large number of locations, the equation defined by the curve fit may be used to calculate strain s based on displacement. F(x) by the following:

where c is half of the thickness along the vertical axis of the material sample 1 and is used as a measure of the compression-tension (C-T) couple that are symmetrical about an elastic curve and describe an internal bending moment within the material sample 1. Stated another way using terms from known flexure equations:

where M is the bending moment at a particular radius of curvature, E is the modulus of elasticity and I is the moment of inertia. It will be appreciated that there are other ways that this relationship could be expressed, and that all are within the scope of the present disclosure. In addition to having results available for user viewing on the display 166B, they can be saved into memory 162 or other tangible storage medium, whether locally or remotely for further calculations, historical recordkeeping or the like.

Although the portable accelerated fatigue testing system 100 is described for use in determining one or more of the modal properties and fatigue behavior of a material sample 1 being analyzed, it will be appreciated that it may be used for various forms of material sample testing and analysis, including those related to structural health monitoring (SHM), operational modal analysis (OMA), operating deflection shape (ODS), deformation monitoring, the aforementioned modal analysis or the like.

In another form, the computer system 160 (whether autonomously or in cooperation with cloud or other remote resources) may employ a trained ML model as discussed herein to perform predictive analytics that could lead to the desired or expected fatigue life behavior of the material sample 1. Thus, upon the use of fatigue data associated with the accelerated fatigue testing disclosed herein, subjecting such data to an ML model may significantly reduce the time and cost required to associate material properties to the component under consideration as well as to one or more underlying manufacturing process parameters. In one form, this may include determining numerical correlations between various failure mechanisms (or factors) and material properties, then creating a network of dependence between failure factors and material properties in order to generate a baseline for a subsequent ML or Al model. From there, the portable accelerated fatigue testing system 100 may demonstrate material test reductions based on the created network and material property correlation. In one form, the ML or Al model is configured to support small dataset input from standardized and accelerated test methods in order to perform a predictive assessment of process-structure properties. In another form, this predictive assessment of processstructure properties may be used as part of a manufacturing operations management process. Regardless of the form, constitutive models may be used to correlate the stress and strain of a particular alloy and geometric configuration of the material sample 1, from which damage parameters and subsequent failure prediction may be ascertained from minimal amounts of data. In one form, the stress, strain, and damage are part of a properties-related workflow (or process) that may be used to correlate data in a decision tree-based workflow that may be used in conjunction with or in lieu of other constitutive models. In one form, an energy-based approach may include performing a simplified stress-strain constitutive model associated with a damage parameter and then extending it to including the influence of a short crack ductile versus brittle crack phenomenon, as well as residual stress effects. In one form, a Bayesian approach and Monte Carlo simulation were applied. By the use of the portable accelerated fatigue testing system 100, optionally in conjunction with an ML model, the amount of data required to gain detailed constitutive model insights may be significantly reduced.

Since the fatigue test is vibration-based where the response is based on factors such as the specimen geometry and material properties, it is possible to make correlations to material properties beyond fatigue life data. Specifically, modal responses are directly correlated to the elastic modulus and density of the material that is used in the material sample 1. Therefore, knowing the frequency and the mode shape (for example, through the vibration pattern) can lead to extracting either of the two properties. With ML, the portable accelerated fatigue testing system 100 can be taught to gather these properties based on previous works, including those that can provide known baseline values. These modal response properties can be correlated to thermophysical properties, where the incorporation of a controlled heating source along with density and temperature distribution from specimen overheating can also lead to capturing thermal conductivity, thermal expansion and specific heat.

In a likewise manner, understanding the impacts of microstructure, porosity, surface roughness, residual stress or the like may be used to predict fatigue and strength performance of AM-produced components. In one form, the portable accelerated fatigue testing system 100 can be used to determine numerical correlations between the failure mechanisms (factors) and material properties. Moreover, additional correlations can be made are modulus to hardness, modulus and hardness to elongation, tensile strength associated with modulus, hardness and fatigue life, as well as others. In one form, a neural network or other ML approach can be used to extract multiple material properties from a single accelerated fatigue test. This ML approach in turn allows a dependence to be determined between failure factors and material properties.

Referring next to FIG. 6, the results of a fatigue test conducted with the portable accelerated fatigue testing system 100 are shown on a screen of the display 166B. Failure is identified when the shaker or other excitation source 110 of FIGS. 1, 2A and 2B experiences an increase in acceleration (shown with a solid line) or the driving frequency (shown with a dashed line) shifts rapidly downward. This is evidenced by the fact that because the phase and the amplitude are locked and being tracked, once a crack (failure) initiates in the material sample 1, the excitation source 110 will need to put more force into the material sample 1 in order to keep the response amplitude constant.

Referring next to FIGS. 7A through 7D, different simulated clamping (that is to say, boundary) conditions of the material sample 1 are shown using a particular clamping (also referred to herein as a monolithic) fixture 115. As shown, by bolting through numerous (six are shown) holes 111, the various parts of the fixture 115 (such as the clamping block 112) are in direct vibration communication with the excitation source 110 (not presently shown). Each of these clamping conditions is expected to produce different responses, modes and patterns. In the first clamping condition (FIG. 7A), a bolt pre-tension is simulated. In the second clamping condition (FIG. 7B), a fixed bolt hole and washer imprint is simulated, showing with particularity the location for the fixed support. The clamping condition of FIG. 7B achieves a lower amount of system damping compared to the condition of FIG. 7A. This in turn means that a smaller amount of energy from the excitation source 110 is required to break the material sample 1. Significantly, the cylindrical shape of the clamping fixture 115 allows a test to have reduced weight, a strong clamping condition and low overall system damping; together, these contribute to low excitation input and high response within the material sample 1. In the third clamping condition (FIG. 7C), a static structural fixed support is simulated for achieving low system damping. As shown, by bolting through numerous (six are shown) holes 111, the various parts of the fixture (such as the clamping block 112) are in direct vibration communication with the excitation source 110. Referring next to FIG. 7D, a fourth clamping condition is shown where the monolithic fixture 115 is similar to the one depicted in FIGS. 7A and 7C, but now shown secured to a portion of the excitation source 110. Certain components that make up the excitation source 110, such as electrical coils and permanent magnets, have been removed for clarity. Upon inclusion of these and other components in an assembled condition, an appropriate signal (such as from one or both of the amplifier 130 and digital control hardware 140) may be sent in order to cause the excitation source 110 to start moving, making it a moving element. The authors of the present disclosure have determined that further reductions in system damping may be achieved by combing the moving excitation source 110 and the clamping block 112 into one monolithic part. Such further reductions may lead to the use of even lower energy input while simultaneously realizing even higher response in the material sample 1.

Referring next to FIGS. 8 A through 8C, the results of both FEA displacement predictions and FEA strain predictions are shown in numerical form for various measurement locations away from the fixed support along the chordwise dimension of an actual material sample 1 that is represented in FIGS. 7A through 7C. Three different plate materials were used, including aluminum alloy AlSilOMg in FIG. 8A, aluminum alloy 6061-T6 in FIG. 8B and stainless steel alloy 316L in FIG. 8C. Each of the plate materials were configured to represent the material sample 1 of FIGS. 1, 2A, 2B and 5A through 5C under two different clamping conditions, one for the bolt pre-tension of FIG. 7A and one for the fixed bolt hole and washer imprint of FIG. 7B. As can be seen, similar damping ratios were employed, as were generally similar excitation frequencies. It will be appreciated that no simulation was conducted for the static structural fixed support of FIG. 7C as the totality of the fixture is not in resonance, while all the static and dynamic loads are in the elastic region of the material sample 1. As such, the stresses on the fixture during operation are note close to any critical failure values of the material sample 1.

Referring next to FIGS. 9 A and 9B, the results of both finite element analysis displacement predictions and strain predictions are shown in graphical form. As can be seen, the first and second clamping conditions of FIGS. 7A and 7B were used. As to be expected, the displacement is at a minimum at the fixed support location. This information may be used to gain a more holistic understanding of how boundary condition effects impact response, and from this to enable an accurate correlation between testing and analysis. As such, results such as those depicted in these figures provides a measure of confidence in the calculation of correlating strain to displacement responses.

Within the present disclosure, it will be understood that the operations, functions, logical blocks, modules, circuits, and algorithm or model steps or events described may be implemented in hardware, software, firmware or any combination thereof. Moreover, if implemented in software, such operations may be stored on or transmitted over as one or more instructions or code on the aforementioned computer-readable medium. The steps or events of a method, algorithm or ensuing model disclosed herein may be embodied in a processor-executable software module, which may reside on a tangible, non-transitory version of such computer- readable medium such that the medium be in any available form that permits access to the events or steps by a processor or related part of a computer. By way of example, and not limitation, such non-transitory computer-readable medium may comprise those previously mentioned, including RAM 162A, ROM 162B, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory or any other form that may be used to store desired program code in the form of instructions or data structures and that may be accessed by a processor or related part of a computer. Combinations of the above should also be included within the scope of non-transitory computer-readable media. Additionally, the operations of a method, algorithm or model may reside as one or any combination or set of codes or instructions on a tangible, non-transitory machine readable medium or computer-readable medium, which may be incorporated into a computer program product. Furthermore, in one non-limiting form, upon having the program code means loaded into memory in general (and in one form into ROM in particular), the processor, microprocessor, controller, microcontroller, SoC or related computational device becomes a specific-purpose machine configured to perform the various computational tasks described herein. In this way, the associated computer becomes a particularly- adapted computer or computer-related data processing device that possesses particular capabilities tied to the resulting instruction set architecture.

Within the present disclosure, one or more of the following claims may utilize the term "wherein" as a transitional phrase. For the purposes of defining features discussed in the present disclosure, this term is introduced in the claims as an open-ended transitional phrase that is used to introduce a recitation of a series of characteristics of the structure and should be interpreted in like manner as the more commonly used open-ended preamble term "comprising" and its variants that do not preclude the possibility of additional acts or structures.

Within the present disclosure, terms such as “preferably”, “generally” and “typically” are not utilized to limit the scope of the claims or to imply that certain features are critical, essential, or even important to the disclosed structures or functions. Rather, these terms are merely intended to highlight alternative or additional features that may or may not be utilized in a particular embodiment of the disclosed subject matter. Likewise, it is noted that the terms “substantially” and “approximately” and their variants are utilized to represent the inherent degree of uncertainty that may be attributed to any quantitative comparison, value, measurement or other representation. As such, use of these terms represents the degree by which a quantitative representation may vary from a stated reference without resulting in a change in the basic function of the subject matter at issue.

Within the present disclosure, the use of the prepositional phrase "at least one of' is deemed to be an open-ended expression that has both conjunctive and disjunctive attributes. For example, a claim that states "at least one of A, B and C" (where A, B and C are definite or indefinite articles that are the referents of the prepositional phrase) means A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B and C together. By way of example, if a claim recites that data is being acquired from at least one of a first sensor, a second sensor and a third sensor, and if such data is being acquired from the first sensor alone, the second sensor alone, the third sensor alone or any combination of the first, second and third sensors, then such data acquisition satisfies the claim. Within the present disclosure, the following claims are not intended to be interpreted based on 35 USC 112(f) unless and until such claim limitations expressly use the phrase "means for" or “steps for” followed by a statement of function void of further structure. Moreover, the corresponding structures, materials, acts and equivalents of all means or step plus function elements in the claims that follow are intended to include any structure, material or act for performing the function in combination with other claimed elements as specifically claimed.

Within the present disclosure, the singular forms “a,” “an” and “the” include plural references unless the context clearly dictates otherwise. The modifier “about” used in connection with a quantity is inclusive of the stated value and has the meaning dictated by the context (for example, it includes at least the degree of error associated with the measurement of the particular quantity). The modifier “about” should also be considered as disclosing the range defined by the absolute values of the two endpoints. For example, the expression “from about 2 to about 4” also discloses the range “from 2 to 4.” The term “about” may refer to plus or minus 10% of the indicated number. For example, “about 10%” may indicate a range of 9% to 11%, and “about 1” may mean from 0.9 to 1.1. Other meanings of “about” may be apparent from the context, such as rounding off, so, for example “about 1” may also mean from 0.5 to 1.4.

For the recitation of numeric ranges herein, each intervening number there between with the same degree of precision is explicitly contemplated. For example, for the range of 6 to 9, the numbers 7 and 8 are contemplated in addition to 6 and 9, and for the range 6.0 to 7.0, the number 6.0, 6.1, 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9 and 7.0 are explicitly contemplated.

Having described the subject matter of the present disclosure in detail and by reference to specific embodiments, it is noted that the various details disclosed in the present disclosure should not be taken to imply that these details relate to elements that are essential components of the various described embodiments, even in cases where a particular element is illustrated in each of the drawings that accompany the present description. Further, it will be apparent that modifications and variations are possible without departing from the scope of the present disclosure, including, but not limited to, embodiments defined in the appended claims. More specifically, although some aspects of the present disclosure may be identified as preferred or particularly advantageous, it is contemplated that the present disclosure is not necessarily limited to these aspects.

It will be apparent to those skilled in the art that various modifications and variations can be made to the described embodiments without departing from the spirit and scope of the claimed subject matter. Thus it is intended that the specification cover the modifications and variations of the various described embodiments provided such modification and variations come within the scope of the appended claims and their equivalents.

Claims

CLAIMS What is claimed is:

1. A portable accelerated fatigue testing system, comprising: a test signal source; an amplifier in signal communication with the test signal source; a support configured to secure a material sample thereto; an excitation source coupled to the support, the excitation source in signal communication with the amplifier; a sensor; and a computer system comprising a processor, digital conversion hardware in signal communication with the sensor, a display and non-transitory computer readable storage comprising computer executable instructions, wherein the processor carries out the computer executable instructions to: output, to the display, a first graphical user interface that is configured to receive a user-generated input command that characterizes at least one parameter of a control signal; control the test signal source to output the control signal that is based upon the usergenerated input command that corresponds to a particular accelerated fatigue test regimen, whereupon: the amplifier generates an excitation drive signal responsive to the control signal from the test signal source; the excitation source imparts a vibration profile into the support, the vibration profile comprising at least the excitation drive signal from the amplifier; and the support imparts vibratory energy corresponding to the vibration profile into the material sample; obtain digital sensor data from the digital conversion hardware corresponding to a vibration pattern that is generated in the material sample in response to receipt of the vibratory energy; measure a deflection of at least a portion of the material sample using the obtained digital sensor data; correlate the deflection to a predetermined fatigue limit of the material sample; and output, in a second graphical user interface on the display, an expected fatigue characteristic of the material sample.

2. The portable accelerated fatigue testing system of claim 1, further comprising: at least one additional sensor in signal communication with the material sample; an input-output coordinator in signal communication with the amplifier and at least one of the sensor and the at least one additional sensor; and network circuitry that provides signal communication between the excitation source, the and at least one of the sensor, the at least one additional sensor and the computer system.

3. The portable accelerated fatigue testing system of claim 1, wherein the control signal comprises a time- varying control signal.

4. The portable accelerated fatigue testing system of claim 1, wherein the material sample is secured to the excitation source without a load frame or a movable crosshead.

5. The portable accelerated fatigue testing system of claim 1 , further comprising a light source configured to illuminate the material sample.

6. The portable accelerated fatigue testing system of claim 1, further comprising a platform upon which at least some of the support, excitation source, sensor, and computer system is commonly disposed.

7. The portable accelerated fatigue testing system of claim 1, wherein the excitation source comprises a multi-axis excitation source.

8. The portable accelerated fatigue testing system of claim 1, wherein the excitation source is selected from the group consisting of a transducer, a speaker, a piezoelectric exciter and an electrodynamic exciter.

9. The portable accelerated fatigue testing system of claim 1 , wherein the portable accelerated fatigue testing system includes a controller with a phase lock loop to adjust the frequency of the input command to maintain the frequency in substantial synchronization with a resonant response frequency within the material sample that corresponds to changes in the vibration pattern.

10. The portable accelerated fatigue testing system of claim 1, wherein the computer executable instructions output the expected fatigue characteristic of the material sample in the second graphical user interface based on a machine learning inference.

11. The portable accelerated fatigue testing system of claim 1 , further comprising a heating source placed in thermal communication with the material sample.

12. A process for performing a fatigue test on a material sample, the process comprising: configuring a portable accelerated fatigue testing system to generate a vibratory profile; inducing, by the portable accelerated fatigue testing system during a first phase of the fatigue test, a vibratory response in the material sample; converting the vibratory response into a first output that corresponds to a peak response of a free edge of the material sample; generating, by the portable accelerated fatigue testing system during a second phase of the fatigue test: a test command for an actual test of the material sample; fatigue test limits of the material sample; and actual fatigue data of the material sample; and acquiring, by the portable accelerated fatigue testing system during a third phase of the fatigue test: an image of the free edge of the material sample while the material sample is in an unexcited state; a calibrated image corresponding to the free edge of the material sample; an input command corresponding to sampling parameters to be used; free edge displacement of the material sample; and an output corresponding to deformations at the free end that are based on a difference between the unexcited state and an excited state that is produced during the actual test of the material sample.

13. The process of claim 12, wherein the vibratory profile comprises a control signal.

14. The process of claim 12, wherein the output corresponding to deformations at the free end that are based on a difference between the unexcited state and an excited state that is produced during the actual test of the material sample comprise a curve fit of free edge displacement that take place at a plurality of locations on the material sample.

15. The process of claim 14, further comprising determining, by the portable accelerated fatigue testing system, a strain amplitude of the material sample based on the curve fit.

16. The process of claim 12, further comprising: detecting, by the portable accelerated fatigue testing system, if a resonant response frequency changes within the material sample; adjusting an input frequency of the input command if the resonant response frequency has changed within the material sample, otherwise maintaining the input frequency of the input command; and using a phase lock loop to ensure that the input frequency of the input command and the resonant response frequency of the material sample remain substantially synchronized throughout at least a portion of the fatigue test.

17. The process of claim 12, wherein at least one of the first and second phases are controlled by varying at least one of a plurality of parameters comprising acceleration, input power and excitation frequency.

18. The process of claim 17, wherein the acceleration is maintained at a substantially constant level while at least one of the other parameters is varied.

19. The process of claim 17, wherein the excitation frequency is maintained at a substantially constant level while the input power is lowered.

20. The process of claim 12, wherein the material sample defines an aperture therein that is predisposed to cause the material sample to respond with a particular vibration pattern in response to the vibratory profile.