CN1475779A - Multi-sensor Noise and Vibration Measurement Analyzer Based on Virtual Instrument Technology - Google Patents

Abstract

The invention relates to a multi-sensor noise and vibration measurement analyzer based on virtual instrument technology, which can be used to measure, analyze and evaluate the noise and vibration of complex mechanical systems. A multi-sensor noise and vibration measurement analyzer based on virtual instrument technology is mainly composed of a sensor, an analog input front-end processor, and a PC. The serial I/O port of the processor is connected with the serial port of the PC, and the PC analyzes and processes the digital signal through the measurement software module; the sensors include noise sensors, vibration sensors and rotational speed sensors, noise sensors, vibration sensors At least one respectively, and the number of analog input front-end processors is determined according to the number of sensors. The invention can simultaneously measure noise and vibration signals, has high measurement accuracy and resolution, and can accurately analyze and evaluate systems and mechanical components.

Description

Multi-sensor Noise and Vibration Measurement Analyzer Based on Virtual Instrument Technology

technical field

The invention relates to a measurement analyzer, in particular to a multi-sensor noise and vibration measurement analyzer based on virtual instrument technology which can be used to measure, analyze and evaluate the noise and vibration of complex mechanical systems.

Background technique

Mechanical components will generate noise and vibration during their movement, both of which are inseparable, and both are embodied as mechanical waves.

In the electromechanical industry, in order to improve product quality, find system faults, or comprehensively evaluate the overall performance of products, etc., it is necessary to measure the noise and vibration of mechanical systems or components. The structure and working principle of the noise and vibration measuring instrument are basically the same, and it is composed of a sensor, a signal conditioning circuit, an A/D converter, a spectrum analysis software and a display.

At present, the domestic noise and vibration measuring instruments have the following problems:

1. Separation of noise measurement and vibration measurement One instrument cannot measure noise and vibration signals at the same time. In a mechanical system, noise and vibration often exist at the same time. Although there is a certain relationship between the two, only by measuring The noise signal obtained indirectly from the vibration signal is inaccurate and not detailed, which will affect the reliability of the system analysis.

2. The structure of the measuring instrument is a traditional form. The instrument is composed of a circuit board, a chassis and an operation panel. Due to limited hardware and software resources and high hardware production costs, the performance and functions of the instrument are greatly limited. For example, in order to improve the noise The resolution of the frequency spectrum adopts the method of segmentation filtering, and the filter bandwidth is 1/3 octave. If a hardware filter is used, the number of filters is limited due to the influence of hardware volume and cost. If a software filter is used , is also limited by hardware and software resources, so it is difficult to improve the measurement accuracy and resolution of traditional instruments.

3. The measuring instrument adopts a single sensor. It is a relatively complex mechanical system. The mechanical parts are connected together, and there are multiple moving parts. The noise and vibration information generated by them are superimposed or mixed together. Using a single-sensor measuring instrument cannot Identify and distinguish specific characteristics of noise and vibration signals, making it difficult to accurately analyze and evaluate systems and mechanical components.

Virtual instrument is a new generation of instrument after digital instrument and intelligent instrument. It uses computer as the software and hardware platform of the instrument, and the sensor and computer are connected through an interface card. Its biggest feature is to make full use of computer resources, break through the concept of traditional instruments, save the panel, chassis and a lot of hardware of the instrument, use software to construct a soft operation panel on the computer screen, and replace the hardware for data processing, analysis and diagnosis. Its calculation function and data processing function are unmatched by traditional instruments.

The Ministry of Science and Technology of the People's Republic of China has made it clear in the "Guidelines for Several Key Projects of the Technology Innovation Fund for Small and Medium-sized Science and Technology Enterprises in 2003" that it will focus on funding the development and industrialization of this type of measuring instrument (see project guide: 4. Optical Mechatronics (2) High Performance , Intelligent instrumentation 5. Virtual instrument and general test platform (1) Sound and vibration analysis virtual instrument).

Multi-sensor information fusion is an emerging subject, and many countries (including China) have listed it as a key technology for the next stage of development. Multi-sensor information fusion is actually the use of computer technology to simulate the ability of the human brain to deal with complex problems comprehensively. In a multi-sensor system, multiple sensors are installed in different parts of the system under test, and the information obtained by each sensor has different characteristics. These information may be mutually supportive or complementary, or may be mutually contradictory or conflicting. Sensor information fusion technology optimizes the combination of information obtained by each sensor to obtain true, effective and reliable information.

The following patents were found from the patent database: "Motor Vibration and Noise Detector" (00216296.2) and "Digital Vibration and Noise Detector" (01238477.1), which use the acceleration vibration sensor to measure the acceleration of the vibration signal, and obtain the speed signal after one integration. The position signal is obtained after the second integration, and the position signal is passed through the A weighting device to obtain the equivalent value of the vibration signal. Because there is a certain relationship between the vibration signal and the noise signal, the size of the noise signal can be obtained indirectly by this method. The method is called the vibration measurement method of noise, and it is impossible to obtain detailed and accurate noise information. These two patents are essentially vibration measuring instruments. "Vibration and Noise Test Device for Bench" (01205906.4) is mainly a test device for engine performance, which combines the output signal of an independent sound level meter (a DC level, whose size reflects the size of the noise) and The output signal of the acceleration sensor is sent to the filter at the same time, and the output signal of the filter is sent to the computer for recording through the A/D converter. This patent essentially applies two separate sets of sound level meter and acceleration measurement system to the engine performance test bench at the same time, and it does not appear from the patent specification and drawings that the measurement device can measure and record noise and vibration at the same time. Therefore, this patent does not have an independent measuring instrument capable of simultaneously measuring noise and vibration, and the output of the sound level meter is the sound pressure level, which is a comprehensive reflection of all frequency components of the noise, and cannot be used to analyze the relationship between sound pressure and frequency. relationship between.

To fundamentally solve the problems of noise and vibration measuring instruments, we must start from the structural design of the instruments. Virtual instrument technology and multi-sensor information fusion technology provide the basis for solving the above problems. The noise and vibration measurement analyzer using virtual instrument technology and multi-sensor information fusion technology has not been reported in China.

Contents of the invention

In order to overcome the problems existing in the current domestic noise and vibration measuring instruments, the purpose of the present invention is to provide a device that can simultaneously measure noise and vibration signals, has high measurement accuracy and resolution, and can accurately analyze the system and mechanical parts. and evaluation of a multi-sensor noise and vibration measurement analyzer based on virtual instrument technology.

In order to achieve the above object, the technical solution of the present invention is: a multi-sensor noise and vibration measurement analyzer based on virtual instrument technology, which is mainly composed of sensors, analog input front-end processors, and PCs, and is characterized in that the output terminal of the sensor and the analog input The input end of the front-end processor is connected, and the serial I/O port of the analog input front-end processor is connected with the serial port of the PC. The sensor converts the measured non-electrical signal into an electrical signal and sends it to the analog input front-end processor. The input front-end processor converts the signal into a digital signal and inputs it into a PC through the serial port, and the working state of the analog input front-end processor is controlled by the PC, and the PC analyzes and processes the digital signal through a measurement software module; the sensor Including noise sensor, vibration sensor and rotational speed sensor, noise sensor and vibration sensor are at least one respectively, and the number of analog input front-end processors is determined according to the number of sensors. Among them, the noise sensor and the vibration sensor can be selected by the user according to the needs, and the speed sensor is used to measure the noise and vibration of the rotating workpiece under different speed conditions.

Described analog input front-end processor is the AD73360 chip of U.S. AD company, and this chip has 6 16 A/D converter channels (6 channels can sample simultaneously), and each channel is made up of signal conditioner, programmable gain amplifier , analog Σ-Δ modulator, sampler and serial I/O port, etc. It can also use other types of analog input front-end processors or use separate chips to form analog input front-end processors.

Described PC machine is equipped with measurement software module, adopts the product LabVIEW virtual instrument software of U.S. NI company to construct virtual instrument software platform, and measurement software module comprises software operation interface, measurement control software module, multi-sensor information fusion software module and spectrum analysis software In the four parts of the module, the user selects the number of noise sensors and vibration sensors through the software operation interface, and determines whether to enable the speed measurement, and then starts the measurement control software module; the measurement control software controls the AD73360 to collect the signals of each sensor in a roving detection mode, and The signal is stored in the internal database; the multi-sensor information fusion software module fuses the signals of each sensor and stores the fused signal in the internal database; the spectrum analysis software module performs spectrum analysis on the noise and vibration signals before and after fusion , and display the spectrogram on the computer screen.

Multi-sensor information fusion is the process of weighting the information of noise sensors and vibration sensors. Let X1(ω), X2(ω),…, Xn(ω) denote the spectrum expressions of the signals measured by n sensors respectively, then The composite expression of n signals is: $x\left(\mathrm{\&omega;}\right)=\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}\mathrm{Ali\; Xl}\left(\mathrm{\&omega;}\right)+\underset{i=1}{\overset{\mathrm{<figure-callout\; id="2"\; label="m\; A"\; filenames="A0312832000061.png,A0312832000081.png"\; state="\{\{state\}\}">m</figure-callout>}}{\mathrm{\&Sigma;}}}A2\mathrm{iX}2\left(\mathrm{\&omega;}\right)+\&CenterDot;\&CenterDot;\&Center\; Dot;+\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}\mathrm{AniXn}\left(\mathrm{\&omega;}\right)$ In the formula, A1i, A2i, ..., Ani represent the weighting coefficients of different frequency components of X1(ω), X2(ω), ..., Xn(ω) respectively, and the value range is [0, 1]. The calculation of the weighting coefficients can be Arithmetic weighting, statistical weighting or other weighting algorithms are used, such as fuzzy algorithm or neural network algorithm.

The present invention has adopted a plurality of sensors (noise sensor, vibration sensor, rotational speed sensor, wherein noise sensor, vibration sensor are respectively at least 1), and is connected with PC by analog input front-end processor, has realized simultaneous measurement of noise and vibration. The purpose of a variety of signals, high measurement accuracy and resolution, and accurate analysis and evaluation of systems and mechanical components. Beneficial effects are: 1. Since the software and hardware resources of the computer are very powerful, so in the spectrum analysis software module, the frequency band of the filter can be made very narrow, greatly improving the resolution and accuracy of the measurement; 2. Not only can determine It can perform various spectrum analysis on random noise and vibration signals, and can perform various statistical analysis and correlation analysis on random noise and vibration signals; 3. Due to the use of multi-sensor information fusion technology, the quality analysis of complex mechanical systems, Fault diagnosis and performance evaluation are more accurate and reliable.

Description of drawings

Fig. 1 is a structural representation of the present invention

Fig. 2 is the software embodiment figure of the present invention

Fig. 3 is the specific embodiment figure of the present invention

Detailed ways

The hardware implementation scheme is described with reference to Fig. 1: noise sensor and vibration sensor 1-1, 1-2, ..., 1-n and speed sensor 2 are connected to analog input front-end processor 3, and analog input front-end processor 3 is connected to computer 4. The software implementation scheme is described with reference to Figure 2: the user selects the number of noise sensors and vibration sensors through the software operation interface, and determines whether to enable the speed measurement, and then starts the measurement control software module; the measurement control software controls the front-end processing of the analog input in a roving detection mode The sensor AD73360 collects the signals of each sensor and stores the signals in the internal database; the multi-sensor information fusion software module fuses the signals of each sensor and stores the fused signals in the internal database; the spectrum analysis software module fuses The noise and vibration signals before and after are subjected to spectrum analysis, and the spectrogram is displayed on the computer screen.

Figure 3 is an application example of measuring and analyzing the rotating shaft of a large motor. The purpose is to analyze whether there is swing vibration, bending vibration, radial vibration and axial vibration caused by unbalanced force or unbalanced force couple or other reasons on the rotating shaft of the motor. At the same time, the relationship between the noise and vibration of the bearing at different speeds and the relationship between the speed of the motor and the overall noise are analyzed. Among the figure, vibration sensors 1-1, 1-2, 1-3 and 1-4 measure the radial and axial vibrations of the motor shaft 5 through bearing blocks 6-1 and 6-2 of the motor shaft 5, and noise sensors 1-5 and 1-6 measure the noise of the bearing, the noise sensors 1-7 measure the noise radiated from the motor casing 7, and the rotational speed sensor 2 measures the rotational speed of the motor shaft 5. The sensors 1-1, 1-2, 1-3, 1-4, 1-5, 1-6, 1-7 and 2 are connected to the analog input front-end processor 3, and the analog input front-end processor 3 is connected to the computer 4. In the implementation of this example, the phase relationship of the vibration sensors 1-1, 1-2, 1-3 and 1-4 is very important to analyze the vibration mode of the shaft, which cannot be achieved with a single sensor measuring instrument Yes, even if multiple vibration measuring instruments are used, it is impossible to measure the phase relationship between the four vibration sensors. The present invention can simultaneously provide the spectrum diagram of the measurement signal of a single noise sensor and vibration sensor, the noise signal spectrum diagram and the vibration signal spectrum diagram after fusion, and the noise and vibration signal spectrum diagram after fusion, so that the noise and vibration signal of the motor shaft can be analyzed. Vibration analysis and comprehensive evaluation. The embodiment of the present invention is incomparable to traditional measuring instruments and their measuring schemes in terms of resolution, accuracy, reliability and detail.

Claims (4)

1. A multi-sensor noise and vibration measurement analyzer based on virtual instrument technology, which is mainly composed of sensors, analog input front-end processors, and PCs. It is characterized in that the output of the sensor is connected to the input of the analog input front-end processor. The serial I/O port of the input front-end processor is connected to the serial port of the PC. The sensor converts the measured non-electrical signal into an electrical signal and sends it to the analog input front-end processor. The analog input front-end processor converts the signal into The digital signal is input into the PC through the serial port, and the working state of the analog input front-end processor is controlled by the PC, and the PC analyzes and processes the digital signal through the measurement software module; the sensor includes a noise sensor, a vibration sensor and a rotational speed sensor, There is at least one noise sensor and one vibration sensor respectively, and the number of analog input front-end processors is determined according to the number of sensors. 2. The multi-sensor noise and vibration measurement analyzer based on virtual instrument technology according to claim 1, characterized in that: the analog input front-end processor is the AD73360 chip of American AD Company, and the AD73360 chip has 6 16-bit A/D converter channel, each channel is composed of signal conditioner, programmable gain amplifier, analog Σ-Δ modulator, sampler and serial I/O port. 3. The multi-sensor noise and vibration measurement analyzer based on virtual instrument technology according to claim 1, characterized in that: said PC is equipped with a measurement software module, and the measurement software module includes a software operation interface and a measurement control software module 1. Multi-sensor information fusion software module and spectrum analysis software module. The user selects the number of noise sensors and vibration sensors through the software operation interface, and determines whether to enable speed measurement, and then starts the measurement control software module; the measurement control software uses roving detection control the AD73360 to collect the signals of each sensor, and store the signals in the internal database; the multi-sensor information fusion software module fuses the signals of each sensor, and stores the fused signals in the internal database; the spectrum analysis software module will The noise and vibration signals before and after fusion are subjected to frequency spectrum analysis, and the frequency spectrum is displayed on the computer screen. 4. The multi-sensor noise and vibration measurement analyzer based on virtual instrument technology according to claim 1, characterized in that: the PC analyzes and processes the digital signal through the measurement software module, including multi-sensor information fusion, multi-sensor information fusion It is the process of weighting the information of the noise sensor and the vibration sensor. Let X1(ω), X2(ω), ..., Xn(ω) denote the spectrum expressions of the signals measured by n sensors respectively, then the n signals The composite expression is: $x\left(\mathrm{\&omega;}\right)=\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}\mathrm{Ali\; Xl}\left(\mathrm{\&omega;}\right)+\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}A2\mathrm{iX}2\left(\mathrm{\&omega;}\right)+\&CenterDot;\&CenterDot;\&CenterDot;+\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}\mathrm{AniXn}\left(\mathrm{\&omega;}\right)$ In the formula, A1i, A2i,..., Ani respectively represent the weighting coefficients of different frequency components of X1(ω), X2(ω),..., Xn(ω).

Images