TWI851986B - Intelligent monitoring system and method for machine equipment - Google Patents

Abstract

The present invention discloses a machine equipment intelligent monitoring system and method, which uses a machine equipment working information sensing module to sense the working information of the machine equipment. The information acquisition module is used to capture and collect working information during the working period in a predetermined acquisition cycle to obtain the working information within the validity period. The information conversion module includes a feature extraction module, a feature selection module and a machine status real-time monitoring chart construction module. The feature extraction module is used to obtain working information features. The feature selection module is used to obtain effective working information. The machine status real-time monitoring chart construction module is used to construct a machine status variation reference chart using a learning algorithm. The machine status variation reference chart is a relationship chart composed of the corresponding working time points and variation values as the vertical axis and horizontal axis respectively. The real-time monitoring display interface module is used to display the relationship chart on the display screen, so that by integrating real-time sensing and artificial intelligence computing technologies, users can conveniently monitor the machine status variation information of the machine equipment in real time.

Description

Intelligent monitoring system and method for machine equipment

This creation is related to a machine equipment intelligent monitoring system and method, especially a machine equipment monitoring technology that can integrate real-time sensing and artificial intelligence computing technologies to allow users to conveniently monitor the mechanical variation status of machine equipment in real time.

According to the current situation, most of the more sophisticated machine tools are controlled by computer numerical value and written with program software to display the working data information of the dedicated machine tool on the local human-machine interface. However, due to the limitation of the computing power and storage capacity of the numerical controller, the relevant working information of the machine tool can only be processed and displayed locally. Therefore, the remote user cannot know the relevant working information of the machine tool in real time. As a result, the remote user is inconvenienced and troubled in monitoring the working data information.

Furthermore, as workpieces are moving towards miniaturization and high precision requirements, the processing accuracy and speed requirements of various machines and equipment (such as machine tools) are becoming increasingly stringent. Although the rapid development of the precision machinery and machine tool industries has led to an increase in the shipment volume of machines and equipment year by year, many manufacturers hope to seize the opportunity and have invested a considerable amount of R&D costs; however, in terms of improving processing quality and stability, they have not been able to further break through the technical bottleneck. The technical bottleneck is that machine tools do not have a set of technologies that can integrate real-time sensing and artificial intelligence computing, so that users cannot monitor the mechanical state of the machine equipment in real time, resulting in monitoring errors that are not discovered when failures are imminent, resulting in a decrease in production quality and yield. In addition, when the machine equipment stops working, the maintenance period causes production scheduling impacts and reduced production capacity, resulting in significant losses.

In addition, the improvement of workpiece dimensional accuracy, surface roughness and material removal rate has always been the focus of metal processing. However, when the cutting depth is too large, strong vibration will be generated between the tool and the workpiece, and chatter may be stimulated, which will leave wavy patterns on the workpiece surface, destroying the surface quality and dimensional accuracy, and accelerating tool wear. In severe cases, it may even damage the spindle of the machine tool. The general metal cutting process is highly nonlinear and complex, so it is indeed difficult to predict using traditional machine learning algorithms, resulting in inconvenience and trouble in machine monitoring.

To improve the above shortcomings, the relevant technical field has developed a patent as shown in the new announcement No. M621978 "Readable barcode real-time display system for machine tools". The patent uses a barcode display module installed in the machine tool to analyze the machine tool's operating information through an external device to obtain machine status messages or warning management messages, and then transmit them to the barcode display module to present them in the form of barcodes. Although the patent allows the client to use a barcode reader to read and obtain machine status information; however, the patent also does not have a set of integrated real-time sensing and artificial intelligence computing technologies to allow users to conveniently monitor the machine equipment's mechanical variation status in real time. Therefore, how to develop a machine tool monitoring technology that can integrate real-time sensing and artificial intelligence computing technology has become a key indicator technology that manufacturers are eager to develop and break through.

In addition, there is another patent as shown in the invention announcement No. I614083 "Method for predicting abnormal signs of the rotating body of the integrated processing machine", which uses a temperature and vibration sensing unit, a capture unit, a computer and a controller system to achieve a diagnostic method for predicting abnormal signs of the rotating body; the capture unit captures the operating state, temperature and vibration signals of the rotating body of the integrated processing machine as a basis for diagnosis, and then analyzes the rotating body state index through computer calculation. Although the patent can make the state of the rotating body easier to grasp through the predictive diagnosis method, thereby improving the utilization rate of the integrated processing machine; however, the patent does not have the functions of electromagnetic flow detection, feature extraction, feature selection, and visual information display, so it takes more computing time, making it difficult for the predictive diagnosis result data to meet the higher accuracy requirements, and it is impossible for users to monitor the variation of machine wear through the visual machine wear monitoring chart, so it is difficult to effectively reduce the probability of machine failure, resulting in inconvenience and trouble in monitoring the machine work data.

In view of the fact that the machine tools and the aforementioned patents are not perfect in functionality and still need to be improved, the inventor of this invention has actively invested in research and development with his accumulated years of rich processing machine research and design capabilities and experience. After continuous design, trial production and testing, the research and development results of this invention have finally been produced.

The first purpose of the present invention is to provide a machine equipment intelligent monitoring system and method, which is mainly built by integrating real-time sensing and artificial intelligence computing technologies, so that users can more conveniently monitor the machine wear data and machine wear monitoring relationship diagrams and other variation status information in real time, so as to obtain the warning information of the machine that is about to change in advance, so as to effectively reduce the probability of machine failure. The technical means to achieve the above-mentioned first purpose is to use a machine equipment working information sensing module to sense the working information of the machine equipment. The information capture module captures and collects the working information during the working period in a predetermined capture cycle to obtain the working information within the validity period. The information conversion module includes a feature extraction module, a feature selection module and a machine state real-time monitoring chart construction module. The feature extraction module is used to obtain working information features. The feature selection module is used to obtain effective working information. The machine state real-time monitoring chart construction module is used to calculate and construct a machine state variation reference chart using a machine learning algorithm. The machine state variation reference chart is a relationship chart composed of the corresponding working time point and variation value as the vertical axis and horizontal axis respectively. The real-time monitoring display interface module is used to display the relationship chart on the display screen.

The second purpose of the present invention is to provide an intelligent monitoring system and method for machine equipment with a visual work information chart display function, which can mainly integrate real-time sensing of processing status and visual chart display of processing status information, so that users can more conveniently monitor the real-time processing and working conditions of the machine equipment, thereby effectively improving the quality and yield of production. The technical means to achieve the above-mentioned second purpose is to use a machine equipment work information sensing module to sense the work information of the machine equipment. The information capture module is used to capture and collect work information during the working period in a predetermined capture cycle to obtain the work information within the validity period. The information conversion module includes a feature extraction module, a feature selection module and a machine status real-time monitoring chart construction module. The feature extraction module is used to obtain the working information features. The feature selection module is used to obtain the effective working information. The machine state real-time monitoring chart construction module is used to construct the machine state variation reference chart by machine learning algorithm. The machine state variation reference chart is a relationship chart composed of the corresponding working time point and variation value as the vertical axis and horizontal axis respectively. The real-time monitoring display interface module is used to display the relationship chart on the display screen. The invention further includes a working information real-time monitoring chart construction module and a working condition real-time monitoring chart construction module; the working information real-time monitoring chart construction module uses the vibration information, temperature information and current information and the corresponding at least one period time domain as the vertical axis and the horizontal axis respectively to respectively construct a vibration information, temperature information and current information real-time monitoring relationship chart; the working condition real-time monitoring chart construction module uses the vibration information, temperature information and current information as the vertical axis and the horizontal axis respectively to respectively construct a vibration information, temperature information and current information real-time monitoring relationship chart; The information and the corresponding at least one cycle time domain are used as the vertical axis and the horizontal axis to construct a real-time monitoring relationship chart of the working condition, and a high point and a low point in a predetermined range above and below each working information value in the real-time monitoring relationship chart are used as the upper control point and the lower control point respectively, and the upper control point and the lower control point are connected to form an upper control line and a lower control line in the at least one cycle time domain and displayed in the real-time monitoring relationship chart of the working condition.

10: Machine equipment working information sensing module

20: Message capture module

30: Message conversion module

31: Feature extraction module

32: Feature selection module

33: Real-time monitoring chart construction module for machine status

34: Work message real-time monitoring chart construction module

35: Construction module for real-time monitoring chart of working conditions

40: Real-time monitoring display interface module

41: Electronic display screen

42: Reference chart of machine variation status

43: Work message real-time monitoring relationship diagram

44: Real-time monitoring relationship chart of working conditions

50:Machinery and equipment

50a: Machine equipment listing module

60: Cloud connection module

70: Digital input module

71: Controller module

Figure 1 is a block diagram of the structure of a specific embodiment of the present invention.

Figure 2 is a schematic diagram of the application implementation of the preferred implementation structure of the present invention.

Figure 3 is a block diagram of the structure of a preferred embodiment of the present invention.

Figure 4 is a schematic diagram of the process implementation of the machine learning algorithm of the present invention.

Figure 5 is a schematic diagram of a specific implementation of the SMB system architecture of the present invention.

Figure 6 is a schematic diagram of the screen display of the real-time monitoring display interface module of the present invention.

Figure 7 is a time domain signal graph aggregated and output by the time domain algorithm of the present invention.

In order to allow the Honorable Review Committee to further understand the overall technical features of the present invention and the technical means for achieving the purpose of the present invention, a specific embodiment is provided with accompanying drawings for detailed description:

Please refer to FIGS. 1 to 3 and 6 to show that the first specific embodiment for achieving the first purpose of the present invention includes a machine equipment working information sensing module 10, a message acquisition module 20, a message conversion module 30, and a real-time monitoring display interface module 40. The machine equipment working information sensing module 10 is used to sense at least one working information generated by the machine equipment 50 during a working period. The message acquisition module 20 is used to continuously acquire and collect the working information sensed by the machine equipment working information sensing module 10 during the working period with an acquisition cycle predetermined by a periodic control module, so as to obtain the working information within the validity period corresponding to at least one cycle time domain. The message conversion module 30 includes a feature extraction module 31, a feature selection module 32 and a device status real-time monitoring chart construction module 33. The feature extraction module 31 is used to extract features of a plurality of valid working messages using a plurality of extraction algorithms to obtain a plurality of valid working message features. The feature selection module 32 is used to analyze a plurality of valid working message features using a selection algorithm to obtain a plurality of valid working messages. The device state real-time monitoring chart construction module 33 is used to construct a device variation state reference chart 42 by using a machine learning algorithm to calculate the multiple valid working information obtained by the feature selection module 32. The device variation state reference chart 42 is a relationship chart composed of the corresponding working time point and variation value as the vertical axis and the horizontal axis respectively. The real-time monitoring display interface module is used to display the device variation state reference chart 42 on an electronic display screen 41 for real-time monitoring by the user.

Please refer to FIGS. 1 to 3 and 6 to show a second specific embodiment for achieving the second purpose of the present invention, which includes a machine equipment working information sensing module 10, a message acquisition module 20, a message conversion module 30, and a real-time monitoring display interface module 40 and other technical elements. The machine equipment working information sensing module 10 is used to sense at least one working information generated by the machine equipment during a working period. The message acquisition module 20 is used to continuously capture and collect the working information sensed by the machine equipment working information sensing module 10 during the working period with a capture cycle predetermined by a periodic control module, so as to obtain the working information within the validity period corresponding to at least one cycle time domain. The message conversion module 30 includes a feature extraction module 31, a feature selection module 32, a machine status real-time monitoring chart construction module 33, a work message real-time monitoring chart construction module 34, and a work condition real-time monitoring chart construction module 35. The feature extraction module 31 is used to extract the features of a plurality of valid work messages using a plurality of extraction algorithms to obtain a plurality of valid work message features. The feature selection module 32 is used to analyze a plurality of valid work message features using a selection algorithm to obtain a plurality of valid work messages. The device state real-time monitoring chart construction module 33 is used to construct a device variation state reference chart 42 by using a machine learning algorithm to calculate the plurality of valid working information obtained by the feature selection module 32. The device variation state reference chart 42 is a relationship chart with the corresponding working time point and variation value as the vertical axis and the horizontal axis respectively. The real-time monitoring display interface module is used to display the device variation state reference chart 42 on an electronic display screen 41 for real-time monitoring by the user. The working information real-time monitoring chart construction module 34 uses the vibration information, temperature information, current information and their corresponding at least one period of time domain as the vertical axis and horizontal axis respectively to construct a working information real-time monitoring relationship chart 43 including the vibration information, temperature information and current information. The real-time working condition monitoring chart construction module 35 uses the vibration information, temperature information and current information and their corresponding at least one cycle time domain as the vertical axis and horizontal axis respectively to construct a real-time working condition monitoring relationship chart 44, and in the real-time working condition monitoring relationship chart 44, a high point and a low point in a predetermined range above and below each working information value are respectively used as an upper control point and a lower control point, and the upper control point and the lower control point are respectively connected to form an upper control line and a lower control line in at least one cycle time domain and displayed in the real-time working condition monitoring relationship chart 44.

Please refer to FIG. 6 , the present invention further includes a machine equipment listing module 50a, which is used to list the machine equipment 50 in operation on the electronic display screen 41, so that the user can select a machine equipment 50 through an operation interface 410, and synchronously start the real-time monitoring display interface module 40 to display the machine variation state reference chart 42, the work information real-time monitoring relationship chart 43 and the work condition real-time monitoring relationship chart 44 on the electronic display screen 41.

Please refer to FIGS. 1-2 and 4-6. The information acquisition module 20 acquires and collects the vibration information, temperature information and current information sensed by the machine equipment working information sensing module 10 to obtain the corresponding vibration information, temperature information and current information within the validity period in at least one cycle time domain; the feature extraction module 31 extracts the features of the vibration information, temperature information and current information within the validity period by using a plurality of extraction algorithms to obtain the features of the vibration information, temperature information and current information within the validity period; the feature selection module 32 selects the algorithm to analyze the vibration information. The vibration information, temperature information and current information characteristics are used to obtain effective vibration information, temperature information and current information; the machine state real-time monitoring chart construction module 33 calculates the effective vibration information, temperature information and current information with the machine learning algorithm to construct a machine variation state reference chart 42. The machine variation state reference chart 42 is a relationship chart composed of the corresponding working time point and variation value as the vertical axis and the horizontal axis respectively. The real-time monitoring display interface 40 module displays the machine variation state reference chart 42 on the electronic display screen 41.

Specifically, the real-time monitoring chart construction module 33 of the machine state calculates the effective vibration information, temperature information and current information of the first workpiece using a machine learning algorithm to construct a reference baseline displayed in the machine variation state reference chart 42.

Specifically, feature selection is also called subset selection, which is mainly used in the field of machine learning. It is a process of combining machine learning algorithms and selecting effective features with discriminative capabilities from the original feature set based on specific performance evaluation indicators to determine the best feature subset to optimize the performance indicators. In short, feature selection is to filter out invalid, non-critical, and repeated or similar discriminative noise features without compromising the performance of the machine learning algorithm, and finally retain only the features that really affect the performance indicators, so as to achieve the purpose of reducing the dimension of the feature space. The selection algorithm used in the present invention can be a principal component analysis method PCA; or a correlation coefficient analysis method, and the following two feature selection algorithms are the better feature selection algorithms used in the present invention:

1. (PCA): Principal component analysis: It is a technique for simplifying a data set. It is a linear transformation. This transformation transforms the data into a new coordinate system so that the largest variance of any data projection is on the first coordinate, the second largest variance is on the second coordinate, and so on. Principal component analysis is often used to reduce the dimension of a data set while maintaining the characteristics of the data set that contribute the most to the variance.

2. (Correlation Coefficient): Correlation coefficient analysis method: It is mainly used to explore the problem between two variables. Accurate correlation analysis produces a correlation coefficient, which is a number between -1 and +1. If it is +1, it means that the two variables have a complete positive linear correlation. If it is -1, it means that the two variables have a complete negative linear correlation. If the correlation coefficient approaches 0, it means that the two variables have no linear correlation.

Specifically, as shown in FIG. 4 , the machine learning algorithm used in the present invention may be a plurality of machine learning algorithms, including a first machine learning algorithm, a second machine learning algorithm, or a third machine learning algorithm. The first machine learning algorithm, the second machine learning algorithm, and the third machine learning algorithm are different machine learning algorithms designed to meet user needs.

The first machine learning algorithm of the present invention is implemented by using a neural network algorithm. The second machine learning algorithm is implemented by using a Gaussian mixture model algorithm. The third machine learning algorithm is implemented by using a recurrent neural network algorithm RNN; or one of the long short-term memory model algorithms LSTM.

In a specific embodiment of the present invention, the machine equipment working information sensing module 10 can be an accelerometer, a thermometer or an ammeter. The sensing object of the accelerometer can be a processing platform, a dividing plate or a processing head for mounting a processing tool of the machine equipment.

In a specific embodiment of the present invention, the sensing object of the thermometer can be at least one of a spindle of the machine equipment 50, a bearing of the spindle, a motor for driving the spindle, and a machining tool driven by the spindle.

In a specific embodiment of the present invention, the sensing object of the ammeter (i.e., the current sensing module shown in FIG5 ) can be a motor used to drive a main shaft of the machine equipment 50.

In addition, please refer to the embodiment shown in Figures 4-5. The digital input module 70 is triggered by the three-color light of the external device to generate an input signal to the controller module 71, and then the controller module 71 activates the machine equipment working information sensing module 10, so that the machine equipment working information sensing module 10 can start sensing work.

Through the above specific implementation description, the present invention does have the following characteristics:

1. The present invention can indeed be implemented by integrating real-time sensing and artificial intelligence computing technologies, allowing users to more conveniently monitor machine wear data and machine wear monitoring relationship diagrams and other variation status information in real time, thereby effectively obtaining warning information about upcoming machine variations in advance, thereby effectively reducing the probability of machine failure.

2. The present invention has the function of visualizing work information chart display, which can integrate real-time sensing of processing status and visualizing chart display of processing status information, so that users can more conveniently monitor the real-time processing and working conditions of machine equipment, thereby effectively improving the quality and yield of production.

The above is only a feasible implementation example of the present invention and is not intended to limit the patent scope of the present invention. Any equivalent implementation based on the content, features and spirit described in the following claims shall be included in the patent scope of the present invention. The structural features of the present invention specifically defined in the claims are not seen in similar articles and are practical and progressive. They have met the requirements for a creation patent. Therefore, an application is filed in accordance with the law. I sincerely request the Jun Bureau to grant the patent in accordance with the law to protect the legitimate rights and interests of the present applicant.

10: Machine equipment working information sensing module

20: Message capture module

30: Message conversion module

31: Feature extraction module

32: Feature selection module

33: Real-time monitoring chart construction module for machine status

34: Work message real-time monitoring chart construction module

35: Construction module for real-time monitoring chart of working conditions

40: Real-time monitoring display interface module

41: Electronic display screen

42: Reference chart of machine variation status

43: Work message real-time monitoring relationship diagram

44: Real-time monitoring relationship chart of working conditions

50:Machinery and equipment

Claims (6)

A machine equipment intelligent monitoring system includes: a machine equipment working information sensing module, which is used to sense a plurality of working information generated by a machine equipment during a working period, and the plurality of working information includes at least two of vibration information, temperature information and current information; a information acquisition module, which is used to continuously acquire and collect the plurality of working information sensed by the machine equipment working information sensing module during the working period with an acquisition cycle predetermined by a periodic control module, so as to obtain a plurality of valid period working information corresponding to at least one cycle time domain; A message conversion module, comprising: a feature extraction module, which is used to extract the features of the plurality of valid working messages by using a plurality of extraction algorithms to obtain a plurality of valid working message features; a feature selection module, which is used to analyze the plurality of valid working message features by using a selection algorithm to obtain a plurality of valid working messages; and a machine state real-time monitoring chart construction module, which is used to calculate the plurality of valid working messages obtained by the feature selection module by using at least one machine learning algorithm to construct a machine state variation reference chart, the machine state variation reference chart The reference chart is a relational chart composed of the corresponding working time points and the variation values as the vertical axis and the horizontal axis respectively; and a real-time monitoring display interface module, which is used to display the machine variation state reference chart on an electronic display screen for real-time monitoring by the user; wherein, it further includes a working information real-time monitoring chart construction module and a working condition real-time monitoring chart construction module; the working information real-time monitoring chart construction module uses the vibration information, temperature information and current information and the corresponding at least one cycle time domain as the vertical axis and the horizontal axis respectively to respectively construct the vibration information, temperature information and working condition real-time monitoring chart. and current information real-time monitoring relationship chart; the real-time working condition monitoring chart construction module uses the vibration information, temperature information and current information and the corresponding at least one cycle time domain as the vertical axis and horizontal axis respectively to construct a real-time working condition monitoring relationship chart, and in the real-time working condition monitoring relationship chart, a high point and a low point in a predetermined range above and below each working information value are respectively used as an upper control point and a lower control point, and the upper control point and the lower control point are respectively connected to form an upper control line and a lower control line in the at least one cycle time domain and displayed in the real-time working condition monitoring relationship chart. The machine equipment intelligent monitoring system as described in claim 1 further includes a machine equipment listing module, which is used to list the machine equipment in the working period on the electronic display screen, so that the user can select one of the machine equipment through an operation interface, and synchronously start the real-time monitoring display interface module to display the machine variation state reference chart, the working information real-time monitoring relationship chart and the working condition real-time monitoring relationship chart on the electronic display screen. The machine equipment intelligent monitoring system as described in claim 1, wherein the message acquisition module captures and collects the vibration message, temperature message and current message sensed by the machine equipment working message sensing module to obtain the corresponding vibration message, temperature message and current message within the validity period in at least one cycle time domain; the feature extraction module extracts the features of the vibration message, temperature message and current message within the validity period by using the plurality of extraction algorithms to obtain the features of the vibration message, temperature message and current message within the validity period; the feature selection module extracts the features of the vibration message, temperature message and current message within the validity period by using the selection algorithms; The algorithm analyzes the characteristics of the vibration information, temperature information and current information to obtain effective vibration information, temperature information and current information; the machine state real-time monitoring chart construction module calculates the effective vibration information, temperature information and current information with the machine learning algorithm to construct a machine variation state reference chart, which is a relationship chart composed of the corresponding working time point and variation value as the vertical axis and horizontal axis respectively; the real-time monitoring display interface module displays the machine variation state reference chart on the electronic display screen. A machine equipment intelligent monitoring method includes: providing a machine equipment working information sensing module, a information acquisition module, a information conversion module and a real-time monitoring display interface module, wherein the information conversion module includes a feature extraction module, a feature selection module and a machine state real-time monitoring chart construction module; using the machine equipment working information sensing module to sense a plurality of working information generated by a machine equipment during a working period, wherein the plurality of working information includes at least two of vibration information, temperature information and current information; using the information acquisition module to control the acquisition cycle predetermined by a periodic control module within the working period; Continuously capturing and collecting the plurality of working messages sensed by the machine equipment working message sensing module to obtain a plurality of valid working messages corresponding to at least one cycle time domain; extracting the features of the plurality of working messages within the validity period by the feature extraction module of the message conversion module using a plurality of extraction algorithms to obtain a plurality of working message features within the validity period; analyzing the plurality of working message features within the validity period by the feature selection module using a selection algorithm to obtain a plurality of valid working messages; and using the machine status real-time monitoring chart construction module to convert the plurality of valid working messages obtained by the feature selection module into A machine learning algorithm is used to calculate and construct a machine variation state reference chart, which is a relationship chart with corresponding working time points and variation values as the vertical axis and horizontal axis respectively; and the real-time monitoring display interface module is used to display the machine variation state reference chart on an electronic display screen for real-time monitoring by the user; wherein, it further includes a working information real-time monitoring chart construction module and a working condition real-time monitoring chart construction module; the working information real-time monitoring chart construction module uses the vibration information, temperature information and current information and the corresponding at least one cycle time domain as the vertical axis and horizontal axis respectively. The real-time monitoring relationship chart of vibration information, temperature information and current information is constructed respectively by using the axis; the real-time monitoring chart construction module uses the vibration information, temperature information and current information and the corresponding at least one cycle time domain as the vertical axis and horizontal axis to construct a real-time monitoring relationship chart of working condition, and in the real-time monitoring relationship chart of working condition, a high point and a low point in a predetermined range above and below each working information value are respectively used as an upper control point and a lower control point, and the upper control point and the lower control point are respectively connected to form an upper control line and a lower control line in the at least one cycle time domain and displayed in the real-time monitoring relationship chart of working condition. The intelligent monitoring method for machine equipment as described in claim 4, wherein the selection algorithm is selected from one of the PCA and the correlation coefficient analysis method. As described in claim 4, the machine equipment intelligent monitoring method, wherein the machine state real-time monitoring chart construction module calculates the effective vibration information, temperature information and current information sensed by the machine equipment during the first workpiece processing using the machine learning algorithm to construct a reference baseline displayed in the machine variation state reference chart.

Images