Driving sound diagnosis system, driving sound diagnosis method, machine learning device, storage medium and storage device of driving sound diagnosis system
By designing a driving sound diagnostic system, including driving sound detection, operating state detection and feature point extraction, the problem of difficulty in determining the cause of abnormal position or speed of the actuator or the driven machinery in the prior art is solved, and the accuracy of the driving sound is achieved Diagnosis and versatility.
Patent Information
- Application Number
- CN201980097381.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-06-17
- Filing Date
- 2019-12-03
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2039-12-03
AI Technical Summary
The prior art is difficult to determine the cause of abnormal position or speed of the actuator or the driven machine, resulting in insufficient accuracy in the diagnosis of the drive sound.
A driving sound diagnostic system is designed, including a driving sound detection unit, an operating state detection unit, a sound vibration time series spectrum acquisition unit, a feature point extraction unit and a key determination unit, through detection of sounds or generated by an actuator or driven machine Vibration, calculate the spectrum, extract feature points, and determine the key factors of driving sound based on feature point data.
The ability to accurately determine the position or speed cause of abnormality of the actuator or driven machinery improves the accuracy and versatility of driving sound diagnosis, and is suitable for different mechanical structures and driving modes.
Smart Images

Figure CN113950615B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a driving sound diagnosis system for diagnosing driving sound of an actuator or a driven machine driven by the actuator, a driving sound diagnosis method, and a machine learning device of the driving sound diagnosis system. Background Art
[0002] Generally, it is known that in a device having a power source such as an electric motor, a large amount of information related to the state of the power source and the driven object is contained in the driving sound of the power source or the driven object of the power source, that is, the mechanical. For example, when a certain abnormality occurs in the power source or the driven object, a sound or vibration different from that in normal times will occur. Therefore, a diagnostic device is known that diagnoses whether the sound or vibration is abnormal. However, when the device is operated for the first time, or when the device is driven after the mechanical structure and driving mode of the device have just been changed, it is difficult to immediately determine whether the driving sound is normal. Therefore, a technology is required to obtain the sound or vibration generated by the device, that is, the driving sound, through a sensor, and easily determine the cause of the driving sound. In particular, a technology is required to perform diagnosis using the driving sound more easily and universally without limiting the mechanical structure and settings of the device.
[0003] Patent document 1 discloses a technique for measuring the sound or vibration generated by a device having a rotating instrument and determining whether there is an abnormality in the device or the cause of the abnormality. According to this technique, first, before the actual operation of the device, a frequency- and time-related feature group is obtained in advance according to the category of abnormal cause for the sound generated by the power source when the abnormal sound occurs and the machine driven by the power source, that is, the driven machine. The frequency-related feature is the frequency at which a vertex is generated from the time change of the spectrum obtained by performing a short-time Fourier transform on the time series data of the measured driving sound. The time-related feature is the time interval at which a vertex is generated for each frequency feature quantity. Moreover, during the actual operation of the device, the frequency- and time-related features obtained in the actual operation by the same method are compared with the frequency- and time-related features obtained in advance, thereby determining whether there is an abnormality and the cause.
[0004] Patent Document 1: Japanese Patent Application Laid-Open No. 10-274558 Summary of the invention
[0005] However, in the technology described in Patent Document 1, the presence or absence of an abnormality in the device and the cause thereof are determined only based on the data obtained by measuring the driving sound generated by the actuator or the driven machine. Therefore, there is a problem that it is difficult to determine the cause of the abnormality that depends on the position or speed of the actuator or the driven machine.
[0006] The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to provide a drive sound diagnosis system capable of identifying the cause of an abnormality that depends on the position or speed of an actuator or a driven machine.
[0007] In order to solve the above-mentioned problems and achieve the purpose, the driving sound diagnosis system involved in the present invention includes a driving sound detection unit, an operation state detection unit, a sound vibration time series spectrum acquisition unit, a feature point extraction unit and a factor determination unit. The driving sound detection unit detects the driving sound generated by the actuator or the driven machine driven by the actuator, that is, the sound or mechanical vibration. The operation state detection unit acquires the driving position, driving speed or force generated by the actuator according to the time series. The sound vibration time series spectrum acquisition unit calculates the spectrum corresponding to each time of the detected time series data of the driving sound, that is, the sound vibration data, and outputs the time series spectrum obtained by associating the power of the calculated spectrum with the frequency and time. The feature point extraction unit extracts the point where the power of the time series spectrum meets the specified conditions with respect to the waveform of the frequency and time as a feature point, and outputs the feature point data obtained by grouping the frequency, time, waveform of the feature point and the driving position, driving speed or force generated by the actuator at the time of the feature point, that is, the operation data. The cause determination unit determines the cause of the detected driving sound by comparing the cause determination condition and the numerical value of the characteristic point data. The cause determination condition determines, for the cause of the driving sound, i.e., each phenomenon occurring in the actuator or the driven machine, a first numerical range when a combination of at least one of the frequency and time of the characteristic points accompanying the phenomenon and the driving position, driving speed or force generated by the driving of the actuator at the time, i.e., the operating data, is set as multidimensional data.
[0008] Effects of the Invention
[0009] The drive sound diagnostic system according to the present invention has an effect of being able to identify the cause of an abnormality that depends on the position or speed of an actuator or a driven machine. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Figure 1 This is a block diagram showing an example of the functional configuration of the driving sound diagnostic system according to the first embodiment.
[0011] Figure 2 This is a diagram showing an example of a hardware configuration when the drive sound diagnostic system according to the first embodiment is applied to an elevator.
[0012] Figure 3 This is a block diagram showing an example of the hardware configuration of the arithmetic processing unit according to the first embodiment.
[0013] Figure 4 This is a schematic diagram showing the embodiment 1. Figure 2 A block diagram showing an example of the functional structure of the arithmetic processing unit.
[0014] Figure 5 This is a flowchart showing an example of a processing procedure of the driving sound diagnostic method according to the first embodiment.
[0015] Figure 6 This is a diagram showing an example of sound vibration data according to the first embodiment.
[0016] Figure 7 This is a diagram showing an example of operating data according to the first embodiment.
[0017] Figure 8 This is a flowchart showing an example of a procedure of synchronization processing of sound vibration data and operation data according to the first embodiment.
[0018] Fig. 9 This is a diagram for explaining a procedure for synchronizing the sound vibration data and the operation data according to the first embodiment.
[0019] Fig.10 It means that Figure 7 An example of segmenting the operation data by operation mode is shown in FIG.
[0020] Fig.11 It is a 3D graph involving the three axes of time, frequency and power. Figure 6 FIG. 1 is a diagram showing an example of time-series spectrum data after frequency conversion of sound vibration data.
[0021] Fig.12 It is a 3D graph showing the three axes of time, frequency and power. Fig.11 A diagram of an example of the vertices of sounds extracted from the time series spectrum data.
[0022] Fig.13 This is a diagram showing an example of a hardware configuration when the drive sound diagnostic system according to the second embodiment is applied to a pickup unit.
[0023] Fig.14 This is a block diagram schematically showing an example of the functional configuration of a server device according to the second embodiment.
[0024] Fig.15 This is a diagram showing an example of recording of device data according to the second embodiment.
[0025] Fig.16This is a block diagram schematically showing an example of a functional configuration of a user terminal according to the second embodiment.
[0026] Fig.17 This is a block diagram showing an example of the hardware configuration of a server device and a user terminal.
[0027] Fig.18 This is a block diagram showing an example of the functional configuration of a factor determination unit in the driving sound diagnostic system according to the third embodiment.
[0028] Fig.19 This is a block diagram showing an example of the functional configuration of a machine learning device of a drive sound diagnosis system according to a fourth embodiment.
[0029] Fig. 20 This is a block diagram schematically showing an example of the functional configuration of a server device according to the fourth embodiment.
[0030] Fig.21 This is a block diagram schematically showing an example of a functional configuration of a user terminal according to the fourth embodiment. DETAILED DESCRIPTION
[0031] Hereinafter, a driving sound diagnosis system, a driving sound diagnosis method, and a machine learning device for a driving sound diagnosis system according to embodiments of the present invention will be described in detail with reference to the drawings. However, the present invention is not limited to these embodiments.
[0032] Implementation method 1.
[0033] Figure 1 1 is a block diagram showing an example of the functional structure of the driving sound diagnosis system according to Embodiment 1. The driving sound diagnosis system 10 includes a driving sound detection unit 11, an operation state detection unit 12, a time synchronization unit 13, an operation mode extraction unit 14, a sound vibration time series spectrum acquisition unit 15, a feature point extraction unit 16, an operation vibration extraction unit 17, and a factor determination unit 18.
[0034] The driving sound detection unit 11 detects the sound or vibration generated by the actuator and the machine driven by the actuator, that is, the driven machine. Hereinafter, the sound or vibration generated by the actuator and the driven machine is referred to as driving sound. The driving sound detection unit 11 is a sensor that detects sound or vibration. A vibration sensor such as a microphone or an accelerometer that detects sound is an example of the driving sound detection unit 11. The driving sound detection unit 11 can detect the sound generated by the driving instrument that drives the actuator.
[0035] The operating state detection unit 12 obtains the operating state of the actuator connected to the driving instrument in a time series. The operating state of the actuator includes the driving position, driving speed, or force generated by the actuator. A motor is an example of an actuator. An encoder for detecting the rotation angle of the motor, a linear scale for detecting the position of the linear motor, a position sensor, a displacement meter, a distance sensor, a speed detector, a current detector, an acceleration sensor, a gyro sensor, and a force sensor are examples of the operating state detection unit 12.
[0036] The time synchronization unit 13 synchronizes the time of the sound vibration data, which is the time series data of the driving sound detected by the driving sound detection unit 11, and the time of the operation data, which is the time series data of the operation state detected by the operation state detection unit 12. In one example, the time synchronization unit 13 obtains the difference in time that serves as a reference for the sound vibration data and the operation data, corrects the time of the sound vibration data or the operation data using the difference, extracts data between the time points that are commonly obtained from the sound vibration data and the operation data, and thereby obtains synchronized sound vibration data and operation data.
[0037] An example of operating data is time series data of current, generated torque, position, and speed. The operating data may be information of a driving instrument or an estimated value based on a portion of detectable data. The position command, speed command, torque command, current command, and voltage command of the motor are stored by the motor drive device, and the drive sound diagnosis system 10 may obtain these commands from the motor drive device as operating data directly, or may obtain the operating data by comparison operation with detectable data or estimation operation including filtering processing.
[0038] In addition, the operation data is not limited to analog signals. Binary data that is turned on when the speed reaches the commanded speed is an example of the operation data. In addition, the operation data does not need to be a set of time and value. A function that uses the time representing the speed pattern at each time as an independent variable is an example of the operation data.
[0039] The operation mode extraction unit 14 divides the operation mode into two or more intervals separated by time based on the operation data obtained by the operation state detection unit 12. Specifically, the operation mode extraction unit 14 analyzes the operation data to distinguish the type of the operation state of the actuator, and extracts the interval of time when the operation state of a specific type of actuator is reached as the operation mode. The operation mode can be determined based on the operation data after being synchronized with the sound vibration data by the time synchronization unit 13, or it can be determined based on the operation data before being synchronized with the sound vibration data by the time synchronization unit 13.
[0040] The sound vibration time series spectrum acquisition unit 15 performs frequency conversion on the sound vibration data after the time synchronization by the time synchronization unit 13, and obtains the time series spectrum of the sound vibration data corresponding to each time. The time series spectrum groups the power of the calculated spectrum with the frequency and time. When performing frequency conversion, the sound vibration time series spectrum acquisition unit 15 only performs frequency conversion on the interval of a specific operation mode among the extraction results of the operation mode extraction unit 14, thereby selecting the time for performing frequency conversion. As an example, the operation mode in which the actuator is stopped and the possibility of measuring the driving sound is low can be excluded from the object of frequency conversion. In this case, the amount of data subjected to frequency conversion processing is reduced, so it is expected to reduce the memory used in the calculation and shorten the processing time.
[0041] The feature point extraction unit 16 extracts the point as a feature point when the power of the time series spectrum obtained by the sound vibration time series spectrum acquisition unit 15 satisfies the predetermined conditions with respect to the waveform of the frequency and time. The feature point extraction unit 16 groups the frequency, time, power of the feature point, the waveform of the feature point, and the operation data of the time of the feature point as feature point data.
[0042] The operation vibration extraction unit 17 extracts a vibration component from the time-synchronized operation data, and acquires vibration data including the frequency or amplitude thereof.
[0043] The cause determination unit 18 determines the cause of the driving sound by comparing the cause determination condition, which is a numerical range of feature points generated in association with each phenomenon determined as the cause of the driving sound, with the feature point data extracted by the feature point extraction unit 16. In this case, the cause determination condition may be a numerical range of vibration data generated in association with each phenomenon determined in advance as the cause of the driving sound. In this case, the cause determination unit 18 estimates the cause of the driving sound by comparing the cause determination condition with the feature point data extracted by the feature point extraction unit 16 and the vibration data extracted by the operation vibration extraction unit 17.
[0044] Furthermore, when the range of values included in the characteristic point is specified for each abnormality of the device, the range of values included in the characteristic point of the device according to the abnormality cause category must be obtained for all types of driven machines. In addition, when the structure of the driven machine is changed, the range of values included in the characteristic point of the device according to the abnormality cause category must also be obtained. There are many types of driven machines, and there are also many changes in the structure of the driven machines. Therefore, it is unrealistic to obtain the range of values included in the characteristic point of the device according to the abnormality cause category for all of them.
[0045] However, in the first embodiment, the range of values included in the characteristic points is determined for each phenomenon that occurs as a cause of the drive sound, not for each abnormality of the device. That is, the characteristic quantity is not specified by a specific driving mode such as the value of the rotational speed of the motor, but is expressed by a conditional expression that does not depend on the driving mode, for example, a conditional expression in which the rotational speed of the motor is set as a variable. Therefore, in the case of different driven machines or in the case of a change in the structure of the driven machine, if the cause of the drive sound is the same, the phenomenon that occurs is also the same, and the same cause judgment condition can be used regardless of the type or structure of the driven machine. As a result, compared with the case where the range of values included in the characteristic points is specified for each abnormality of the device, the time required for the preparation of the diagnosis can be shortened, and the diagnosis of the drive sound can be implemented universally.
[0046] The feature point data groups the frequency, time, power, waveform of the feature point, and operation data at the time of the feature point. Therefore, it includes not only information related to sound or vibration, but also information related to the position or speed of the actuator or driven machine. That is, in the factor determination unit 18, when determining the factor of the driving sound, the position or speed of the actuator or driven machine is also included in the determination, so the factor of the abnormality that depends on the position or speed of the actuator or driven machine can be easily determined.
[0047] In addition, Figure 1 In the embodiment, the time synchronization unit 13, the operation mode extraction unit 14, and the operation vibration extraction unit 17 may be appropriately included in the driving sound diagnosis system 10 according to the performance or device structure required for diagnosis, or may be removed from the driving sound diagnosis system 10. In one example, when the sound vibration data and the operation data are obtained by the same measuring instrument or device, an AD (Analog to Digital) converter with the same acquisition timing is used, thereby synchronizing the acquired data. That is, the sound vibration data and the operation data can be synchronized without providing the time synchronization unit 13. In this case, by removing the time synchronization unit 13, it is expected that the memory used in the driving sound diagnosis system 10 can be reduced and the processing time can be shortened.
[0048] Figure 2 FIG. 1 is a diagram showing an example of a hardware configuration when the drive sound diagnostic system according to Embodiment 1 is applied to an elevator. Figure 2As shown, the diagnosed object 100 includes a motor 110 as an actuator, a driven machine 120, and a driving device 130 that drives the motor 110. In addition, the diagnosed object 100 is provided with a driving sound diagnosis system 10A including a driving sound detection unit 11, an operation state detection unit 12, and a calculation processing unit 140.
[0049] The motor 110 is a servo motor that controls the current of the armature winding based on the difference between the drive command and the rotation information. The motor 110 can be any actuator that generates power by receiving energy or an electrical signal from an instrument that drives the motor 110. In this example, the motor 110 is controlled by the drive device 130. However, the state quantity of the motor 110 controlled by the drive device 130 is not limited to the current. Oil pressure, air pressure, heat, and ultrasonic waves are examples of the state quantity of the motor 110 controlled by the drive device 130. In addition, the motor 110 is not limited to generating a rotational force, and can also be a linear motor driven in the translation direction.
[0050] The driven machine 120 is a lifter that transports a driven object in the up-down direction according to the rotation of the motor 110. The driven machine 120 includes a bracket 121 that fixes the motor 110 to a frame, a gear box 122 that amplifies the rotational force generated by the motor 110 and transmits it to a ball screw 123, the ball screw 123 that converts the rotation of the motor 110 into a movement in the up-down direction, and a coupling 124 that connects the gear box 122 and the ball screw 123.
[0051] In addition, the elevator has: a slider 125, which is driven in the up and down directions by the rotation of the ball screw 123; a workbench 126, which is fixed to the slider 125 and carries the workpiece; a linear guide 127, which guides the movement of the slider 125 in the up and down directions so that it can slide freely; and a bracket 128, which fixes the ball screw 123 to the linear guide 127 via a bearing so that it can rotate freely.
[0052] Here, the driven machine 120 is exemplified as a lift having a ball screw 123, but the driven machine 120 may be any machine as long as it generates sound or vibration in response to the rotation of the motor 110. A screw, a belt, a gear, a cam, a link mechanism, a bearing, or a seal, or a machine combining these elements is an example of the driven machine 120.
[0053] The drive device 130 is connected to the motor 110 via a cable 151. The drive device 130 includes a motor driver 131, a motor control device 132, and a display 133. The motor driver 131 supplies power for driving the motor 110 to the motor 110. In addition, the motor driver 131 drives the motor 110 according to a drive command transmitted from the motor control device 132.
[0054] The motor control device 132 sends an electrical signal such as a command position or command speed to the motor driver 131, whereby the motor driver 131 controls the amount and timing of the current supplied to the motor 110. The display 133 displays various states of the entire system including the driving sound diagnosis system 10A and the diagnosis object 100, and notifies the user.
[0055] In the first embodiment, the motor driver 131 is connected to the driving sound detector 11 and the operating state detector 12 , and has a function of relaying communication between the driving sound detector 11 and the operating state detector 12 and the motor control device 132 .
[0056] The motor control device 132 is a controller that gives a drive command including a position or speed pattern of the motor 110 to the motor driver 131. The motor control device 132 is a control device including a PLC (Programmable Logic Controller), a CPU (Central Processing Unit) for motor driving, a DSP (Digital Signal Processor), a pulse generator, and the like.
[0057] Furthermore, the display 133 acquires the state of the entire system including the driving sound diagnosis system 10A from at least one of the motor driver 131, the motor control device 132, or the calculation processing unit 140 through communication, and displays it in a form that is easy for the user to see. The display 133 may have a built-in liquid crystal display. By including the display 133 in the system, after the calculation processing unit 140 performs the diagnosis of the driving sound, the result of the diagnosis of the driving sound is displayed on the display 133 through communication, and the user can be informed in an easily understandable manner.
[0058] The driving device 130 may be any device that drives at least one electric motor 110 , and may be a part of the present embodiment or a combination of a plurality of the parts.
[0059] The calculation processing unit 140 is a processing device that can execute the diagnostic processing of the drive sound diagnosis system 10A described later as software. In the first embodiment, the CPU of the microcomputer built in the motor control device 132 realizes the function of the calculation processing unit 140 . Figure 3 1 is a block diagram showing an example of the hardware configuration of the calculation processing unit according to Embodiment 1. The calculation processing unit 140 includes a processor 141 and a memory 142. The processor 141 and the memory 142 are connected via a bus 143. A CPU and a GPU (Graphics Processing Units) are examples of the processor 141.
[0060] Figure 4 This is a schematic diagram showing the embodiment 1. Figure 2 The block diagram of an example of the functional structure of the operation processing unit. The operation processing unit 140 is a software executed by the CPU of the microcomputer and includes a time synchronization unit 13, an operation mode extraction unit 14, a sound vibration time series spectrum acquisition unit 15, a feature point extraction unit 16, an operation vibration extraction unit 17, and a factor determination unit 18. Figure 1 and Figure 2 The same components are denoted by the same reference numerals, and their description is omitted.
[0061] exist Figure 2 , the structure in which the operation processing unit 140 is built into the motor control device 132 is shown, but the embodiment is not limited to this structure. The operation processing unit 140 may be built into the motor driver 131, which is another device of the drive device 130, or may be built into the display 133. In addition, another microcomputer may be installed in the motor control device 132 as the operation processing unit 140. Furthermore, it may be connected to the drive device 130 as another independent device.
[0062] The driving sound detection unit 11 is a microphone that detects the driving sound generated by the diagnosis object 100. The detected driving sound is sent to the calculation processing unit 140 via the cable 152 and the driving device 130. Figure 2 In the example of , the driving sound detection unit 11 is fixed to the elevator. However, the driving sound detection unit 11 only needs to be able to detect the driving sound of the diagnosis object 100, and the installation method is not limited to being fixed to the driven machine 120. The driving sound detection unit 11 may be fixed to the motor 110, or may be separated from the driven machine 120 at a distance that allows the driving sound to be detected.
[0063] When the sound source of the phenomenon to be diagnosed is limited, the range of the sound collection target or the range that can be collected can be limited by placing the microphone adjacent to the sound source or using a directional microphone. By limiting the sound collection, the surrounding noise that hinders the diagnosis can be reduced, thereby reducing misdiagnosis.
[0064] In addition, a plurality of microphones may be used as the driving sound detection unit 11. By collecting the same sound or vibration through a plurality of microphones, it is possible to reduce erroneous diagnosis due to noise.
[0065] Furthermore, a sound recorder may be used to record the driving sound generated by the diagnostic object 100 in a time series as sound vibration data. A smart phone or a sound recorder is an example of a sound recorder. In addition, an animation may be captured by a camera or the like, and only the sound portion may be extracted as sound vibration data.
[0066] In addition, a storage unit for storing the driving sound generated in a time series as sound vibration data may be provided in the driving sound detection unit 11, and the stored sound vibration data may be sent to an upper computing device as needed. In this case, when the driving sound diagnosis is not performed, the communication between the driving sound detection unit 11 and the computing unit 140 is not performed, and when the driving sound diagnosis is performed, the time series sound vibration data stored in the driving sound detection unit 11 is aggregated and sent. By setting the structure as described above, the processing involved in the communication between the driving sound detection unit 11 and the computing unit 140 can be reduced.
[0067] The operating state detection unit 12 is an encoder mounted on the motor 110 to detect the rotation angle of the motor 110. The data of the rotation angle detected by the encoder is sent to the calculation processing unit 140 via the cable 153 and the drive device 130 as the operating state.
[0068] The installation location of the operating state detection unit 12 is not limited to the case where it is fixed to the motor 110. The operating state detection unit 12 may be fixed to the driving unit of the machine driven by the motor 110, that is, the driven machine 120. Similar to the driving sound detection unit 11, a storage unit that stores the detected time-series operating state as operating data may be provided in the operating state detection unit 12, and the stored operating data may be sent to a higher-level computing device as needed.
[0069] The calculation processing unit 140 receives the driving sound detected by the driving sound detection unit 11 and the operating state detected by the operating state detection unit 12 via the cables 152 and 153 , and diagnoses the cause of the driving sound.
[0070] In order to exchange corresponding data with the drive sound detector 11 and the operating state detector 12 , the arithmetic processing unit 140 is preferably connected to the drive sound detector 11 and the operating state detector 12 via a high-speed network with no data delay.
[0071] Next, the operation of the diagnostic object 100 will be described. In order to transport a workpiece (not shown) by the elevator, the user of the elevator of the driven machine 120 inputs a drive instruction of the motor 110 into the motor control device 132. The motor control device 132 sends a drive instruction to the motor driver 131 based on information including the drive mode and action timing input by the user. The motor driver 131 controls the motor drive current in accordance with the received drive instruction to drive the motor 110. The elevator, that is, the motor 110 which is the power source, rotates, thereby rotating the ball screw 123, and the slider 125 connected to the ball screw 123 and the workbench 126 connected to the slider 125 move up and down.
[0072] In one example, the user of the lifter places the workpiece on the worktable 126 while the position of the worktable 126 is moving vertically below the linear guide 127 through the above-mentioned operation. The worktable 126 is rotated by the motor 110, thereby moving vertically above the linear guide 127. The lifter transports the workpiece placed on the worktable 126 in accordance with this movement.
[0073] The diagnostic object 100 generates a driving sound when driven by the motor 110. Specifically, the diagnostic object 100 generates a driving sound caused by torque pulsation of the motor 110, translation and torsional rigidity of the ball screw 123, connection rigidity of the coupling 124, meshing rigidity of the gears, movement, deformation or collision of the machine, sliding friction between the linear guide 127 and the slider 125, etc. The driving sound generated by the diagnostic object 100 is detected by the driving sound detection unit 11, and the rotation angle of the motor 110 is detected by the operating state detection unit 12, and is transmitted to the motor driver 131, the motor control device 132, and the calculation processing unit 140 via the cables 152 and 153. However, the transmission method does not necessarily have to be wired, and may be wireless or via a recording medium.
[0074] The display 133 communicates with the motor control device 132 as appropriate, and displays various information required by the user including the rotation angle of the motor 110 detected by the operating state detection unit 12 and the presence or absence of abnormality in the entire system of the diagnosis target 100 .
[0075] In the first embodiment, the time synchronization unit 13, the operation mode extraction unit 14, the sound vibration time series spectrum acquisition unit 15, the feature point extraction unit 16, the operation vibration extraction unit 17, and the cause determination unit 18 are configured on a single hardware, namely, the operation processing unit 140, but they may be divided into separate hardware. As an example, the time synchronization unit 13 may be separated from the operation processing unit 140 and may be configured as software on a microcomputer of the motor driver 131. In this case, the motor driver 131 that performs control processing at a higher speed than the motor control device 132 performs the time synchronization processing with strict time requirements, and the operation mode extraction unit 14, the sound vibration time series spectrum acquisition unit 15, the feature point extraction unit 16, the operation vibration extraction unit 17, and the cause determination unit 18 with relatively less time restrictions are performed by the operation processing unit 140 of the motor control device 132. As a result, the performance required for the operation processing unit 140 can be reduced, and even a device with low operation processing capability can realize the diagnosis of the cause of the drive sound.
[0076] In addition, the realization of each function is not limited to the software realization on the CPU of the microcomputer, and electronic circuits such as ASIC (Application Specific Integrated Circuits), FPGA (Field Programmable Gate Array) or CPLD (Complex Programmable Logic Device) may be used.
[0077] Next, a description will be given of a procedure for diagnosing a driving sound in the driving sound diagnosis system 10A that uses the sound or vibration generated by the diagnosis target 100 as a diagnosis target. Figure 5 1 is a flowchart showing an example of the processing sequence of the drive sound diagnosis method involved in Embodiment 1. In Embodiment 1, as described above, when the driven machine 120 is driven by the driving of the motor 110, the drive sound is generated due to the movement, deformation, collision, rigidity, friction, etc. of the machine. In addition, the magnitude and frequency of the sound pressure of the drive sound are different depending on the driving method of the motor 110. Therefore, in Embodiment 1, the diagnosis method of the drive sound generated by the driven machine 120 is described.
[0078] The driving sound detection unit 11 of the driving sound diagnosis system 10A detects the driving sound (step S11 ), and the operating state detection unit 12 detects the operating state of the motor 110 (step S12 ).
[0079] The time synchronization unit 13 imports the driving sound from the driving sound detection unit 11 as time-series sound vibration data (step S13 ). Figure 6: is a figure which shows an example of the sound vibration data concerning Embodiment 1. In this figure, the horizontal axis shows time, and the vertical axis shows amplitude.
[0080] Furthermore, the time synchronization unit 13 imports the operating status from the operating status detection unit 12 as time-series operating data (step S14 ). Figure 7 1 is a diagram showing an example of operation data according to Embodiment 1. In this diagram, the horizontal axis represents time, and the vertical axis represents rotation angle, rotation speed, and current. Figure 7 , as operation data, a rotation angle Po of the motor 110 from a position serving as a reference of the ball screw 123 , a rotation speed w of the motor 110 , and a motor current i are shown.
[0081] Here, the acquisition of the sound vibration data and the acquisition of the operation data need to import the data at the same time. However, if it is the same time, the acquisition time of the imported data can be different, and the time when some data are obtained can be different from the same time.
[0082] In addition, regarding the frequency of the sound or vibration generated by the diagnostic object 100, when its frequency band is foreseen through some kind of insight, it is desirable to space out the sampling of the sound vibration data and the operation data until the frequency band of the object. By spacing out the processing, the amount of memory to be used and the processing time can be reduced. In the case of spacing out the processing, in order to perform frequency conversion of the sound vibration data by subsequent processing compared with the processing of the driving sound detection unit 11 and the operating state detection unit 12, it is desirable to perform filtering processing implemented by a noise removal filter. A low-pass filter, a high-pass filter, a band-pass filter, a notch filter, and a band-stop filter are examples of noise removal filters. In addition, as a noise removal filter, the exemplified filter can be applied alone or in plurality. As described above, the effect of reducing the foldback noise caused by spacing out can be expected.
[0083] Next, the time synchronization unit 13 performs synchronization processing of the sound vibration data and the operation data (step S15 ). Figure 8 This is a flowchart showing an example of a procedure of synchronization processing of sound vibration data and operation data according to the first embodiment.
[0084] First, the difference between the reference time of the sound vibration data and the operation data is obtained (step S31). Here, the reference time refers to the time set as time 0 in each data. When the sound vibration data and the operation data are not synchronized, the timing set as time 0 of each data is set separately, so even if the acquisition time of the data is the same, the timing actually obtained is sometimes different. Therefore, by finding the difference between the actual time of the timing set as time 0 in each data, the time of each data is corrected.
[0085] As a method for obtaining the difference in reference time, the following method is cited, that is, designing in such a way that the timing of obtaining data in advance becomes a certain time difference, and setting the certain time difference as the difference in reference time. In addition, the following method is cited, that is, accessing a common master clock in advance to set the acquisition timing of each data to be the same, and setting the difference in reference time to 0. Alternatively, the following method is also cited, that is, the time of the common master clock at the timing of starting data acquisition is stored as a timestamp, and the difference in the timestamps of the sound vibration data and the operation data is set as the difference in reference time.
[0086] Fig. 9 FIG. 2 is a diagram for explaining the procedure for synchronizing the sound vibration data and the operation data according to the first embodiment. Fig. 9 In the example, the time 0 of the sound vibration data Da is the acquisition time of the data, and the time 0 of the operation data Do is the acquisition time of the data. The time 0 of the sound vibration data Da is the reference time t1, and the time 0 of the operation data Do is the reference time t2. In this case, the difference Δt between the reference times is t2-t1.
[0087] Next, the acquisition time of each data is corrected based on the obtained reference time difference (step S32). Specifically, the obtained reference time difference is added to the acquisition time of each data acquired first among the sound vibration data and the operation data as a correction.
[0088] exist Fig. 9 In the example, the data acquired first is the sound vibration data Da. Therefore, the time obtained by adding the difference Δt of the reference time to the acquisition time t1 of the sound vibration data Da becomes the reference time t2 of the operation data.
[0089] return Figure 8 , the time intervals at which the sound vibration data and the operation data are commonly obtained are calculated (step S33). An example of a specific calculation method is the following method, that is, the time interval from the acquisition start time of the data acquired later in the aforementioned processing to the earlier time interval of the last acquisition time of the sound vibration data and the operation data is set as the time interval at which the sound vibration data and the operation data are commonly obtained. Fig. 9 In the example of , the time interval Δtc commonly acquired is from the time t2 when the operation data Do is acquired to the time t3 when the sound and vibration data Da is last acquired.
[0090] Then, the calculated data of the uncommon time that is not between the times that are commonly acquired is discarded (step S34). Fig. 9In the example of , the sound vibration data Da before time t2 and the operation data Do after time t3 are discarded. Thus, the sound vibration data and the operation data can be synchronized, and the synchronization process of the sound vibration data and the operation data is completed.
[0091] return Figure 5 The operation mode extraction unit 14 estimates the operation state of the driven machine 120 based on the time-synchronized operation data and extracts the operation mode (step S16). In the first embodiment, the operation mode of the driven machine 120 is divided into three modes: stopped, running at a constant speed, or accelerating or decelerating.
[0092] <Stop>
[0093] The section where the rotation of the motor 110 stops is defined as being stopped. Here, the rotation stop of the motor 110 means that the sum of the amount of change in the data of the rotation angle of the motor 110 detected by the operating state detection unit 12 converges to a specific threshold value within a certain period of time. At this time, the section where the sum of the amount of change in the data of the rotation angle converges to a specific threshold value is defined as a position stop section. The certain period of time included in the condition can be determined in accordance with the average value of the detection period of the sound vibration data and the operating data.
[0094] <Constant speed operation>
[0095] The interval in which the rotation speed of the motor 110 is constant and not stopped is defined as constant speed operation. Here, constant speed means that the sum of the change amount of the speed data converges to a specific threshold value within a certain period of time, as in the case of stopping. At this time, the interval in which the sum of the change amount of the speed data converges to the specific threshold value is defined as the interval in which the speed is constant and the time is longer than the predetermined time.
[0096] <Acceleration and deceleration>
[0097] The section that is neither stopped nor in constant speed operation is defined as acceleration / deceleration. With respect to the driving direction, the section when the acceleration is positive can be defined as acceleration, and the section when the acceleration is negative can be defined as deceleration. Fig.10 It means that Figure 7 FIG. 1 is a diagram showing an example of segmenting the operation data by the operation mode. In this example, the operation mode is acceleration / deceleration during the period from time 0 to t11, constant speed operation during the period from time t11 to t12, acceleration / deceleration during the period from time t12 to t13, and stop after time t13.
[0098] return Figure 5The sound vibration time series spectrum acquisition unit 15 performs frequency conversion on the sound vibration data synchronized by the time synchronization unit 13 and calculates the spectrum of the sound vibration data at each time (step S17). The spectrum of the sound vibration data is the spectrum data of the time series obtained by associating the power of the calculated spectrum with the frequency and time.
[0099] The frequency conversion is preferably performed by short-time Fourier transform (STFT) or wavelet transform. By obtaining a time series spectrum by these methods, the workload and processing of filter design etc. can be reduced.
[0100] Here, when the frequency band of the driving sound generated by the diagnostic object 100 is predicted by some insight, a filter that extracts the frequency components in the predicted frequency band before frequency conversion can be applied to the sound vibration data. By applying the filter, the driving sound can be diagnosed more accurately.
[0101] In the first embodiment, the sound vibration time series spectrum acquisition unit 15 performs frequency conversion on only the sound vibration data corresponding to the period of constant speed operation extracted by the operation pattern extraction unit 14 among the time synchronized sound vibration data, and discards the data of other periods.
[0102] Fig.11 It is a 3D graph that shows the time, frequency and power axes. Figure 6 FIG. 1 is a diagram showing an example of time series spectrum data after frequency conversion of sound vibration data. In this diagram, the vertical axis represents power, and two orthogonal axes in a plane perpendicular to the vertical axis represent time and frequency.
[0103] return Figure 5 The feature point extraction unit 16 extracts feature points whose frequency and time of power of the spectrum meet the prescribed conditions for the spectrum of the sound vibration data represented by the time series data obtained by the sound vibration time series spectrum acquisition unit 15 (step S18). In addition, the feature point extraction unit 16 generates feature point data related to the extracted feature points (step S19). The feature point data is obtained by grouping the frequency, time, power, waveform of the feature point and the operation data of the time of the feature point.
[0104] An example of a condition used when extracting a feature point by the feature point extraction unit 16 is a vertex, a peak, an edge line, and a saddle point. A case where a vertex is obtained as a feature point will be described.
[0105] First, the spectrum of the time series sound vibration data contains noise, so a low-pass filter is applied to the time axis and the frequency axis. The time constant of the low-pass filter is determined by the sampling period of the sound vibration data and the frequency conversion resolution. In addition, instead of using a low-pass filter, the Hilbert transform can be applied to the spectrum of the time series.
[0106] Next, a power threshold is determined for the spectrum of the time series after the low-pass filter is applied, and the spectrum is divided into a region where the power is greater than the threshold and a region where the power is less than the threshold. The central value of the spectrum of the time series is an example of a power threshold.
[0107] Then, from the point x=(t x , f x , p x ), get the next 4 points. Here, t x , f x , p x They respectively represent the time, frequency and power of point x.
[0108] (1) At frequency f x , get the time t x The next moment t x+1 The point x1(t x+1 , f x , p1)
[0109] (2) At frequency f x , get the time t x A moment before t x-1 The point x2(t x-1 , f x , p2)
[0110] (3) At the acquisition time t x , frequency is f x The next frequency f x+1 The point x3(t x , f x+1 , p3)
[0111] (4) At the acquisition time t x , frequency is f x The frequency f of a x-1 The point x4(t x , f x-1 , p4)
[0112] Next, the powers of the five points, which are the four points x1, x2, x3, and x4 mentioned above plus the original point x, are compared. When the power of point x is greater than that of the other four points among the five points, that is, when point x surrounded by the other four points x1, x2, x3, and x4 becomes the point with the highest power among the five points, point x is set as a candidate for the vertex. In addition, in this example, the case where the point with the highest power is extracted using point x and two points adjacent to point x in the direction of the time axis and two points adjacent to point x in the direction of the frequency axis is shown, but the implementation method is not limited to this. That is, the following method is sufficient, that is, using multiple points and point x that surround point x on the plane formed by the time axis and the frequency axis, when the power of point x surrounded by multiple points becomes the largest, point x is extracted as the point with the highest power.
[0113] Finally, the vertex candidates are arranged in descending order of power, and a certain number of points are extracted as maxima from the largest.
[0114] Here, the candidate points may be arranged in descending order of power, and a certain number of points obtained from the highest power may be used as vertices, or all candidate points whose power exceeds a predetermined threshold may be used as vertices. In addition, these methods may be combined.
[0115] By acquiring the peaks of the waveform of the power of the time series spectrum with respect to the frequency and time as feature points as described above, the amount of calculation for the cause diagnosis described later can be reduced by a method with a small amount of processing.
[0116] When no operating data is obtained at the time of the feature point, the operating data at the time closest to the feature point is used, or the operating data at the time of the feature point is interpolated based on the operating data at multiple times before and after the feature point, thereby determining the operating data at the time of the feature point.
[0117] Fig.12 It is a 3D graph showing the three axes of time, frequency and power. Fig.11 1 is a diagram showing an example of vertices of sound extracted from the time series spectrum data of . In this figure, vertices 1, 2, and 3 extracted by the method described above are shown.
[0118] return Figure 5 The operation vibration extraction unit 17 extracts the vibration component from the operation data after time synchronization and obtains the vibration data (step S20). The frequency, amplitude or phase of the vibration component is an example of vibration data. This process can be implemented independently from the sound vibration time series spectrum acquisition process of step S17.
[0119] An example of a method of extracting a vibration component is a method of obtaining a time from a peak of operation data to a next peak during constant speed operation.
[0120] Next, the factor determination unit 18 compares the feature point data generated in step S19 and the vibration data obtained in step S20 with the factor determination conditions registered for each factor of the driving sound, thereby determining the cause of the driving sound (step S21). Here, for each factor determination condition registered for each factor of the driving sound, with respect to the feature point data, for each phenomenon occurring in the motor 110 or the driven machine 120, which is a factor of the driving sound, a numerical range is determined when a combination of at least one of the frequency and time included in the feature point accompanying the phenomenon and the driving position, driving speed, or driving force generated by the actuator, i.e., the motor 110 at the above time, i.e., the operating data is set as multidimensional data. With respect to the vibration data, for each phenomenon occurring in the motor 110 or the driven machine 120 as a factor of the driving sound, a numerical range including the vibration component occurring with the phenomenon is determined.
[0121] The factor determination unit 18 compares the acquired feature point data with the numerical range for each factor, thereby determining the generation factor of the sound vibration data. In addition, the factor determination unit 18 compares the acquired vibration data with the numerical range for each factor, that is, compares the vibration component of the operation data with the apex of the sound, thereby determining the generation factor of the vibration data.
[0122] As an example, when dirt is attached to the linear guide 127, the friction between the slider 125 and the linear guide 127 increases and a scraping sound is generated, the scraping sound is generated only when the slider 125 is in a specific position interval. Therefore, when the position of the motor 110 at the time when the characteristic point is generated is a specific position, or is concentrated in a certain width interval P, the diagnosis object 100 is determined by the factor determination unit 18 as generating a sound in the interval P. Here, the width of the interval P can be determined based on the ratio of the difference between the maximum value and the minimum value of the data relative to the position in the time series operation data. In addition, the factor determination condition can be any condition that can check the distribution degree of the position of the motor 110 at the time of the characteristic point is concentrated in a specific position. For example, the number of characteristic points or their ratio that are separated from the interval P of a certain width, the average and dispersion of the position of the characteristic point at the time, or the correlation coefficient between the frequency and the position of the characteristic point are calculated, and it can be checked that the calculated value is within a predetermined range.
[0123] In addition, as another example, in the case of a center deviation of two shafts connected by the coupling 124, it is known that a characteristic sound is generated at a frequency twice the rotation speed of the coupling 124. Therefore, the rotation speed of the coupling 124 is calculated by multiplying the speed of the motor 110 from the operating data at the time of occurrence of the characteristic point by the conversion ratio of the gear box 122, and when the calculated rotation speed and the frequency of the characteristic point are in a positive proportional relationship of 2 times, the factor determination unit 18 determines that the center deviation of the two shafts connected by the coupling 124. In this case, instead of the speed of the motor 110, the current of the motor 110 may be obtained as the operating data, and the value may be accumulated to replace the speed.
[0124] Furthermore, as an example of other conditions, when the machine resonates mechanically by the drive of the motor 110, a resonant sound of a specific resonant frequency f is generated at a specific speed v that excites the mechanical resonance. Therefore, when the rotation speed of the motor 110 at the time when the characteristic point occurs is a specific speed or is concentrated in a certain width interval V, and the frequency of the characteristic point is a specific frequency or is concentrated in a certain width interval F, the diagnosis object 100 is determined by the cause determination unit 18 as generating a sound of the frequency f generated by mechanical resonance at the rotation speed v.
[0125] Furthermore, when the frequency of the vibration obtained by the operation vibration extraction unit 17 and the period of the characteristic point at which the frequency f is generated by the mechanical resonance are the same, the factor determination unit 18 determines that the mechanical resonance is excited by the vibration component of the motor 110 at a certain interval due to the rotation of the motor 110. That is, by comparing the vibration component of the operation data with the peak of the sound, the factor determination unit 18 determines that the mechanical resonance is periodically vibrated by the vibration component of the drive, and the resonance is excited.
[0126] As described above, the factor determination unit 18 uses the registered factor determination conditions to check the feature point data extracted by the feature point extraction unit 16, or the feature point data and the vibration data extracted by the operation vibration extraction unit 17, thereby determining the factor of the drive sound that depends on the position or speed of the actuator or the driven machine 120. The registered factor determination conditions can be checked as a binary tree search by sorting similar conditions in advance. This reduces the number of conditions checked for factor determination and shortens the determination time. The above is the end of the processing of the drive sound diagnosis method.
[0127] The drive sound diagnosis system 10, 10A according to the first embodiment synchronizes the time series sound vibration data related to the drive sound generated by the actuator or the driven machine 120 with the time series operation data related to the operation state of the actuator, and extracts the operation mode. In addition, feature points are extracted from the frequency spectrum of the time series sound vibration data obtained by performing time-frequency analysis on the sound vibration data, and feature point data is generated by grouping the frequency, time, power, waveform of the feature point and the operation data of the time of the feature point. Moreover, the feature point data is compared with the factor discrimination conditions prepared in advance, thereby discriminating the generation factor of the drive sound. As described above, the generation factor of the drive sound is diagnosed based on the sound vibration data and the operation data, so that the generation location of the drive sound caused by the attachment of foreign matter to the driven machine 120 can be determined, and the abnormal cause that depends on the position or speed of the actuator or the driven machine 120 can be easily determined. In addition, the cause of the frequency change of the sound generated according to the speed of the actuator or the driven machine 120 can also be discriminated. Furthermore, even when the entire device vibrates, it is possible to determine whether the sound vibration is caused by operation.
[0128] In addition, the driving sound diagnosis system 10, 10A diagnoses the cause of the sound or vibration based on the sound vibration data and operation data of the characteristic point and the time of occurrence. Therefore, even if there is a device that generates a loud operation sound near the diagnosis object, it is possible to suppress misdiagnosis based on the causal relationship between the time of occurrence and the operation data.
[0129] Furthermore, the drive sound diagnosis system 10, 10A diagnoses the cause of the sound generation by combining the frequency spectrum of the sound vibration data with the operation data, so when determining the change in the frequency spectrum of the sound vibration data, the influence of the change in the frequency spectrum of the sound vibration data caused by the change in the operation mode can be removed. In particular, when the operation mode changes according to the condition of the workpiece, the influence on the frequency spectrum of the sound vibration data can also be removed by the operation data, so the drive sound diagnosis system 10, 10A can perform appropriate diagnosis.
[0130] In addition, the driving sound diagnosis system 10, 10A substitutes the characteristic quantity obtained by time-frequency analysis of the sound vibration data into a conditional expression that does not depend on the driving mode, thereby diagnosing the cause of the driving sound. Therefore, it is not necessary to prepare normal or abnormal operation data in advance. Therefore, even if the structure of the driving instrument or machine is changed, the diagnosis of the driving sound can be universally and immediately implemented.
[0131] The case of the center offset of two shafts connected by the coupling 124 is used for explanation. It is known that the sound is a characteristic sound produced at a frequency twice the frequency of the rotation speed of the coupling 124. By changing the method of using the machine, the drive mode changes from the rotation speed of the motor 110 of 10 revolutions per second to the rotation speed of the motor 110 of 20 revolutions per second. In the technology described in Patent Document 1, a threshold value of the peak frequency of the sound when detecting an abnormality must be set for each drive mode. An example of the threshold value of the peak frequency is represented by the following formula (1) before the drive mode is changed, and is represented by the following formula (2) after the drive mode is changed.
[0132] Peak frequency before change = 10 × transformation ratio of gear box 122 × 2 ± error [Hz]... (1)
[0133] Peak frequency after change = 20 × transformation ratio of gear box 122 × 2 ± error [Hz]... (2)
[0134] On the other hand, in the present embodiment, the peak frequency which is a characteristic amount of the sound generated by the displacement of the coupling 124 is expressed by the following equation (3).
[0135] Peak frequency = frequency of rotation speed × transformation ratio of gear box 122 × 2 ± error [Hz]... (3)
[0136] That is, the peak frequency of the sound generated by the offset of the coupling 124 including equations (1) and (2) is expressed by a conditional expression with the rotation speed of the motor 110 as a variable. In this case, for each driving mode, it is not necessary to prepare a conditional expression in advance for each rotation speed of the motor 110. Moreover, the frequency of the rotation speed can be obtained based on the obtained operating data, so by using equation (3), the offset of the coupling 124 can be determined regardless of the operating mode. That is, the characteristic quantity of the sound, that is, the frequency, is substituted into a conditional expression that is independent of the driving mode for determination. In addition, here, the sound generated in the case of the center offset of the two axes connected by the coupling 124 is described, but for other factors, the characteristic quantity obtained by performing time-frequency analysis on the vibration data is substituted into a conditional expression that is independent of the driving mode, thereby making it possible to diagnose the cause of the driving sound.
[0137] Then, the cause of the problem is diagnosed based on the driving sound generated by the driven machine 120. By using the driving sound for diagnosis, it is possible to perform diagnosis without selecting a machine even when the machine rigidity is low.
[0138] The driving sound diagnosis system 10, 10A uses the results of time-frequency analysis of the sound and vibration data to diagnose the cause of the sound or vibration. By performing time-frequency analysis, it is possible to reduce errors in the diagnosis of the cause due to sensor misdetection or noise.
[0139] Furthermore, the driving sound diagnosis system 10, 10A diagnoses the cause of the driving sound based on the characteristic amount obtained by time-frequency analysis of the sound vibration data. This can reduce the data to be diagnosed and the processing time. In addition, the cause of the driving sound can be estimated by using the frequency.
[0140] In addition, the driving sound diagnosis system 10, 10A estimates the operating state of the instrument based on the operating data by the operating mode extraction unit 14, and extracts the operating mode. Moreover, only the sound vibration data corresponding to the period of constant speed operation is frequency-converted, and the sound vibration data in acceleration and deceleration, in which the drive of the actuator is difficult to stabilize, can be discarded. Therefore, compared with the case where the sound vibration data in acceleration and deceleration is still used, the accuracy of diagnosis is improved, and the sound vibration data that is not suitable for diagnosis is discarded, thereby reducing the amount of memory and processing time used in the calculation process.
[0141] Furthermore, the drive sound diagnosis system 10, 10A determines the period when the driven machine 120 is not in motion based on the operation data, and omits the sound vibration data during the period when the driven machine 120 is not in motion, thereby preventing false detection caused by sounds not based on driving. Furthermore, by omitting the frequency conversion when the driven machine 120 is not in motion, the processing cost can be reduced.
[0142] Furthermore, when investigating the cause of the driving sound, the driving sound that is to be investigated can be excluded by identifying the operation mode including the driving sound that is to be investigated. In particular, when a plurality of actuators are provided, the actuator that is the cause of the driving sound can be estimated.
[0143] In addition, the driving sound diagnosis system 10, 10A is defined as being in constant speed operation when the sum of the changes in the operating data, i.e., the speed data, converges to a specific threshold value within a certain period of time in the operating mode extraction unit 14. As a result, when the speed of the actuator converges to a certain width, the driving sound generated by the driven machine 120 is stable and easily becomes a homogeneous state, so the cause of the sound can be determined more accurately. In addition, the sound of a stable state can be extracted without determining the speed, so the same cause diagnosis can be performed when the driving mode is different and it is in constant speed operation. In addition, in order to estimate the cause, it is desired to extract the sound of the actuator or the driven machine 120 when it is stable at multiple speeds.
[0144] In addition, the driving sound diagnosis system 10, 10A includes the operation vibration extraction unit 17 that extracts the vibration component from the operation data, so that the cause determination unit 18 can compare the vibration component with the occurrence time of the characteristic point. Thus, it can be determined that the mechanical resonance is periodically vibrated by the driving vibration component, and the resonance is excited.
[0145] Furthermore, driving sound diagnosis system 10 , 10A detects characteristic points and the degree to which the position of motor 110 at the time of the characteristic points is concentrated at specific positions. This makes it possible to determine whether the detected characteristic points are characteristic points caused by factors dependent on the operating data.
[0146] In addition, the drive sound diagnosis system 10, 10A extracts the feature point as a point at which the waveform of the power of the time series spectrum with respect to the frequency and time becomes a vertex. By reducing the number of diagnoses by extraction, the processing by the factor determination unit 18 can be reduced. By calculating the correlation coefficient between the vertex of the spectrum and the operating data, it is possible to determine whether the detected vertex is caused by a factor dependent on the operating data.
[0147] Furthermore, the driving sound diagnosis system 10, 10A determines the cause of the driving sound generated by the driven machine 120 and notifies the user of the elevator via the display 133. As described above, the user of the elevator can detect that there is a problem with the driving sound of the driven machine 120 and take appropriate measures based on the diagnosis result.
[0148] Implementation method 2.
[0149] Fig.13 1 is a diagram showing an example of a hardware configuration when the drive sound diagnosis system according to Embodiment 2 is applied to a pickup unit. In this example, the diagnosis target 200 is a pickup unit that transfers a workpiece 291 flowing on a belt conveyor 225 to another belt conveyor 226 .
[0150] like Fig.13As shown, the diagnostic object 200 includes a plurality of motors 211, 212, 213, 214 and an actuator 215, a driven machine 220, and a drive device 230. A drive sound diagnostic system 10B is provided in the diagnostic object 200. The drive sound diagnostic system 10B includes a drive sound detection unit 11, an operation state detection unit 12, a wireless network device 240, a server device 250, a user terminal 260, and a network device 270. The server device 250 and the user terminal 260 are connected via a communication line 280. In the driven machine 220, the up-down direction is defined as the Z direction, the extending direction of the belt conveyors 225 and 226 in a plane perpendicular to the Z direction is defined as the Y direction, and the direction perpendicular to the Y direction and the Z direction is defined as the X direction.
[0151] The driven machine 220 is a pickup unit and includes a linear guide 221 , a ball screw 222 , a linear guide 223 , a head 224 , and two belt conveyors 225 and 226 .
[0152] The linear guide 221 is a guide that fixes the driving direction of the linear motor, that is, the motor 212. In this example, the linear guide 221 is arranged on the upper part of two belt conveyors 225 and 226 arranged in parallel, extending in the X direction. The head 224 is connected to the linear guide 221 via the motor 212, the ball screw 222 and the linear guide 223. When the motor 212 is driven, the movable direction of the motor 212 is limited to the extension direction of the linear guide 221. Fig.13 In the example, the movable direction of the motor 212 is limited to the X direction.
[0153] The ball screw 222 is connected to the motor 211, and the head 224 is driven in the Z direction by the rotation of the motor 211. The linear guide 223 is a guide that limits the driving direction of the ball screw 222 to the Z direction. The linear guide 223 is fastened to the motor 211 and the ball screw 222, and is fixed to the movable part of the motor 212. Thus, the ball screw 222 is driven along the linear guide 221 by the drive of the motor 212.
[0154] The head 224 is a worktable driven in the Z direction by the ball screw 222. The head 224 has a shape extending in the Y direction and has an actuator 215 having a mechanism for holding the workpiece 291 at the bottom. The actuator 215 is a vacuum suction cup that holds the workpiece 291 by a vacuum adsorption mechanism. The workpiece 291 held by the actuator 215 can be moved up and down by the ball screw 222. The workpiece 291 is a pickup object of the pickup unit.
[0155] The belt conveyor 225 is a feeder that feeds the workpiece 291 from the positive side to the negative side in the Y direction. The belt conveyor 225 is connected to the motor 213. The built-in conveyor belt is driven by the rotation of the motor 213, thereby conveying the workpiece 291 on the conveyor belt.
[0156] The belt conveyor 226 is an unloader that conveys the workpiece 291 from the positive side to the negative side in the Y direction. The belt conveyor 226 is connected to the motor 214. The built-in conveyor belt is driven by the rotation of the motor 214, thereby conveying the workpiece 291 on the conveyor belt.
[0157] The motor 211 is connected to the ball screw 222. The motor 211 is a servo motor that receives a current controlled by the drive device 230 and rotates a shaft.
[0158] The motor 212 is connected to the linear guide 221. The motor 212 is a linear servo motor that receives a current controlled by the drive device 230 and drives the motor 212 in the X direction.
[0159] The motor 213 is connected to the belt conveyor 225, and the motor 214 is connected to the belt conveyor 226. The motors 213 and 214 are stepping motors that receive pulse-shaped electrical signals from the driving device 230 and rotate their shafts.
[0160] The actuator 215 is a plurality of vacuum chucks that receive an electrical signal from the driving device 230 and perform suction or removal of the workpiece 291. The workpiece 291 that is picked up is held or released by the suction or removal action.
[0161] The drive device 230 includes a plurality of motor drivers 231 , 232 , and 233 and a motor control device 234 .
[0162] The motor driver 231 is connected to the motor 211 via a cable, and supplies driving power to the motor 211 while referring to the rotation angle of the motor 211. The motor driver 232 is connected to the motor 212 via a cable, and supplies driving power to the motor 212 while referring to the position of the motor 212. The motor driver 233 is connected to the motors 213 and 214 via a cable, and sends pulse-shaped commands to the motors 213 and 214.
[0163] The motor control device 234 manages the motor drivers 231, 232, and 233, and controls the driving of each motor 211, 212, 213, and 214. The motor drivers 231, 232, and 233 and the motor control device 234 are connected by cables, and can exchange information with each other through communication. In addition, the motor control device 234 is connected to the actuator 215 through a cable (not shown), and controls the adsorption and removal of the workpiece 291 by the actuator 215 through electrical signals. Since the adsorption or removal of the workpiece 291 is performed, the vacuum pump is an example of the actuator 215. When the vacuum pump is activated, the adsorption of the workpiece 291 is performed, and when the vacuum pump is not activated, the removal of the workpiece 291 is performed.
[0164] The driving sound detection unit 11 is a microphone that is arranged adjacent to the driven machine 220 and detects the sound generated by the diagnostic object 200. The driving sound detection unit 11 is provided with a communication unit 241A. An example of the communication unit 241A is a wireless communication device. The driving sound detection unit 11 groups the detected driving sound with a time stamp of the time when the sound was detected, and transmits the detected driving sound to the server device 250 via the communication unit 241A.
[0165] The operating state detection unit 12 is connected to the drive device 230 via a cable, and is a recorder that obtains information from the drive device 230 as the operating state. The operating state detection unit 12 communicates appropriately with the drive device 230, obtains various data including the operating state, and records them. The operating state detection unit 12 obtains the speed instructions of the motors 211, 212, 213, and 214 and the adsorption state of the actuator 215 as the operating state of the diagnostic object 200, and outputs them to the time synchronization unit 13. Binary data of adsorption or detachment is an example of the adsorption state.
[0166] The operating state detection unit 12 is provided with a communication unit 241B. An example of the communication unit 241B is a wireless communication device. The operating state detection unit 12 combines the acquired operating state with a time stamp of the time when the operating state was acquired, and transmits the combination to the server device 250 via the communication unit 241B.
[0167] exist Fig.13 , a case where one drive sound detection unit 11 and one operating state detection unit 12 are provided for the motors 211, 212, 213, 214 and the actuator 215 of multiple axes is shown, but a structure in which multiple drive sound detection units 11 or multiple operating state detection units 12 are provided may also be provided. In particular, in the case of a driven machine 220 having a structure that excludes interference with drive sounds of other motors by shielding or the like, by providing the drive sound detection unit 11 for each of the motors 211, 212, 213, 214 and the actuator 215, more accurate diagnosis can be expected.
[0168] The wireless network device 240 is a device that performs wireless communication with the communication units 241A and 241B. The wireless network device 240 has a communication unit 241C. A wireless LAN (Local Area Network) is constructed by the wireless network device 240 and the communication units 241A, 241B, and 241C, and the wireless network device 240 becomes an access point for the communication units 241A, 241B, and 241C. In addition, the wireless network device 240 also has the function of a router, and relays the communication between each terminal and the network of the communication line 280. In addition, the communication units 241A, 241B, 241C and the wireless network device 240 synchronize the time between the wireless devices including the communication units 241A, 241B, 241C and the wireless network device 240 as needed. The purpose is to set the time used in the driving sound detection unit 11 and the operating state detection unit 12 to an accurate time.
[0169] In the second embodiment, the functional processing unit for performing factor identification using the sound vibration data and the operation data is distributedly provided in the server device 250 and the user terminal 260. The server device 250 and the user terminal 260 are connected via a communication line 280.
[0170] If the server device 250 is instructed from the user terminal 260 to perform factor identification processing related to the driven machine 220, it uses the time series data of the driving sound, i.e., the sound vibration data, and the time series data of the operating status, i.e., the operating data, obtained from the driven machine 220 to generate device data containing feature point data.
[0171] Fig.14 2 is a block diagram schematically showing an example of the functional structure of the server device involved in Embodiment 2. The server device 250 is an information processing device provided on the network of the communication line 280. The server device 250 has the functions of the time synchronization unit 13, the operation mode extraction unit 14, the sound vibration time series spectrum acquisition unit 15, and the feature point extraction unit 16. That is, the server device 250 has the time synchronization unit 13, the operation mode extraction unit 14, the sound vibration time series spectrum acquisition unit 15, and the feature point extraction unit 16 as applications on the server device 250. In addition, in Embodiment 2, a structure in which the operation vibration extraction unit 17 is omitted is exemplified.
[0172] The server device 250 also includes a communication unit 251 that communicates with the driving sound detection unit 11 and the operating state detection unit 12 , and with the user terminal 260 .
[0173] Furthermore, the server device 250 has a device data storage unit 252. The device data storage unit 252 is a database such as RDBMS (Relational Database Management System) or Not only SQL. The server device 250 collects device data including the drive sound and the operating state sent to the network from the drive sound detection unit 11 and the operating state detection unit 12 into a database, and stores it after processing as needed. As an example, as described in Implementation Example 1, the synchronization of the sound vibration data and the operating data is obtained, the spectrum data of the time series is obtained for the sound vibration data, the feature points are extracted from the spectrum data, and the device data including the feature point data is stored in the device data storage unit 252, and the device data includes the frequency, time, power, waveform of the feature point and the operating data of the time of the feature point are grouped.
[0174] Fig.15 1 is a diagram showing an example of a record of device data according to Embodiment 2. One record of device data includes a registration time, an operation mode, feature point data, and operation data. In addition, by analyzing the data collected by the server device 250 through deep learning or the like, it is possible to discriminate factors with higher accuracy.
[0175] The user terminal 260 instructs the server device 250 to determine the cause of the driving sound of the driven machine 220, performs a process of determining the cause using the device data received from the server device 250, and displays the result. The user terminal 260 is an information processing device owned by the user of the driven machine 220. An example of the user terminal 260 is a notebook or desktop personal computer.
[0176] Fig.16 2 is a block diagram schematically showing an example of a functional configuration of a user terminal according to Embodiment 2. The user terminal 260 has the function of the factor determination unit 18 described in Embodiment 1. That is, the user terminal 260 has the factor determination unit 18 as an application on the user terminal 260 .
[0177] In addition, the user terminal 260 includes a communication unit 261, an input unit 262, and a display unit 263. The communication unit 261 communicates with the server device 250 via the communication line 280. As an example, an instruction to execute the process of factor identification input by the user from the input unit 262 is sent to the server device 250, and various information including device data is received from the server device 250. In addition, the communication unit 261 is connected to the drive sound detection unit 11 and the operation state detection unit 12, and can also obtain sound vibration data with the drive sound as time series data and operation data with the operation state as time series data.
[0178] The input unit 262 is an input interface with the user. A keyboard or a mouse is an example of the input unit 262. The input unit 262 is used to input instructions for executing a process for determining a cause of the driven machine 220.
[0179] The display unit 263 displays information necessary for executing the process of factor identification of the driven machine 220. A liquid crystal display is an example of the display unit 263. The display unit 263 displays the result of factor identification or device data acquired via a network.
[0180] The network device 270 is a communication device that relays communication among the drive sound detection unit 11 and the operation state detection unit 12 provided in the driven machine 220 , the server device 250 , and the user terminal 260 . A router is an example of the network device 270 .
[0181] like Fig.13 As shown, by configuring the drive sound diagnostic system 10B via a network, it is possible to perform drive sound diagnosis even in a remote location different from the factory where the driven machine 220 is installed.
[0182] The server device 250 and the user terminal 260 that perform the driving sound diagnosis are realized by the information processing device as described above. Fig.17 4 is a block diagram showing an example of the hardware configuration of a server device and a user terminal. The server device 250 and the user terminal 260 include a computing device 401, a memory 402, a storage device 403, a communication device 404, an input device 405, and a display device 406. The computing device 401, the memory 402, the storage device 403, the communication device 404, the input device 405, and the display device 406 are connected via a bus 407.
[0183] The computing device 401 is a processor represented by a CPU that performs computing processing. The memory 402 functions as a work area for storing data used by the computing device 401 during computing processing. The storage device 403 stores computer programs, information, etc. The communication device 404 has a communication function with other devices connected to the network. The input device 405 receives input from the operator. The input device 405 is a keyboard, a mouse, etc. The display device 406 outputs a display screen. The display device 406 is a monitor, a display, etc. In addition, a touch panel in which the input device 405 and the display device 406 are integrated can also be used.
[0184] Fig.14The functions of the time synchronization unit 13 , the operation mode extraction unit 14 , the sound vibration time series spectrum acquisition unit 15 , and the feature point extraction unit 16 are realized by the calculation device 401 reading and executing a computer program stored in the storage device 403 .
[0185] in addition, Fig.16 The functions of the factor determination unit 18 shown are realized by the arithmetic device 401 reading and executing a computer program stored in the storage device 403 .
[0186] Next, the operation of the diagnostic object 200 will be described. In order to transport the workpiece 291 by the pickup unit, the user of the pickup unit inputs the drive instructions of the motors 211, 212, 213, 214 and the actuator 215 to the motor control device 234. As an example, the user inputs the instructions to the motor control device 234 from the user terminal 260. Then, the instructions are sent to the motor control device 234 via the network device 270, the communication line 280, the server device 250, the wireless network device 240, the communication unit 241C, and the communication unit 241B. The motor control device 234 generates and sends the drive instructions to the motor drivers 231, 232, and 233 according to the input instructions.
[0187] The motor drivers 231 , 232 , and 233 control the motor driving current according to the received driving command to drive the motors 211 , 212 , 213 , and 214 and the actuator 215 .
[0188] When the motor 211 is driven, the ball screw 222 connected to the motor 211 rotates, and the head 224 and the actuator 215 connected to the ball screw 222 move in the Z direction.
[0189] When the motor 212 is driven, the linear guide 223 connected to the motor 212 , the motor 211 , and the instrument connected to the motor 211 move in the X direction.
[0190] The motor 213 is driven to rotate the belt conveyor 225, and the workpiece 291 on the belt conveyor 225 moves from the positive side to the negative side in the Y direction. Similarly, the motor 214 is driven to rotate the belt conveyor 226, and the workpiece 291 on the belt conveyor 226 moves from the positive side to the negative side in the Y direction.
[0191] The actuator 215 adsorbs the workpiece 291 moving on the belt conveyor 225 at a timing of contacting the actuator 215 directly below, thereby holding the workpiece 291. Thus, when the actuator 215 moves upward by driving the motor 211 after adsorption, the state in which the workpiece 291 and the actuator 215 are in contact can be maintained.
[0192] Then, the motor 211 rotates to move the actuator 215 upward. At this time, the workpiece 291 is sucked by the actuator 215, so as the actuator 215 rises, the workpiece 291 also rises and becomes separated from the belt conveyor 225.
[0193] Next, the motor 212 is driven toward the positive side in the Y direction, thereby moving the workpiece 291 directly above the belt conveyor 226. Then, the motor 211 is rotated again, thereby placing the workpiece 291 on the belt conveyor 226. The actuator 215 releases the workpiece 291 at the timing when the held workpiece 291 contacts the belt conveyor 226, thereby placing the workpiece 291 on the belt conveyor 226. After the actuator 215 is released, the motors 211 and 212 are driven to return the actuator 215 to the starting position. Finally, the motor 214 is rotated, thereby carrying the workpiece 291 toward the negative side in the Y direction of the belt conveyor 226.
[0194] During the above-described operation, the diagnostic object 200 generates a driving sound when it is driven. The driving sound generated by the diagnostic object 200 and the rotation angles of the motors 211, 212, 213, 214 are recorded and acquired by the driving sound detection unit 11 and the operation state detection unit 12, respectively, and transmitted to the server device 250 via wireless communication.
[0195] The server device 250 communicates with the drive sound detection unit 11 and the operating state detection unit 12 at any time through the wireless network device 240, obtains various information related to the diagnostic object 200 including the drive sound and the operating state, and stores the information in the device data storage unit 252. As a pre-processing for storage in the device data storage unit 252, the time synchronization unit 13, the operating mode extraction unit 14, the sound vibration time series spectrum acquisition unit 15, and the feature point extraction unit 16 perform processing on the sound vibration data and operating data of the acquired time series data, and calculate the operating mode and feature point data. The device data storage unit 252 stores the feature point data calculated by the time synchronization unit 13, the operating mode extraction unit 14, the sound vibration time series spectrum acquisition unit 15, and the feature point extraction unit 16, the operating data synchronized with the time of the feature point, and other necessary information related to the diagnostic object 200.
[0196] In response to the user's request, the user terminal 260 obtains various system information required by the user, such as the presence or absence of system abnormalities, through the server device 250, and displays the information on the display unit 263 of the user terminal 260. In addition, the user can use the database stored in the device data storage unit 252 of the server device 250 to perform analysis required for the maintenance and operation of the diagnosis object 200 and the driving sound diagnosis system 10B.
[0197] Next, a driving sound diagnosis procedure of driving sound diagnosis system 10B in Embodiment 2 will be described. The basic driving sound diagnosis procedure is substantially the same as that in Embodiment 1, and the following description will focus on the differences from Embodiment 1.
[0198] In Embodiment 2, the time synchronization unit 13 of the server device 250 synchronizes the sound vibration data and the operation data using the time stamps grouped in the driving sound detection unit 11 and the operation state detection unit 12. Thus, even when the real-time communication cannot be ensured due to poor communication quality, etc., the sound vibration data and the operation data can be synchronized.
[0199] In addition, when data at a specific time is required in the operation mode extraction unit 14 of the server device 250, data having a time stamp at a time specified by the time stamp is extracted. By extracting in this way, even if the acquisition cycles of the sound vibration data and the operation data are different, they can be analyzed as synchronized data.
[0200] As another method in the case where the detection periods of the sound vibration data and the operation data are different, after filtering the data to have the same period, a method of spacing the data apart or conversely interpolating the data during the period by a method such as linear interpolation may be considered.
[0201] In addition, the operation mode extraction unit 14 divides the process and sets the operation mode according to which motor among the multiple motors 211, 212, 213, 214 and the actuator 215 is mainly operated. Specifically, the operation mode extraction unit 14 determines which motor 211, 212, 213, 214 or actuator 215 is operated based on the operation data, according to the speed instructions of the motors 211, 212, 213, 214 and the data of the crimping state of the actuator 215, and determines the operation mode. In one example, the motor 213 mainly operates when the workpiece is moved in, the motors 211 and 212 mainly operate when the workpiece is lifted, the motor 214 mainly operates when the workpiece is moved out, and the actuator 215 mainly operates when the pump is operated. Therefore, in one example, the operation mode extraction unit 14 divides the operation mode when the workpiece is moved in, when the workpiece is lifted, when the workpiece is moved out, and when the pump is operated.
[0202] The factor determination unit 18 of the user terminal 260 compares the feature point data with the numerical range generated by the factor determination condition, i.e., the server device 250, to determine the factor of the driving sound. As a method for generating the numerical range in the server device 250, the numerical range may be generated based on a parameter estimated as a factor based on the knowledge, or may be generated by machine learning based on operation data and sound vibration data. In addition, the server device 250 may dynamically generate the factor determination condition before determining it by the factor determination unit 18. In this case, the factor diagnosis can be performed based on a more urgent case, so that a more accurate diagnosis can be performed.
[0203] In addition, in Embodiment 2, if Figure 1 As shown, the configuration is such that the operation vibration extracting unit 17 is not provided. Therefore, the factor determining unit 18 determines the factor using feature point data instead of vibration data.
[0204] The drive sound diagnosis system 10B according to the second embodiment compares the characteristic point data obtained by performing time-frequency analysis on the sound vibration data and the operation data with the numerical range generated by the server device 250 for each operation mode, thereby diagnosing the cause of the drive sound. Therefore, it is not necessary to prepare normal or abnormal operation data in advance, and even when the structure of the drive instrument or machine is changed, the drive sound diagnosis can be performed universally and immediately.
[0205] In addition, according to the drive sound diagnosis system 10B involved in the second embodiment, it is possible to immediately determine the cause of the drive sound generated by the driven machine 220 driven by the plurality of motors 211, 212, 213, 214 and the actuator 215. In particular, it becomes easy to determine which motor drives the part where the cause exists. Thus, when there is a problem in the drive sound of the driven machine 220, the user can take appropriate measures according to the cause, and the time required for the treatment can be shortened.
[0206] Implementation method 3.
[0207] Fig.18 1 is a block diagram showing an example of the functional structure of the cause determination unit in the driving sound diagnosis system according to Embodiment 3. The driving sound diagnosis system according to Embodiment 3 replaces the cause determination unit 18 of the operation processing unit 140 in Embodiment 1 with a cause determination unit 18A, which estimates the cause of the driving sound using a learner that has learned the cause determination of the driving sound. In addition, the same reference numerals are attached to the same components as those in Embodiment 1 and their description is omitted, and the parts different from Embodiment 1 are described below.
[0208] The factor determination unit 18A includes a learning result storage unit 31 and a factor estimation unit 32. The learning result storage unit 31 stores the learning result of machine learning performed in advance, which estimates the factor of the driving sound corresponding to the feature point data extracted by the feature point extraction unit 16. The factor estimation unit 32 performs a calculation process based on the learning result of the learning result storage unit 31.
[0209] The factor determination unit 18A receives the feature point data extracted by the feature point extraction unit 16 as input, and performs calculation processing based on the learning result of the learning result storage unit 31 by the factor estimation unit 32 to identify the factor of the driving sound.
[0210] At this time, the learning result stored in the learning result storage unit 31 may be a result obtained by machine learning using all feature point data or a result obtained by machine learning using part of the feature point data. In addition, the factor inference unit 32 extracts input data according to the feature point data used in the learning of the learning result storage unit 31 and performs calculation processing.
[0211] In addition, as a learning model of machine learning that introduces the learning results stored in the learning result storage unit 31, K-nearest neighbor method, decision tree, support vector machine, and kernel approximation are mentioned. In addition, deep learning can be used as a learning model.
[0212] The factor determination unit 18A in the third embodiment uses a learner that has completed learning of the factor determination of the driving sound to make a determination. Thus, it is possible to provide a more accurate factor determination of the driving sound. In addition, the factor determination unit 18A uses a learner that has performed machine learning using the same data as the factor determination unit 18, so it is possible to make a determination that is not dependent on the driving mode. Thus, the learning results stored in the learning result storage unit 31 can be applied to various driving devices.
[0213] Implementation method 4.
[0214] Fig.19 1 is a block diagram showing an example of the functional structure of the machine learning device of the drive sound diagnosis system involved in Embodiment 4. The machine learning device 50 enables the server device 250 and the user terminal 260 having the function processing unit for performing cause discrimination using sound vibration data and operation data in Embodiment 2 to have the function of performing machine learning. In addition, the same reference numerals are attached to the same structural elements as those in Embodiments 1 and 2, and their description is omitted, and the parts different from Embodiments 1 and 2 are described.
[0215] The machine learning device 50 includes a driving sound detection unit 11 , an operation state detection unit 12 , a sound vibration time series spectrum acquisition unit 15 , a feature point extraction unit 16 , a factor acquisition unit 51 , a factor learning unit 52 , and a learning result storage unit 53 .
[0216] The factor acquisition unit 51 acquires data related to the factors causing the driving sound. For example, the factor acquisition unit 51 may acquire data related to the factors causing the driving sound that are measured and input by the operation of the designer or the user. Alternatively, the factor acquisition unit 51 may acquire the diagnosis results of the factors causing the driving sound involved in Embodiment 1 or Embodiment 2, or may acquire the diagnosis results of the factors causing the driving sound in other driving sound diagnosis systems.
[0217] The factor learning unit 52 learns the generation factor of the drive sound according to the training data set created based on the combination of the feature point data extracted by the feature point extraction unit 16 and the data related to the generation factor of the drive sound obtained by the factor acquisition unit 51. The learning model used by the factor learning unit 52 can use the learning model mentioned in the embodiment 3. The diagnostic data used in the training data set can use the data of multiple driven machines 220. In one example, by connecting to multiple driven machines 220 through a communication line 280 or the like and collecting diagnostic data, a more accurate diagnosis can be made. The learning result storage unit 53 stores the learning results obtained by the factor learning unit 52.
[0218] Fig. 20 2 is a block diagram schematically showing an example of the functional structure of the server device involved in Embodiment 4. The server device 250A is an information processing device provided on the network of the communication line 280. The server device 250A includes a time synchronization unit 13, an operation mode extraction unit 14, a sound vibration time series spectrum acquisition unit 15, a feature point extraction unit 16, a factor learning unit 52, a learning result storage unit 53, and a communication unit 251.
[0219] The server device 250A processes the device data including the driving sound and the operating state sent from the driving sound detection unit 11 and the operating state detection unit 12 to the communication line 280 as needed, and learns and stores them. As an example, first, as in Embodiment 2, all or part of the feature point data, which is a group of the frequency, time, power, waveform of the feature point, and the operating data at the time of the feature point, are input to the factor learning unit 52. Next, the generation factors of the driving sound are obtained from the user terminal 260 via the communication line 280 and the communication unit 251, and are combined to form a training data set. Finally, the generation factors of the driving sound are learned by the factor learning unit 52 using the training data set, and the results are stored in the learning result storage unit 53.
[0220] Fig.21 This is a block diagram schematically showing an example of the functional configuration of a user terminal according to Embodiment 4. The user terminal 260A includes a communication unit 261 , an input unit 262A, a display unit 263 , a factor determination unit 18 , and a factor acquisition unit 51 .
[0221] The factor acquisition unit 51 acquires data related to the generation factor of the drive sound from the user, for example. The acquired data related to the generation factor of the drive sound is transmitted to the server device 250A via the communication unit 261 .
[0222] The input unit 262A is an input interface with the user. The input unit 262A serves as an interface for the user to input an instruction to execute a process for determining a factor of the driven machine 220 and for the user to input data related to the factor of the drive sound acquired by the factor acquisition unit 51 .
[0223] The drive sound diagnosis system according to Embodiment 4 has both a machine learning function for the cause of the drive sound and a cause identification function using the result of the machine learning function. Thus, the cause diagnosis of the drive sound can be utilized and the accuracy of the cause diagnosis can be continuously improved through learning.
[0224] Furthermore, by connecting to a plurality of driven machines 220 or user terminals 260A via the communication line 280, more accurate factor diagnosis can be used in a plurality of locations regardless of the installation location of the driven machines 220. As a result, diagnosis with high versatility can be learned.
[0225] The driving sound diagnosis system according to the fourth embodiment has both the machine learning function of the factors of the driving sound and the factor determination function using the result of the machine learning function. However, the machine learning device 50 of the driving sound diagnosis system may be configured separately, and the factor determination using the learning result may be a function of another device. An example of the factor determination device using the learning result is the device of the third embodiment. Fig.18 The structure shown.
[0226] The configuration described in the above embodiment is an example of the content of the present invention, and can be combined with other known technologies. Part of the configuration can be omitted or changed without departing from the scope of the present invention.
[0227] Description of the label
[0228] 10, 10A, 10B driving sound diagnosis system, 11 driving sound detection unit, 12 operating state detection unit, 13 time synchronization unit, 14 operating mode extraction unit, 15 sound vibration time series spectrum acquisition unit, 16 feature point extraction unit, 17 operating vibration extraction unit, 18, 18A factor determination unit, 31, 53 learning result storage unit, 32 factor inference unit, 50 machine learning device, 51 factor acquisition unit, 52 factor learning unit, 100, 200 diagnosis object, 110, 21 1, 212, 213, 214 electric motors, 120, 220 driven machines, 130, 230 driving devices, 131, 231, 232, 233 motor drivers, 132, 234 motor control instruments, 133 display, 140 calculation processing unit, 215 actuator, 250, 250A server device, 251, 261 communication unit, 252 device data storage unit, 260, 260A user terminal, 270 network equipment, 280 communication line.
Claims
1. A driving sound diagnostic system, characterized in that: have: a driving sound detection unit for detecting driving sound, which is sound or mechanical vibration generated by the actuator or a driven machine driven by the actuator; an operation state detection unit that obtains a driving position, a driving speed, or a force generated by driving of the actuator in a time series; a sound vibration time series spectrum acquisition unit, which calculates the spectrum corresponding to each time of the detected time series data of the driving sound, that is, the sound vibration data, and outputs the time series spectrum obtained by grouping the power of the calculated spectrum with the frequency and the time; a feature point extraction unit, which extracts as a feature point a point at which the power of the time series spectrum is a waveform of at least one of a top, a peak, an edge, and a saddle point relative to the frequency and the waveform at the time, and outputs feature point data obtained by grouping the frequency of the feature point, the time, the waveform of the feature point, and operation data, wherein the operation data is a driving position, a driving speed, or a force generated by driving of the actuator at the time of the feature point; as well as A cause determination unit determines the cause of the detected drive sound by comparing a cause determination condition with the numerical value of the feature point data, wherein the cause determination condition determines a first numerical range when a combination of the frequency and at least one of the moments included in the feature points accompanying the phenomenon and the operation data is set as multidimensional data for each phenomenon occurring in the actuator or the driven machine, and the operation data is a drive position, a drive speed or a force generated by the drive of the actuator at the moment.
2. The driving sound diagnostic system according to claim 1, characterized in that: The apparatus further comprises an operation mode extraction unit that divides the operation state of the driven machine into two or more intervals separated by time based on the operation data. The sound vibration time-series spectrum acquisition unit calculates the spectrum using the sound vibration data of at least one point detected between a start time and an end time of the interval.
3. The driving sound diagnostic system according to claim 2, characterized in that: The operation pattern extraction unit extracts a section in which the speed of the actuator converges to a certain width for a time period longer than a predetermined time period.
4. The driving sound diagnostic system according to any one of claims 1 to 3, characterized in that: It also includes an operation vibration extraction unit for extracting vibration components from the operation data, The cause determination unit determines the cause of the detected driving sound by comparing the vibration component with a second numerical range based on a comparison of the numerical value of the feature point data with the first numerical range, and the second numerical range includes the vibration component generated along with the phenomenon for each of the phenomena.
5. The driving sound diagnostic system according to any one of claims 1 to 3, characterized in that: The factor determination unit includes: a learning result storage unit that stores a learning result of machine learning for estimating factors of the feature point data and the sound for the operation data; as well as A factor estimating unit receives the multi-dimensional data as an input, executes a calculation process based on the learning result stored in the learning result storage unit, and determines a factor of generation of the drive sound.
6. The driving sound diagnostic system according to claim 4, characterized in that: The factor determination unit includes: a learning result storage unit that stores a learning result of machine learning for estimating factors of the feature point data and the sound for the operation data; as well as A factor estimating unit receives the multi-dimensional data as an input, executes a calculation process based on the learning result stored in the learning result storage unit, and determines a factor of generation of the drive sound. 7 . A storage medium storing a program for causing a computer to execute the functions of the driving sound diagnosis system according to claim 1 .
8. A driving sound diagnosis method, characterized in that: The process includes the following steps: detecting a driving sound, which is a sound or mechanical vibration generated by an actuator or a driven machine driven by the actuator; Obtaining a driving position, a driving speed, or a force generated by driving of the actuator in a time series; Calculating the frequency spectrum corresponding to each moment of the detected time series data of the driving sound, i.e., the sound vibration data, and outputting the time series frequency spectrum obtained by grouping the power of the calculated frequency spectrum with the frequency and the moment; Extracting at least any one of the apex, peak, ridge and saddle point of the waveform of the power of the time series spectrum relative to the frequency and the waveform at the time as a feature point, and outputting feature point data obtained by grouping the frequency of the feature point, the time, the waveform of the feature point and operation data, wherein the operation data is the driving position, driving speed or force generated by driving of the actuator at the time of the feature point; as well as The cause of the detected driving sound is determined by comparing the cause determination condition with the numerical value of the feature point data. The cause determination condition determines the numerical range when the combination of the frequency and at least one of the moments included in the feature point accompanying the phenomenon and the operating data is set as multidimensional data for each phenomenon occurring in the actuator or the driven machine. The operating data is the driving position, driving speed or force generated by the driving of the actuator at the moment. 9 . A storage medium storing a program for causing a computer to execute the steps of the drive sound diagnosis method according to claim 8 .
10. A machine learning device for a driving sound diagnosis system, characterized in that: have: a driving sound detection unit for detecting driving sound, which is sound or mechanical vibration generated by the actuator or a driven machine driven by the actuator; an operation state detection unit that obtains a driving position, a driving speed, or a force generated by driving of the actuator in a time series; a sound vibration time series spectrum acquisition unit, which calculates the spectrum corresponding to each time of the detected time series data of the driving sound, that is, the sound vibration data, and outputs the time series spectrum obtained by grouping the power of the calculated spectrum with the frequency and the time; a feature point extraction unit, which extracts as a feature point a point at which the power of the time series spectrum is a waveform of at least one of a top, a peak, an edge, and a saddle point relative to the frequency and the waveform at the time, and outputs feature point data obtained by grouping the frequency of the feature point, the time, the waveform of the feature point, and operation data, wherein the operation data is a driving position, a driving speed, or a force generated by driving of the actuator at the time of the feature point; a factor learning unit for learning the cause of the drive sound based on the multidimensional data by combining at least one of the frequency and the time included in the characteristic point accompanying the occurrence of the phenomenon and the operation data for each phenomenon occurring in the actuator or the driven machine, wherein the operation data is a driving position, a driving speed, or a force generated by driving of the actuator at the time; a learning result storage unit for storing the learning result obtained by the factor learning unit; as well as A factor acquisition unit acquires a generation factor of the drive sound given to the factor learning unit. 11 . A storage medium storing a program for causing a computer to execute the functions realized by the machine learning device of the drive sound diagnosis system according to claim 10 .
12. A storage device storing the program according to any one of claims 7, 9 and 11.
Citation Information
Patent Citations
Rotating equipment abnormality diagnostic method and device therefor
JP1998274558A
Abnormal engine sound fault on-line diagnostic system and diagnostic method
CN103558029A
Method and system for detection of faults in pump assembly via handheld communication device
CN107076155A
