Diagnostic Devices

Abstract

The present invention provides a diagnostic device, which includes: a control unit that controls a diagnostic operation of driving a transmission belt; a first tension calculation unit that calculates a tension value of a transmission belt when the transmission belt is free of wear based on data obtained from the diagnostic operation; a second tension calculation unit that calculates a tension value of a second transmission belt including a factor of a tension drop in the transmission belt and a factor of wear of the transmission belt; and a third tension calculation unit that calculates a degree of wear of the transmission belt based on the tension value of the first transmission belt and the tension value of the second transmission belt. Thus, the diagnostic device estimates the degree of wear of the transmission belt or supports abnormality diagnosis.

Description

Technical Field

[0001] The present invention relates to a diagnostic device, and in particular to a diagnostic device for diagnosing the degree of wear of a transmission belt for transmitting power. Background Art

[0002] Among industrial machines, there is a machine that transmits the rotational power of an electric motor to a shaft via a power transmission unit such as a timing belt. Japanese Patent Publication No. 2013-216030, Japanese Patent Publication No. 2000-193560, Japanese Patent Publication No. 2007-90477, Japanese Patent Publication No. 11-262932, Japanese Patent Publication No. 9-178546, and Japanese Patent Publication No. 2018-167424 propose a method for indirectly estimating the state of a transmission belt by performing a specific action for maintenance of a machine that uses a timing belt as a power transmission unit of an industrial machine.

[0003] The state of the transmission belt can be estimated based on phenomena such as delay in the movement of the shaft (relative to the movement of the motor) and difference in the amount of movement observed when performing a specific action. The phenomenon observed when performing such a specific action causes a deviation from the normal state of the transmission belt due to factors such as tension reduction and belt wear (wear, cracking, peeling, etc. of the broken surface). In this way, the deviation from the normal state of the transmission belt is related to multiple factors, but in the conventional estimation of the state of the transmission belt in a specific action, the abnormality of the transmission belt as a result of including these multiple factors is measured and estimated.

[0004] Usually, when checking the tension of the transmission belt by manual operation, the rotation of the transmission belt is stopped, and the machine is disassembled to expose the transmission belt. At this time, the wear state of the transmission belt can be determined by visually inspecting the transmission belt. However, there are also parts of the transmission belt of the machine that are hidden in the shadows of other parts and difficult to see visually. In addition, the maintenance of the transmission belt by visual inspection depends on the skills of the operator. Therefore, at a site where many machines are installed, fewer skilled operators cannot perform maintenance work on all these machines by visual inspection. Therefore, the current situation is that many machines only have the adjustment of the transmission belt tension as a maintenance item, and the wear state of the transmission belt is not visually inspected and maintained.

[0005] On the other hand, when performing a specific diagnostic operation such as frequency scanning, especially when judging the state of the timing belt based on the natural vibration frequency, as described above, the abnormality occurs as a result of the overlap of two factors caused by the decrease in tension of the timing belt and the wear of the timing belt. Therefore, it is difficult to grasp the wear of the timing belt only through the above-mentioned specific diagnostic operation. Summary of the invention

[0006] In view of the above-mentioned situation, a structure capable of diagnosing the degree of wear of the transmission belt by a relatively simple method other than visual inspection is desired.

[0007] One embodiment of the present invention is a diagnostic device for diagnosing the degree of wear of a power transmission belt provided in an industrial machine, the diagnostic device comprising: a control unit for controlling a diagnostic operation for driving the transmission belt; a first tension calculation unit for calculating a tension value of the transmission belt when the transmission belt is estimated to be free of wear based on data obtained from the diagnostic operation; a second tension calculation unit for calculating a tension value of the second transmission belt including a tension drop factor of the transmission belt and a wear factor of the transmission belt; and a third tension calculation unit for calculating the degree of wear of the transmission belt based on the tension value of the first transmission belt and the tension value of the second transmission belt. According to one embodiment of the present invention, the degree of wear of the transmission belt can be diagnosed by a relatively simple method other than visual inspection. In particular, the transmission belt used in an injection molding machine is a part that is difficult for an operator to visually inspect during maintenance, and since there are various types of metal molds, it is difficult to generate a general determination formula and a machine learning model for determining wear based only on natural vibration frequency. Therefore, the method of the present invention that can determine wear of the transmission belt by a relatively simple method other than visual inspection is effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The above and other objects and features of the present invention will be apparent from the following embodiments described with reference to the accompanying drawings.

[0009] Figure 1 This is a schematic hardware configuration diagram of a diagnostic device according to one embodiment.

[0010] Figure 2 This is a schematic block diagram showing the functions of the diagnostic device according to the first embodiment.

[0011] Figure 3 This is a schematic block diagram showing the functions of the first tension calculation unit according to the first embodiment.

[0012] Figure 4 Shows an example of frequency response data.

[0013] Figure 5 This is a schematic block diagram showing the functions of the second tension calculation unit according to the first embodiment.

[0014] Figure 6 This is a schematic block diagram showing the functions of the third tension calculation unit according to the first embodiment.

[0015] Figure 7 This is a schematic block diagram showing the functions of the second tension calculation unit according to the second embodiment.

[0016] Figure 8 This is a schematic block diagram showing the functions of the second tension calculation unit according to the third embodiment. DETAILED DESCRIPTION

[0017] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0018] Figure 1 1 is a schematic hardware structure diagram of the main parts of a diagnostic device according to an embodiment of the present invention. The diagnostic device 1 can be installed as a control device for controlling industrial machinery such as an injection molding machine based on a control program, for example. In addition, the diagnostic device 1 can be installed on a personal computer arranged in parallel with a control device for controlling industrial machinery based on a control program, a personal computer connected to the control device via a wired / wireless network, a unit computer, a fog computer 6, or a cloud server 7. In this embodiment, an example of installing the diagnostic device 1 as a control device for controlling industrial machinery based on a control program is shown.

[0019] The CPU 11 included in the diagnostic device 1 of the present embodiment is a processor that controls the entire diagnostic device 1. The CPU 11 reads a system program stored in the ROM 12 via the bus 22, and controls the entire diagnostic device 1 according to the system program. The RAM 13 temporarily stores temporary calculation data, display data, and various data input from the outside.

[0020] The nonvolatile memory 14 is composed of, for example, a battery-backed memory or SSD (Solid State Drive) not shown in the figure, and maintains the storage state even if the power of the diagnostic device 1 is turned off. The nonvolatile memory 14 stores control programs and data read from the external device 72 via the interface 15, control programs and data input via the input device 71, and control programs and data acquired from other devices via the network 5. The control programs and data stored in the nonvolatile memory 14 can also be expanded in the RAM 13 when executed / used. In addition, various system programs such as well-known analysis programs are pre-written in the ROM 12.

[0021] The interface 15 is connected to the CPU 11 of the diagnostic device 1 and an external device 72 such as a USB device. For example, a control program and setting data for controlling industrial machinery are read from the external device 72. In addition, the control program and setting data edited in the diagnostic device 1 can be stored in an external storage unit via the external device 72. The PLC (Programmable Logic Controller) 16 executes the ladder program and outputs signals to the industrial machinery and the peripheral devices of the industrial machinery (such as tool changers, actuators such as robots, sensors such as temperature sensors and humidity sensors installed on the industrial machinery) via the I / O unit 19 to control the above-mentioned industrial machinery and the above-mentioned peripheral devices. In addition, the PLC 16 receives signals from various switches and peripheral devices of the operating panel equipped on the main body of the industrial machinery via the I / O unit 19, and transmits them to the CPU 11 after necessary signal processing.

[0022] The interface 20 connects the CPU 11 of the diagnostic device 1 to a wired or wireless network 5. Other industrial machines 3, fog computers 6, cloud servers 7, etc. are connected to the network 5, and these devices and the diagnostic device 1 exchange data with each other.

[0023] The display device 70 outputs and displays various data read from the memory, data obtained as a result of executing a program, etc., data output from the machine learning device 100 described later, etc. via the interface 17. In addition, the input device 71 composed of a keyboard, a pointing device, etc. transmits instructions based on the operator's operation, data input by the operator, etc. to the CPU 11 via the interface 18.

[0024] The axis control circuit 30 for controlling the axis of the industrial machine receives the axis movement instruction amount from the CPU 11, and outputs the axis instruction to the servo amplifier 40. The servo amplifier 40 receives the instruction and drives the servo motor 50 for moving the axis of the industrial machine. The servo motor 50 has a built-in position / speed detector (not shown), and feeds back the position / speed feedback signal from the position / speed detector to the axis control circuit 30. The axis control circuit 30 performs position / speed feedback control using the position / speed feedback signal.

[0025] In addition, Figure 1 In the hardware structure diagram of FIG. 1 , only one of the axis control circuit 30, servo amplifier 40, and servo motor 50 is shown, but in reality, a number corresponding to the number of axes of the industrial machine to be controlled is prepared. In addition, at least one of the servo motors 50 is connected to a predetermined axis of the industrial machine via a transmission belt as a power transmission unit.

[0026] The interface 21 connects the CPU 11 to the machine learning device 100. The machine learning device 100 includes a processor 101 that uniformly controls the entire machine learning device 100, a ROM 102 that stores a system program, etc., a RAM 103 for temporary storage in various processes involved in machine learning, and a non-volatile memory 104 for storing learning models, etc. The machine learning device 100 can observe various information that can be obtained by the diagnostic device 1 (for example, data indicating the operating state of the servo motor 50, detection values ​​of temperature sensors and humidity sensors not shown, etc.) via the interface 21. In addition, the diagnostic device 1 obtains the processing results output from the machine learning device 100 via the interface 21. The diagnostic device 1 stores or displays the obtained processing results, and sends them to other devices via the network 5, etc.

[0027] Figure 2 Functions included in the diagnostic device 1 according to the first embodiment of the present invention are shown as a schematic block diagram. Figure 1 The CPU 11 of the diagnostic device 1 and the processor 101 of the machine learning device 100 execute system programs to control the operations of the respective components of the diagnostic device 1 and the machine learning device 100 , thereby realizing the respective functions of the diagnostic device 1 of this embodiment.

[0028] The diagnostic device 1 of this embodiment includes a control unit 110, a first tension calculation unit 120, a second tension calculation unit 130, and a third tension calculation unit 140. A control program 200 for controlling the servo motor 50 included in the industrial machine is stored in advance in the RAM 13 or the nonvolatile memory 14 of the diagnostic device 1.

[0029] Figure 1The CPU 11 included in the diagnostic device 1 shown in the figure executes the system program read from the ROM 12, and mainly performs the calculation processing using the RAM 13 and the non-volatile memory 14 by the CPU 11, the control processing of each part of the industrial machine using the axis control circuit 30 and the PLC 16, and the input and output processing via the interface 18, thereby realizing the control unit 110. The control unit 110 analyzes the program blocks of the control program 200, and controls each part of the industrial machine based on the analysis results. For example, when the program blocks of the control program 200 instruct to drive each axis of the industrial machine, the control unit 110 generates movement instruction data based on the instructions based on the program blocks and outputs it to the servo motor 50. In addition, when the program blocks of the control program 200 instruct to operate the peripheral devices such as the sensor 4 installed in the industrial machine, the control unit 110 generates a predetermined signal to operate the peripheral device and outputs it to the PLC 16. On the other hand, the control unit 110 obtains the speed feedback gain of the servo motor 50 and the detection value data detected by the sensor 4 such as the temperature sensor and the humidity sensor, and outputs it to each tension calculation unit as needed.

[0030] The control program 200 includes a program block for a diagnostic action for causing the servo motor 50 that drives the transmission belt in advance to perform a scanning action at a rotation speed (frequency) within a predetermined range. In addition, the control program 200 includes a program block for an instruction to obtain the speed feedback gain of the servo motor 50 in the scanning action as time series data, and includes a program block for an instruction to obtain the detection value data of the sensor 4 in at least any one of the start of the scanning action, during the scanning action, and at the end of the scanning action. The diagnostic action based on the control program 200 can be performed at multiple positions of the timing belt, and multiple diagnostic actions can also be performed under multiple (initial) conditions such as different mechanical postures and different diagnostic actions such as different meshing positions of the transmission belt. The loss of the transmission belt is generally generated locally, and there are also multiple states of loss. In this regard, by performing diagnosis based on multiple data obtained at different transmission belt positions and conditions, it is also possible to deal with partial transmission belt loss with high accuracy.

[0031] Figure 1The CPU 11 of the diagnostic device 1 and the processor 101 of the machine learning device 100 respectively execute the system programs read from the ROM 12 and ROM 102, and mainly perform the calculation processing using the RAM 13 and the non-volatile memory 14 by the CPU 11, the calculation processing using the RAM 103 and the non-volatile memory 104 by the processor 101, and the input and output processing via the interface 18 and the interface 21, thereby realizing the first tension calculation unit 120. The first tension calculation unit 120 acquires the speed feedback gain acquired from the servo motor 50 and the detection value data detected by the sensor 4 when the industrial machine is operating, and performs calculation to estimate the tension value of the transmission belt equipped in the industrial machine based on the acquired data. The first tension calculation unit 120 receives the acquired data from the control unit 110. The tension value of the transmission belt calculated by the first tension calculation unit 120 is the estimated value of the tension of the transmission belt when the acquired data is acquired when the industrial machine with a non-wearing transmission belt is subjected to a specific diagnostic operation.

[0032] Figure 3 The first tension calculation unit 120 of the present embodiment is a block diagram showing the general functions of the first tension calculation unit 120. The first tension calculation unit 120 includes a first preprocessing unit 122 and a first estimating unit 124. In addition, a first learning model storage unit 126 is provided on the RAM 103 or the nonvolatile memory 104. The first learning model storage unit 126 stores a first learning model that has been learned for estimating the tension value of the transmission belt in the industrial machine to which the transmission belt without loss is installed.

[0033] The first preprocessing unit 122 generates state data for the machine learning process of the machine learning device 100 based on the acquired data acquired from the control unit 110, and outputs the state data to the machine learning device 100. The first preprocessing unit 122 generates, for example, estimation data S1 including frequency response data representing a frequency-gain characteristic obtained by frequency analyzing the speed feedback gain of the servo motor 50 acquired from the control unit 110. The estimation data S1 generated by the first preprocessing unit 122 may be any data as long as it is suitable as an input to the first learning model stored in the first learning model storage unit 126.

[0034] As an example, the frequency response data as the estimation data S1 may be data obtained by sampling the gain value within a predetermined frequency range at a predetermined frequency cycle. In this case, the frequency response data as the estimation data S1 preferably includes a range including a primary resonant frequency and a primary antiresonant frequency as a pair. Figure 4 is a diagram showing an example of frequency analysis of the speed feedback gain. Figure 4In the example, there is a trough of the anti-resonance frequency at the position fa, and there is a peak of the resonance frequency at the position fb. The first pre-processing unit 122 generates a predetermined range before and after the position of the peak of the resonance frequency and the trough of the anti-resonance frequency by using the sequence data of the frequency-gain group sampled in the predetermined frequency period as the frequency response data used in the estimation data S1. In addition, in addition to the above, the frequency response data as the estimation data S1 may also include a range that includes a resonance frequency and an anti-resonance frequency in pairs of two or more times. In addition, the first pre-processing unit 122 may also include a sequence of values ​​of a predetermined range around the anti-resonance frequency and the resonance frequency in the frequency response data as the estimation data S1.

[0035] The first estimating unit 124 performs an estimation process of the tension value of the transmission belt using the first learning model stored in the first learning model storage unit 126 based on the estimation data S1 generated by the first preprocessing unit 122, and outputs the estimation result. The estimation process performed by the first estimating unit 124 is an estimation process using the first learning model generated by the well-known supervised learning. For example, when the first learning model stored in the first learning model storage unit 126 is generated as a neural network (convolutional neural network), the first estimating unit 124 inputs the estimation data S1 generated by the first preprocessing unit 122 into the neural network, and outputs the estimated value of the tension value of the transmission belt as the output thereof as the estimation result. The tension value of the transmission belt estimated by the first estimating unit 124 is output to the third tension calculation unit 140 as the first transmission belt tension value.

[0036] The first learning model stored in the first learning model storage unit 126 is a learning model that generates a learning model by performing supervised learning when a specific diagnostic action is performed in an industrial machinery equipped with a lossless transmission belt with various transmission belt tension states, using input data generated based on the acquired data (including input data of frequency response data representing frequency-gain characteristics) and the tension value of the transmission belt of the industrial machinery under the tension state of each transmission belt as label data.

[0037] Figure 1The CPU 11 of the diagnostic device 1 and the processor 101 of the machine learning device 100 respectively execute the system programs read from the ROM 12 and ROM 102, and mainly perform the calculation processing using the RAM 13 and the non-volatile memory 14 and the input processing via the interface 18 by the CPU 11, thereby realizing the second tension calculation unit 130. The second tension calculation unit 130 acquires or even detects the tension value of the transmission belt of the industrial machine. The tension value of the transmission belt acquired or detected by the second tension calculation unit 130 is the tension value of the transmission belt including the tension drop factor of the timing belt and the wear factor of the timing belt.

[0038] Figure 5 The second tension calculation unit 130 of the present embodiment is a block diagram showing the general functions of the second tension calculation unit 130. The second tension calculation unit 130 includes a tension measurement value acquisition unit 132.

[0039] The tension measurement value acquisition unit 132 acquires the tension value of the transmission belt of the industrial machine input by the operator via the input device 71. The operator, for example, stops the transmission belt of the industrial machine and measures the tension value of the transmission belt using a tension meter such as a sonic transmission belt tension meter or a micrometer. Then, the operator inputs the measured transmission belt tension value via the input device 71. The tension measurement value acquisition unit 132 outputs the acquired transmission belt tension value to the third tension calculation unit 140 as the second transmission belt tension value.

[0040] Figure 1 The CPU 11 of the diagnostic device 1 and the processor 101 of the machine learning device 100 respectively execute system programs read from the ROM 12 and ROM 102 , and mainly perform calculation processing using the RAM 13 and the nonvolatile memory 14 by the CPU 11 , thereby realizing the third tension calculation unit 140 .

[0041] Figure 6 The third tension calculation unit 140 of the present embodiment is a block diagram showing the general functions of the third tension calculation unit 140. The third tension calculation unit 140 includes a difference calculation unit 142.

[0042] The difference calculation unit 142 calculates the difference between the tension value of the first transmission belt input from the first tension calculation unit 120 and the tension value of the second transmission belt input from the second tension calculation unit 130. The tension value of the first transmission belt is a tension value of the transmission belt when only the tension drop of the timing belt is a factor, which is estimated based on data detected from the current industrial machine. The tension value of the second transmission belt is a tension value of the transmission belt that includes the tension drop factor of the timing belt and the wear factor of the timing belt. The difference calculation unit 142 calculates the value related to the wear state of the timing belt by calculating the difference between these two values. The difference calculation unit 142 can also calculate a predetermined value (for example, loss [mm] and loss rate [%) indicating the wear degree of the timing belt by applying the calculated difference value to a predetermined calculation formula determined in advance. Furthermore, the tension value of the first transmission belt and the tension value of the second transmission belt can be input in advance, and a third learning model for estimating the wear degree of the predetermined transmission belt can be generated, and the calculation for estimating the wear degree of the timing belt can be performed using the third learning model. The value calculated by the difference calculation unit 142 may be displayed and output on the display device 70 , or may be sent and output to other computers such as the fog computer 6 or the cloud server 7 via the network 5 .

[0043] The diagnostic device 1 of the present embodiment having the above-mentioned structure can calculate the degree of wear of the transmission belt of the industrial machine based on the data acquired during the specific diagnostic operation of the industrial machine and the manual measurement result performed by the operator. The control program 200 automatically acquires the data based on the specific diagnostic operation of the industrial machine, and even an unskilled operator can easily perform the manual measurement based on the tension of the transmission belt according to the manual, so that the wear of the transmission belt can be grasped relatively easily.

[0044] As a variation of the diagnostic device 1 of this embodiment, the third tension calculation unit 140 of the diagnostic device 1 may also be configured to compare a predetermined threshold value set in advance with the calculated degree of wear of the transmission belt, and output information indicating that the state of the transmission belt is abnormal when the calculated degree of wear of the transmission belt exceeds the predetermined threshold value. With such a configuration, not only the degree of wear of the transmission belt can be displayed, but also the normality / abnormality of the transmission belt can be diagnosed.

[0045] Figure 7 This is a block diagram showing the general functions of the second tension calculation unit 130 included in the diagnosis device 1 according to the second embodiment. Other functions of this embodiment are the same as those described in the first embodiment.

[0046] Figure 1The CPU 11 of the diagnostic device 1 and the processor 101 of the machine learning device 100 respectively execute the system programs read from the ROM 12 and ROM 102, and mainly perform the calculation processing using the RAM 13 and the non-volatile memory 14 and the input and output processing via the interface 18 by the CPU 11, thereby realizing the second tension calculation unit 130 of the present embodiment. The second tension calculation unit 130 of the present embodiment includes a belt tension measuring unit 133.

[0047] The transmission belt tension measuring unit 133 measures the tension value of the transmission belt of the industrial machine by using a known automatic transmission belt tension measuring method, for example, as described in the six patent documents in the related art column of this specification. By using the known automatic transmission belt tension measuring method, the tension value of the transmission belt including the tension drop factor of the timing belt and the wear factor of the timing belt can be measured. In addition, since the known automatic transmission belt tension measuring method is widely known through other documents, a detailed description of this specification is omitted. The transmission belt tension value measured by the transmission belt tension measuring unit 133 is output to the third tension calculation unit 140 as the tension value of the second transmission belt.

[0048] The diagnostic device 1 of the present embodiment having the above-described configuration can calculate the degree of wear of the transmission belt of the industrial machine based on data acquired during a specific diagnostic operation of the industrial machine and measurement results of a predetermined automatic measurement technique.

[0049] Figure 8 This is a block diagram showing the general functions of the second tension calculation unit 130 included in the diagnosis device 1 according to the third embodiment. Other functions of this embodiment are the same as those described in the first embodiment.

[0050] Figure 1 The CPU 11 of the diagnostic device 1 shown and the processor 101 of the machine learning device 100 respectively execute the system programs read from the ROM 12 and the ROM 102, and mainly perform calculation processing using the RAM 13 and the non-volatile memory 14 by the CPU 11 and calculation processing using the RAM 103 and the non-volatile memory 104 by the processor 101, and input and output processing via the interface 18 and the interface 21, thereby realizing the second tension calculation unit 130 of the present embodiment.

[0051] The second tension calculation unit 130 of this embodiment includes a second preprocessing unit 134 and a second estimating unit 135. In addition, a second learning model storage unit 136 is provided on the RAM 103 or the nonvolatile memory 104. The second learning model storage unit 136 stores a second learning model learned for estimating the tension value of the transmission belt in the industrial machine.

[0052] The second tension calculation unit 130 of the present embodiment acquires the speed feedback gain acquired from the servo motor 50 and the detection value data detected by the sensor 4 when the industrial machine is in motion, and performs calculation to estimate the tension value of the transmission belt provided in the industrial machine based on the acquired data. The second tension calculation unit 130 receives the acquired data from the control unit 110. The tension value of the transmission belt calculated by the second tension calculation unit 130 is an estimated value of the tension of the transmission belt when the acquired data is acquired when the industrial machine is subjected to a specific diagnostic operation.

[0053] The second preprocessing unit 134 generates state data for the machine learning process of the machine learning device 100 based on the acquired data acquired from the control unit 110, and outputs the state data to the machine learning device 100. The second preprocessing unit 134 generates, for example, estimation data S2 including frequency response data representing a frequency-gain characteristic obtained by frequency analyzing the speed feedback gain of the servo motor 50 acquired from the control unit 110. The estimation data S2 generated by the second preprocessing unit 134 may be any data as long as it is suitable as an input to the second learning model stored in the second learning model storage unit 136.

[0054] The second estimation unit 135 performs an estimation process of the tension value of the transmission belt using the second learning model stored in the second learning model storage unit 136 based on the estimation data S2 generated by the second preprocessing unit 134, and outputs the estimation result. The estimation process performed by the second estimation unit 135 is an estimation process using the second learning model generated by the well-known supervised learning. For example, when the second learning model stored in the second learning model storage unit 136 is generated as a neural network (convolutional neural network), the second estimation unit 135 inputs the estimation data S2 generated by the second preprocessing unit 134 into the neural network, and outputs the estimated value of the tension value of the transmission belt as the output thereof as the estimation result. The tension value of the transmission belt estimated by the second estimation unit 135 is output to the third tension calculation unit 140 as the second transmission belt tension value.

[0055] The second learning model stored in the second learning model storage unit 136 is a learning model that generates a learning model by performing supervised learning when a specific diagnostic action is performed with various belt tension states in an industrial machine equipped with belts in various wear states, using input data generated based on the acquired data (including input data of frequency response data representing frequency-gain characteristics) and the belt tension value of the industrial machine under each belt tension state as label data.

[0056] The diagnostic device 1 of the present embodiment having the above-described structure can use data acquired during a specific diagnostic operation of the industrial machine to calculate the degree of wear of the transmission belt of the industrial machine based on the tension value of the first transmission belt estimated using the first learning model and the tension value of the second transmission belt estimated using the second learning model.

[0057] As mentioned above, although one embodiment of the present invention has been described, the present invention is not limited to the example of the above-mentioned embodiment, and can be implemented in various forms by adding appropriate changes.

Claims

1. A diagnostic device for diagnosing the degree of wear of a power transmission belt of an industrial machine, characterized in that: The above-mentioned diagnostic device has: a control unit that controls a diagnostic operation of driving the transmission belt; a first tension calculation unit for calculating a tension value of the transmission belt when the transmission belt is free of wear, i.e., a first transmission belt tension value, based on the data obtained from the diagnosis operation; a second tension calculation unit for calculating a tension value of the second transmission belt including a tension drop factor of the transmission belt and a wear factor of the transmission belt; as well as a third tension calculation unit for calculating a degree of wear of the transmission belt based on the tension value of the first transmission belt and the tension value of the second transmission belt, The third tension calculation unit calculates a difference between the tension value of the first transmission belt and the tension value of the second transmission belt, or applies the calculated difference to a predetermined calculation formula to calculate a value representing the degree of wear of the transmission belt, or inputs the tension value of the first transmission belt and the tension value of the second transmission belt in advance to generate a learning model for estimating a predetermined degree of wear of the transmission belt, and estimates the degree of wear of the transmission belt using the learning model.

2. The diagnostic device according to claim 1, characterized in that The diagnostic operation is a frequency sweep operation, and the first tension calculation unit estimates the tension of the first transmission belt using a feedback signal obtained by the operation as an input.

3. The diagnostic device according to claim 1 or 2, characterized in that: The first tension calculation unit estimates the tension of the first transmission belt using a first learning model obtained by learning using an input including data acquired using a transmission belt without loss.

4. The diagnostic device according to claim 1, characterized in that The above-mentioned diagnostic action is performed at multiple positions or under multiple conditions of the transmission belt.

5. The diagnostic device according to claim 1, characterized in that The second tension calculation unit calculates a tension value of the second transmission belt including a tension drop factor of the transmission belt and a wear factor of the transmission belt by means different from that of the first tension calculation unit.

6. The diagnostic device according to claim 1, characterized in that The second tension calculation unit estimates the tension of the second transmission belt using a second learning model obtained by learning including input of data acquired using transmission belts having various degrees of wear.

7. The diagnostic device according to claim 1, characterized in that The second tension calculation unit acquires the tension value of the transmission belt measured by the measuring device, and sets the acquired value as the tension value of the second transmission belt.

8. The diagnostic device according to claim 1, characterized in that The above-mentioned industrial machine is an injection molding machine.

Images