Fault diagnosis method and system for transmission system

Abstract

The invention discloses a fault diagnosis method and system for a transmission system. The method comprises the steps that a friction power generation sensor is installed on a coupler of the transmission system; collecting output signals of the friction power generation sensor of the transmission system under different working conditions; performing feature extraction and calibration on the output signal to obtain a data set; training and testing a support vector machine by using the data set to obtain a fault diagnosis model; and intelligent fault diagnosis of the transmission system is carried out based on the fault diagnosis model. The friction power generation sensor is installed on the elastic body of the coupler in a specific installation mode, the amplitude of output signals of the friction power generation sensor is effectively improved, the characteristic difference of the output signals under different working conditions is made to be obvious, the accuracy of subsequent model fault type recognition is effectively improved, the structure is compact, installation is convenient and fast, and the application range is wide; a machine learning algorithm is combined to realize automatic identification of transmission system fault types, and the intelligent degree of industrial automation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of transmission system fault diagnosis, and in particular to a transmission system fault diagnosis method and system. Background Art

[0002] With the development of industrial intelligent manufacturing and artificial intelligence, transmission system sensors play a key role in modern industry. They are widely used to monitor and control the operating status of transmission systems to ensure the safety and reliability of equipment. In recent years, they tend to develop in the aspects of high precision, high sensitivity, multi-sensor data fusion, and automated processing.

[0003] At present, there are many implementation plans for transmission system fault diagnosis, most of which use vibration probes as sensor devices. Although they have wide applicability and can respond to various faults, their vibration signals are mixed, the causes of vibration are complex and difficult to identify, and the difference between normal signals and fault signals is very small. Therefore, it is difficult to analyze the fault type through their vibration signals.

[0004] Therefore, the prior art still needs to be improved and developed. Summary of the invention

[0005] The main purpose of the present invention is to provide a transmission system fault diagnosis method and system, aiming to solve the problem in the prior art that when using sensors to diagnose transmission system faults, the difference between the healthy signal and the fault signal is very small, resulting in difficulty in identifying the fault type.

[0006] To achieve the above object, the present application provides a transmission system fault diagnosis method, the transmission system fault diagnosis method comprising:

[0007] Install a friction-generating sensor on the coupling of the transmission system;

[0008] Collecting the output signal of the friction power generation sensor of the transmission system under different working conditions;

[0009] Performing feature extraction and calibration on the output signal to obtain a data set;

[0010] Using the data set to train and test a support vector machine to obtain a fault diagnosis model;

[0011] The target output signal of the friction power generation sensor in the transmission system to be diagnosed is collected in real time, the target output signal is input into the fault diagnosis model to perform transmission system fault diagnosis, and the fault type of the transmission system to be diagnosed is output.

[0012] Optionally, the coupling includes two steel claws and a plum blossom-shaped elastic body as an intermediate body connecting the two steel claws;

[0013] The friction power generation sensor is installed on the coupling of the transmission system, specifically:

[0014] The friction power generation sensor is embedded in the gap between the plum blossom-shaped elastomer and the steel claw, and the friction power generation sensor is installed on any surface of several extrusion surfaces of the plum blossom-shaped elastomer in a specific installation method; wherein, the specific installation method refers to an installation method that makes the output signal amplitude of the friction power generation sensor greater than a preset value and the difference in the output signal characteristics of the friction power generation sensor under different working conditions of the transmission system is greater than a threshold.

[0015] Optionally, the friction power generation sensor consists of a three-layer structure of a friction power generation sheet, a conductive copper foil and a PI insulating tape. The friction power generation sensor is used to collect the vibration signal between the plum blossom-shaped elastomer and the steel claw, and convert the vibration signal into a voltage signal.

[0016] Optionally, the collecting the output signal of the friction power generation sensor of the transmission system under different working conditions specifically includes:

[0017] Collecting the output signal of the friction power generation sensor when the transmission system is in a healthy state and using it as a healthy signal;

[0018] Collecting the output signal of the friction power generation sensor under the simulated fault state of the transmission system and using it as a fault signal;

[0019] The simulated fault conditions include belt drive slippage, drive shaft misalignment and gear wear.

[0020] Optionally, the extracting and calibrating the output signal to obtain a data set specifically includes:

[0021] Performing a fast Fourier transform on the output signal to obtain a frequency domain signal corresponding to the output signal;

[0022] Respectively extracting a time domain eigenvalue of the output signal and a frequency domain eigenvalue of a frequency domain signal corresponding to the output signal, wherein both the time domain eigenvalue and the frequency domain eigenvalue include a standard deviation, a root mean square, a waveform factor, and a peak-to-peak value;

[0023] Marking the healthy signal and the fault signal in the output signal to obtain the label of the output signal;

[0024] A data set is constructed based on the feature values ​​and labels of the output signals.

[0025] Optionally, the using the data set to train and test a support vector machine to obtain a fault diagnosis model specifically includes:

[0026] Dividing the data set into a training set and a test set;

[0027] Based on the improved particle swarm algorithm, determining the optimal penalty factor and the optimal kernel function parameter of the support vector machine;

[0028] According to the optimal penalty factor and the optimal kernel function parameter, the support vector machine is trained using the training set to obtain a classification model of the support vector machine;

[0029] The classification model of the support vector machine is tested using the test set to verify the performance of the classification model of the support vector machine, and the verified classification model of the support vector machine is used as a fault diagnosis model for transmission system fault diagnosis.

[0030] In addition, to achieve the above-mentioned purpose, the present invention also provides a transmission system fault diagnosis system, which is used to implement the steps in any of the above-mentioned transmission system fault diagnosis methods, wherein the transmission system fault diagnosis system includes a transmission system test bench and a friction power generation sensor;

[0031] The transmission system test bench includes a base and a variable frequency motor, a speed sensor, a gear box, a torque sensor, a conveyor pulley, a conveyor belt, a magnetic powder brake and a control console arranged on the base. The motor shaft of the variable frequency motor is connected to the input shaft of the gear box through a coupling. The coupling includes two steel claws and a plum blossom-shaped elastomer as an intermediate connecting the two steel claws. The friction power generation sensor is embedded in the gap between the plum blossom-shaped elastomer and the steel claws. The friction power generation sensor is installed on any surface of several extrusion surfaces of the plum blossom-shaped elastomer in a specific installation manner.

[0032] Optionally, the friction power generation sensor consists of a three-layer structure of a friction power generation sheet, a conductive copper foil and a PI insulating tape. The friction power generation sensor is used to collect the vibration signal between the plum blossom-shaped elastomer and the steel claw, and convert the vibration signal into a voltage signal.

[0033] Optionally, the fault diagnosis system of the transmission system also includes a conductive slip ring, an oscilloscope and a host computer, the conductive slip ring is installed on the motor shaft where the coupling is located, the friction power generation sensor is connected to one end of the conductive slip ring, the other end of the conductive slip ring is connected to one end of the oscilloscope, and the other end of the oscilloscope is connected to the host computer.

[0034] Optionally, the conductive slip ring is used to derive the output signal of the friction power generation sensor, the oscilloscope is used to collect the output signal, the host computer uses MATLAB software to extract features and calibrate the output signal to obtain a data set, and the host computer also uses a support vector machine algorithm to classify and identify the data set and output the fault type of the transmission system.

[0035] Beneficial effects: Compared with the prior art, the present invention provides a fault diagnosis method and system for a transmission system, the method comprising: installing a friction power sensor on the coupling of the transmission system; collecting the output signal of the friction power sensor under different working conditions of the transmission system; extracting features and calibrating the output signal to obtain a data set; using the data set to train and test a support vector machine to obtain a fault diagnosis model; and performing intelligent diagnosis of transmission system faults based on the fault diagnosis model. The present invention uses a specific installation method to install a friction power sensor on the elastic body of the coupling, which effectively improves the amplitude of the output signal of the friction power sensor, makes the output signal characteristics of different working conditions significantly different, and effectively improves the accuracy of subsequent model identification of fault types, and has a compact structure, convenient installation, and a wide range of applications; combined with a machine learning algorithm to realize automatic identification of transmission system fault types, it is conducive to improving the intelligence of industrial automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of a preferred embodiment of the conventional system fault diagnosis method of the present invention;

[0037] Figure 2 is a structural diagram of the coupling and the friction power generation sensor of the present invention;

[0038] Figure 3 is a structural diagram of the elastomer of the present invention;

[0039] Figure 4 It is a schematic diagram of installing the bilateral electrodes on the elastic body of the present invention;

[0040] Figure 5 It is a schematic diagram of the installation of a single-side electrode on an elastic body of the present invention;

[0041] Figure 6 This is a comparison diagram of the average voltage of the fault signal and the healthy signal under the 1st to 6th installation modes of the single-side electrode 5-piece friction power generation sensor of the present invention;

[0042] Figure 7 This is a comparison diagram of the average voltage of the fault signal and the healthy signal of the 5-piece friction power generation sensor with a single electrode of the present invention under the 7th to 12th installation modes;

[0043] Figure 8 It is a schematic diagram of the structure of the slip fault of the belt transmission system of the present invention;

[0044] Fig. 9 It is a voltage diagram of a healthy signal and a fault signal under two installation modes of the present invention, namely, "single-side electrode 1-2-3-4-5-RLLRL" and "single-side electrode 1-2-3-4-5-RLRLR";

[0045] Fig.10It is a data table diagram of characteristic values ​​and output values ​​in the support vector machine algorithm of the present invention;

[0046] Fig.11 is a flow chart of the support vector machine algorithm of the present invention;

[0047] Fig.12 It is a comparison chart of the prediction results of the training set and the test set in the support vector machine algorithm of the present invention;

[0048] Fig.13 This is a graph showing the signal changes of the present invention after 6000 consecutive cycles of operation;

[0049] Fig.14 It is a structural diagram of a preferred embodiment of the conventional system fault diagnosis system of the present invention;

[0050] Fig.15 It is a circuit connection schematic diagram of the friction power generation sensor of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] With the rapid progress of information technology and industrial generation technology, modern industrial equipment has gradually become larger, more complex, faster, more integrated and intelligent, and the requirements for equipment stability and safety are becoming higher and higher. In order to prevent mechanical equipment, electrical equipment, and transmission systems from shutting down due to aging, mechanical failure and other failures, which affects normal production and life, it is necessary to study appropriate fault diagnosis methods to quickly identify the type of fault, perform repairs in a timely manner, and avoid major economic losses. Fault diagnosis, as one of the key technologies to maintain safe and reliable operation of equipment, aims to quickly and accurately determine the type of fault, etc., and reduce the operation and maintenance costs of equipment.

[0053] The transmission system is a key component in the operation of mechanical equipment and plays a key role in transmitting kinetic energy. The high failure rate of the transmission system is an important reason for the downtime of mechanical equipment, so real-time monitoring and fault diagnosis of the transmission system are of great significance to ensure the operation of equipment. With the development of industrial intelligent manufacturing and artificial intelligence, transmission system sensors play a key role in modern industry. They are widely used to monitor and control the operating status of the transmission system to ensure the safety and reliability of the equipment. In recent years, it has tended to develop in the aspects of high precision, high sensitivity, multi-sensor data fusion, and automated processing.

[0054] At present, there are many implementation plans for transmission system fault diagnosis, most of which use vibration probes as sensor devices. Although they have wide applicability and can respond to various faults, their vibration signals are mixed, the causes of vibration are complex and difficult to identify, and the difference between normal signals and fault signals is very small. Therefore, it is difficult to analyze the fault type through their vibration signals.

[0055] In order to solve the above technical problems, the present invention provides a transmission system fault diagnosis method and system, the transmission system fault diagnosis method comprising: installing a friction power sensor on the coupling of the transmission system; collecting the output signal of the friction power sensor of the transmission system under different working conditions; extracting features and calibrating the output signal to obtain a data set; using the data set to train and test a support vector machine to obtain a fault diagnosis model; collecting the target output signal of the friction power sensor in the transmission system to be diagnosed in real time, and inputting the target output signal into the fault diagnosis model to perform transmission system fault diagnosis. The transmission system fault diagnosis system comprises a transmission system test bench and a friction power sensor, the transmission system test bench comprises a base and a variable frequency motor, a speed sensor, a gear box, a torque sensor, a conveyor pulley, a conveyor belt, a magnetic powder brake and a control console arranged on the base, the motor shaft of the variable frequency motor is connected to the input shaft of the gear box through a coupling, the coupling comprises two steel claws and a plum blossom-shaped elastic body as an intermediate connecting the two steel claws, the friction power sensor is embedded in the gap between the plum blossom-shaped elastic body and the steel claw, and the friction power sensor is installed on any of the 12 extrusion surfaces of the plum blossom-shaped elastic body in a specific installation manner. The present invention installs a friction power generation sensor on the plum blossom-shaped elastomer of the coupling in a specific installation method, and uses the friction power generation sensor to collect the vibration signal between the elastomer and the steel claw and convert it into a voltage signal. The voltage signal is used to reflect the working state of the transmission system, and then the voltage signal is processed and analyzed using a machine learning algorithm to automatically identify the corresponding fault type. The present application uses a specific installation method to install a friction power generation sensor on the elastomer of the coupling, which effectively improves the amplitude of the output signal of the friction power generation sensor, so that the characteristics of the output signal under different working conditions are obviously different, and the accuracy of subsequent model identification of fault types is effectively improved. In addition, the structure is compact, the installation is convenient, and the application range is wide. The present application combines a machine learning algorithm to realize the automatic identification of transmission system fault types, which is conducive to improving the intelligence level of industrial automation.

[0056] The application content is further explained below through the description of embodiments in conjunction with the accompanying drawings.

[0057] A preferred embodiment of the fault diagnosis method of the transmission system of the present invention is as follows: Figure 1 As shown, specifically including:

[0058] S1. Install a friction-generating sensor on the coupling of the transmission system.

[0059] In one implementation of this embodiment, Figure 2 As shown, the coupling 4 includes two steel claws 1 and a plum blossom-shaped elastomer 2 as an intermediate connecting the two steel claws 1, and the friction power generation sensor 3 is composed of a three-layer structure of a friction power generation sheet 301, a conductive copper foil 302 and a PI insulating tape 303; the friction power generation sensor is installed on the coupling of the transmission system in detail: the friction power generation sensor is embedded in the gap between the plum blossom-shaped elastomer and the steel claw, and the friction power generation sensor is installed on any surface of several extrusion surfaces of the plum blossom-shaped elastomer (the elastomer in this embodiment has 12 extrusion surfaces) in a specific installation method; wherein, the specific installation method refers to an installation method in which the amplitude of the output signal of the friction power generation sensor is greater than a preset value (the preset value in this embodiment is set to 5V) and the difference in the output signal characteristics of the transmission system under different working conditions is greater than a threshold, that is, the difference in the output signal characteristics is obvious.

[0060] Specifically, the coupling is used to connect the motor shaft of the variable frequency motor with the input shaft of the gearbox, including a plum blossom-shaped elastomer and two steel claws. The friction power generation sensor is embedded in the gap between the elastomer and the steel claws. The friction power generation sensor is composed of multiple units connected in parallel and installed at different positions of the 12 extrusion surfaces of the elastomer. The elastomer serves as an intermediate connecting the two steel claws, which can transmit the speed and torque of the motor shaft to the input shaft of the rear gearbox. Therefore, the friction power generation sensor embedded therein can convert the vibration signal on the motor shaft into a voltage signal. It can be understood that the friction power generation sensor is installed on the coupling elastomer in a specific arrangement. When the transmission system test bench is working, a small contact separation and friction are generated between the steel claws and the elastomer of the coupling, so that the friction power generation sensor installed thereon generates a friction electric signal, which can reflect the working state of the transmission system. According to the principle of friction power generation, the present invention realizes friction power generation self-driven sensing through a special assembly method, and effectively improves the amplitude of the output signal, and significantly enhances the ability to distinguish fault types.

[0061] Different installation methods of triboelectric sensors respond differently to faults. Through experimental exploration, we found a specific installation method that makes the output signals in healthy and faulty states have significant differences in amplitude, frequency and other characteristics, which is conducive to improving the fault recognition rate.

[0062] Specifically, different installation methods of triboelectric sensors generate different output signals. In order to explore the best installation method of triboelectric sensors on the coupling, the present invention conducted batch experiments. Figure 3 As shown, the coupling's elastic body 2 has 6 blade branches, and each blade branch has two mounting surfaces (L and R). After permutations and combinations and elimination of equivalent repetitive schemes, there are 75 different mounting methods. Figure 4 As shown, Figure 4The schematic diagram of the installation of the bilateral electrode 1-2-4 is shown in FIG. 1 , 2 and 4. A friction power generation sensor 3 is installed on each of the left and right sides of the leaf branches 1, 2 and 4 of the elastic body 2. Figure 5 As shown, Figure 5 This is a schematic diagram of the installation of a single-sided electrode 1-2-3-5-LRLR, with a friction power generation sensor 3 installed on the left side of the 1-blade branch of the elastomer 2, on the right side of the 2-blade branch, on the left side of the 3-blade branch, and on the right side of the 5-blade branch.

[0063] Before each round of experiments, the friction power sensor needs to be installed on the coupling. There are 75 specific installation methods as mentioned above. The number and position of the friction power sensor installation will affect the quality of the signal. The present invention compares the sensor output signals under 75 different friction power sensor installation methods and obtains the best signal installation method.

[0064] Specifically, after collecting healthy and fault signals under various installation methods, in order to select the best installation method, the following definition of "good signal" was made: the signal amplitude is greater than 5V and the fluctuation is stable within 6 seconds; the fault signal is significantly different from the healthy signal, for example, its signal peak-to-peak value, frequency, waveform factor, etc. must be significantly different to facilitate subsequent fault identification. Based on this standard, some installation methods with better effects were selected. Taking the simulation of belt drive slip as an example, Figure 6 and Figure 7 As shown in the figure, the average voltage comparison of each fault and healthy signal under 12 installation methods of 5-piece friction power generation sensor with single-side electrode is shown. Figure 6 and Figure 7The middle horizontal axis is the number of the healthy signal or fault signal under different installation methods. 1a represents the healthy signal under the installation method of "single-side electrode 1-2-3-4-5-RLRLL", and 1b represents the fault signal under the installation method of "single-side electrode 1-2-3-4-5-RLRLL"; 2a represents the healthy signal under the installation method of "single-side electrode 1-2-3-4-5-RLRLR", and 2b represents the fault signal under the installation method of "single-side electrode 1-2-3-4-5-RLRLR"; 3a represents the healthy signal under the installation method of "single-side electrode 1-2-3-4-5-RLLRL", and 3b represents the fault signal under the installation method of "single-side electrode 1-2-3-4-5-RLLRL"; 4 a indicates the healthy signal under the installation mode of "single-side electrode 1-2-3-4-5-RLLLL", 4b indicates the fault signal under the installation mode of "single-side electrode 1-2-3-4-5-RLLLL"; 5a indicates the healthy signal under the installation mode of "single-side electrode 1-2-3-4-5-RLLLR", 5b indicates the fault signal under the installation mode of "single-side electrode 1-2-3-4-5-RLLLR"; 6a indicates the healthy signal under the installation mode of "single-side electrode 1-2-3-4-5-RLLRR", 6b indicates the fault signal under the installation mode of "single-side electrode 1-2-3-4-5-RLLRR"; 7a indicates the healthy signal under the installation mode of "single-side electrode 1-2-3-4-5-LRLLL" 7a indicates the healthy signal under the installation mode of “single-side electrode 1-2-3-4-5-LRLLL”, 8b indicates the fault signal under the installation mode of “single-side electrode 1-2-3-4-5-RRLLL”; 9a indicates the healthy signal under the installation mode of “single-side electrode 1-2-3-4-5-RRLLR”, 9b indicates the fault signal under the installation mode of “single-side electrode 1-2-3-4-5-RRLLR”; 10a indicates the healthy signal under the installation mode of “single-side electrode 1-2-3-4-5-LLLLL”, 10b indicates the fault signal under the installation mode of “single-side electrode 1-2-3- 11a represents the healthy signal under the installation mode of “single-side electrode 1-2-3-4-5-LLRRR”, 11b represents the fault signal under the installation mode of “single-side electrode 1-2-3-4-5-LLRRR”; 12a represents the healthy signal under the installation mode of “single-side electrode 1-2-3-4-5-LRRRR”, 12b represents the fault signal under the installation mode of “single-side electrode 1-2-3-4-5-LRRRR”; among them, the two installation modes of “single-side electrode 1-2-3-4-5-RLLRL” and “single-side electrode 1-2-3-4-5-RRLLR” have the best effect in all installation schemes and are determined to be the best installation methods.

[0065] The present invention optimizes the installation method of the friction power generation sensor to increase the signal amplitude from less than 5V to more than 10V, and the signal is significantly different from the normal state in the case of faults such as belt transmission slippage.

[0066] S2. Collect the output signal of the friction power generation sensor of the transmission system under different working conditions.

[0067] In one implementation of this embodiment, collecting the output signal of the friction power generation sensor of the transmission system under different working conditions specifically includes:

[0068] S21, collecting the output signal of the friction power generation sensor when the transmission system is in a healthy state and using it as a healthy signal;

[0069] S22, collecting the output signal of the friction power generation sensor in the transmission system simulated fault state and using it as a fault signal;

[0070] Among them, the simulated fault conditions include belt drive slippage, drive shaft misalignment and gearbox wear.

[0071] Specifically, the sensor embedded in the coupling designed based on the principle of friction power generation can respond differently to normal operation, belt drive slippage, drive shaft misalignment and other working conditions, and output different signals. In order to realize automatic fault diagnosis of the transmission system, it is necessary to collect the sensor signal under the fault first. Therefore, the output signal of the friction power generation sensor under the simulated fault condition of the transmission system is collected by presetting the fault, and compared with the signal under the healthy state, which lays the foundation for subsequent fault diagnosis.

[0072] For example: Figure 8 As shown, Figure 8 This is a schematic diagram of the structure of a belt drive system slipping fault. The driving wheel 16 rotates, and the driven wheel 17 and the conveyor belt 14 slip. In this embodiment, the magnetic powder brake is used as the load of the transmission system. The braking power of the magnetic powder brake is adjusted to increase the load, causing the belt drive to slip, thereby simulating a belt drive system slipping fault. The differential power of the magnetic powder brake is 2kw, and the input speed of the three-phase motor is 600rpm. Measure the signals of the healthy and slipping states. After the equipment is installed, turn on the motor, return the magnetic powder brake to zero, and start recording the healthy state signal when all the drive shafts work normally. After saving, adjust the magnetic powder brake immediately. When the differential power is around 2kw, the belt drive overloads and slips. The output signal of the friction power sensor at this time is recorded as the slipping state signal (fault signal). Fig. 9 As shown, Fig. 9 (a) is the health signal collected under the installation mode of "single-side electrode 1-2-3-4-5-RLLRL", Fig. 9 (b) is the slip fault signal recorded under the installation mode of "single-side electrode 1-2-3-4-5-RLLRL". Fig. 9(c) is the health signal collected under the installation mode of "single-side electrode 1-2-3-4-5-RLRLR", Fig. 9 (d) is the slip fault signal recorded under the installation mode of “single-side electrode 1-2-3-4-5-RLRLR”.

[0073] S3. Extract and calibrate the output signal to obtain a data set.

[0074] In an implementation of this embodiment, feature extraction and calibration are performed on the output signal to obtain a data set, which specifically includes:

[0075] S31, performing fast Fourier transform on the output signal to obtain a frequency domain signal corresponding to the output signal;

[0076] S32, respectively extracting the time domain eigenvalues ​​of the output signal and the frequency domain eigenvalues ​​of the frequency domain signal corresponding to the output signal, wherein both the time domain eigenvalues ​​and the frequency domain eigenvalues ​​include standard deviation, root mean square, waveform factor and peak-to-peak value;

[0077] S33, calibrating the healthy signal and the fault signal in the output signal to obtain a label of the output signal;

[0078] S34. Construct a data set based on the feature values ​​and labels of the output signal.

[0079] In this embodiment, 10 groups of healthy signals and fault signals are collected respectively to avoid the influence of accidental conditions. Feature extraction is performed on the 20 groups of signals, and four characteristic values ​​of standard deviation (Std), root mean square (RMS), waveform factor (SF), and peak-to-peak value (PP) are selected here. The difference between different signals can be accurately analyzed by extracting the characteristic values ​​in the signal. Among them, the standard deviation (Std) reflects the degree of discreteness of the waveform signal. The larger the value, the higher the discreteness of the signal; the root mean square (RMS) reflects the effective value of the signal. The higher the amplitude and effective proportion of the signal, the greater its effective value; the waveform factor reflects the shape characteristics of the signal, such as the degree of distortion of the waveform, which means the ratio of the signal root mean square to the rectified average value; the peak-to-peak value (PTP) is the difference between the maximum positive value and the maximum negative value in the waveform diagram, reflecting the range of signal amplitude variation. The larger the peak-to-peak value, the larger the signal variation range.

[0080] In order to analyze the signal more comprehensively and understand the amplitude and frequency characteristics of the signal, the signal was subjected to fast Fourier transform (FFT), and four eigenvalues ​​of the time domain and frequency domain of the 20 groups of signals were extracted to form a 20×9 data set, of which the first 8 columns are the time domain and frequency domain eigenvalues ​​of the data, and the 9th column is the output value. The healthy data is calibrated as output 1, and the fault data is calibrated as output 2.

[0081] S4. Use the data set to train and test the support vector machine to obtain a fault diagnosis model.

[0082] In one implementation of this embodiment, a support vector machine is trained and tested using a data set to obtain a fault diagnosis model, specifically including:

[0083] S41, dividing the data set into a training set and a test set;

[0084] S42. Determine the optimal penalty factor and optimal kernel function parameters of the support vector machine based on the improved particle swarm algorithm;

[0085] S43, training the support vector machine using the training set according to the optimal penalty factor and the optimal kernel function parameter to obtain a classification model of the support vector machine;

[0086] S44. Testing the classification model of the support vector machine using the test set to verify the performance of the classification model of the support vector machine, and using the verified classification model of the support vector machine as a fault diagnosis model for transmission system fault diagnosis.

[0087] In this embodiment, a support vector machine (SVM) algorithm is used to classify and identify the data set to achieve automatic identification and diagnosis of healthy and faulty data. The basic idea is to find an optimal hyperplane (a straight line in two-dimensional space) by mapping the data to a high-dimensional space, so that one side of the hyperplane is one category and the other side is another category. Among them, the steps of using the support vector machine algorithm to achieve automatic fault diagnosis are training first and then prediction. First, a large amount of data is used for training to find an optimal hyperplane, that is, a trained model, and then the hyperplane is used to classify the test data. Finally, the deviation between the classification result and the manual calibration result is compared, and the predicted error value is output.

[0088] There are two important parameters that need to be confirmed in the support vector machine algorithm, the penalty factor C and the kernel function parameter g. The penalty factor C is used to adjust the classifier's attention to errors. The smaller the C value, the higher the classifier's tolerance for errors, but too small a value may lead to poor classification results; the larger the C value, the stricter the classification, but too large a value will make the classifier overly sensitive to the training data, resulting in overfitting. The kernel function parameter g represents the width of the kernel function. The smaller g is, the narrower the bandwidth of the kernel function, making the SVM model simpler, but too small a value may not capture the complex structure in the data, affecting the classification accuracy; the larger the g value is, the wider the bandwidth of the kernel function is, which can capture more subtle structures in the data, but it also increases the risk of overfitting. Fig.10 As shown, the present invention has obtained the best penalty factor C of 10 and the kernel function parameter g of 0.1 through multiple experiments and comparison of the output prediction errors.

[0089] Based on the above, if Fig.11As shown in the figure, the overall SVM algorithm to implement transmission system fault diagnosis process includes: obtaining the initial signal and performing low-pass filtering to remove high-frequency interference; extracting the signal time domain and frequency domain eigenvalues ​​to obtain a data set; then inputting the above 20 columns of 8-dimensional data sets, dividing 80% into training sets and 20% into test sets, adjusting the penalty factor to 10, the kernel function parameter to 0.1, and running the training model; after running for a period of time, the output prediction results are obtained. In this experiment, Fig.12 As shown in the figure, the true values ​​of the training set and the test set are compared with the predicted values. It can be seen from the figure that the true values ​​are highly consistent with the predicted values, and their root mean square errors are less than 0.02, showing a very high fault diagnosis accuracy.

[0090] S5. Collect the output signal of the friction power generation sensor in the transmission system to be diagnosed in real time, and input the output signal into the fault diagnosis model to perform transmission system fault diagnosis.

[0091] In order to test the durability of the sensor, the present invention conducted a long-term experiment of 6000 cycles. Fig.13 As shown, the experiment proves that the voltage signal of the friction power generation sensor of the present invention has no obvious attenuation after running continuously and smoothly for 6000 cycles, from about 7V at the beginning to about 8.5V at the end. Although there is a slight change in the middle, the output voltage is stable at about 8V throughout. This shows that the friction power generation sensor of the present invention has good durability and is suitable for long-term work.

[0092] This embodiment also provides a transmission system fault diagnosis system, which is used to implement the steps in the transmission system fault diagnosis method described above, such as Fig.14 As shown, a preferred embodiment of the fault diagnosis system of the transmission system includes a transmission system test bench and a friction power generation sensor. The transmission system test bench includes a base 8 and a variable frequency motor 9, a speed sensor 10, a gear box 11, a torque sensor 12, a conveyor pulley 13, a conveyor belt 14, a magnetic powder brake 15 and a control console arranged on the base 8. The motor shaft of the variable frequency motor 9 is connected to the input shaft of the gear box 11 through a coupling 4, as shown in FIG. Figure 2 As shown, the coupling 4 includes two steel claws 1 and a plum blossom-shaped elastomer 2 as an intermediate connecting the two steel claws 1. The friction power generation sensor 3 is embedded in the gap between the plum blossom-shaped elastomer 2 and the steel claws 1. The friction power generation sensor 3 is installed on any surface of several extrusion surfaces (the elastomer in this embodiment has 12 extrusion surfaces) of the plum blossom-shaped elastomer 3 in a specific installation manner.

[0093] Specifically, the present invention includes a self-made friction power generation sensor 3, a coupling 4, a commercial conductive slip ring 5, an oscilloscope 6, a host computer 7 and a transmission system test bench, the test bench includes a base 8, a variable frequency motor 9, a speed sensor 10, a gear box 11, a torque sensor 12, a conveyor pulley 13, a conveyor belt 14, a magnetic powder brake 15 and a control console, and the overall device is as follows Fig.14 As shown. Figure 2 As shown, the friction power generation sensor 3 is installed in the gap between the elastomer 2 and the steel claw 1 of the coupling 4. The friction power generation sensor 3 is composed of a three-layer structure of a friction power generation sheet 301, a conductive copper foil 302, and a PI insulating tape 303. The friction power generation material can be one or more of PTFE film, PVDF film, PDMS film, etc. The coupling elastomer 2 is an intermediate connecting the two steel claws 1 (the other is not shown), which can transmit the speed and torque of the motor shaft to the input shaft of the rear gearbox, so the friction power generation sensor 3 embedded therein can convert the vibration signal on the shaft into an electrical signal. Furthermore, the sheet-shaped friction power generation sensor can be installed on any of the 12 extrusion surfaces of the plum blossom-shaped elastomer, and multiple sensors can be connected in parallel to form a multi-sensor array by connecting the conductive copper foil of each friction power generation sensor. The plum blossom-shaped coupling is a standard part, and the conductive slip ring and wire are commercial products. The oscilloscope model is RIGOL DS1000Z-E.

[0094] In this embodiment, a triboelectric sensor is installed on the elastomer of the plum blossom coupling in a specific arrangement. When the transmission system test bench is working, a slight contact separation and friction is generated between the steel claws of the coupling and the elastomer, so that the triboelectric sensor installed thereon generates a triboelectric signal, which can reflect the working state of the transmission system. In other words, the triboelectric sensor of the present invention is installed on the coupling connecting the motor shaft and the gearbox shaft, and monitors the working state of the entire transmission system by detecting the working state of the motor shaft in real time.

[0095] Different installation methods of triboelectric sensors respond differently to faults. Through experimental exploration, a specific installation method is found, so that the signals output in healthy and faulty states have significant differences in amplitude, frequency and other characteristics, which is conducive to improving the fault recognition rate. Identifiable transmission system fault types include belt drive slip, drive shaft misalignment, gear wear, etc. After using a specific triboelectric sensor installation method, experiments are conducted and data are collected, and the voltage signal data is collected using an oscilloscope. The signal is processed and calibrated and classified using a host computer (such as a PC) using MATLAB software, and the model is trained by a support vector machine (SVM) algorithm, and finally the fault signal is diagnosed and identified. According to the principle of triboelectric generation, the present invention realizes triboelectric self-driven sensing through a special assembly method, and effectively improves the amplitude of the output signal, significantly enhances the ability to distinguish fault types, and combines machine learning algorithms to realize automatic diagnosis of real transmission system faults. Its advantages are sensitive signals, convenient installation, and easy to form an automated sensor network, which is conducive to improving the intelligence of industrial automation.

[0096] In one implementation of this embodiment, Figure 2 As shown, the friction power generation sensor 3 is composed of a three-layer structure of a friction power generation sheet 301, a conductive copper foil 302 and a PI insulating tape 303. The friction power generation sensor 3 is used to collect the vibration signal between the plum blossom-shaped elastic body 2 and the steel claw 1, and convert the vibration signal into a voltage signal. The sensor based on the principle of friction power generation in this design can directly collect the friction and vibration signals between the coupling elastic body and the steel claw, and convert them into voltage signals through the principle of friction power generation. It has the characteristics of rapid response, self-drive and compact structure.

[0097] In one implementation of this embodiment, the fault diagnosis system of the transmission system also includes a conductive slip ring, an oscilloscope and a host computer. The conductive slip ring is installed on the motor shaft where the coupling is located. The friction power generation sensor is connected to one end of the conductive slip ring, the other end of the conductive slip ring is connected to one end of the oscilloscope, and the other end of the oscilloscope is connected to the host computer.

[0098] Specifically, the friction power sensor is installed in the coupling, and two signal wires are connected to the inner ring wire of the conductive slip ring. The conductive slip ring is installed on the motor shaft where the coupling is located. The outer ring is fixed to the base. The inner ring wire is connected to the outer ring wire through the brush inside the conductive slip ring, and finally connected to the signal probe of the oscilloscope. The output end of the oscilloscope is connected to the host computer. The sampling frequency of the oscilloscope is 2kHz, and the sampling time is 6 seconds. Fig.15 As shown, the signal probe at one end of the oscilloscope 6 is connected to the conductive copper foil 302 of the friction power generation sensor 3 through the conductive slip ring 5, and the other end of the oscilloscope 6 and the steel claw 1 of the coupling 4 are both grounded.

[0099] In one implementation of this embodiment, a conductive slip ring is used to derive the output signal of a friction power sensor, an oscilloscope is used to collect the output signal, and a host computer uses MATLAB software to perform feature extraction and calibration on the output signal to obtain a data set. The host computer also uses a support vector machine algorithm to classify and identify the data set and output the fault type of the transmission system.

[0100] Specifically, the voltage signal generated by the friction power sensor is derived by a wire through a conductive slip ring installed on the transmission shaft to avoid wiring harness entanglement caused by rotation. The derived signal is connected to the oscilloscope via the oscilloscope's signal probe to preview and save the signal. The oscilloscope is connected to the host computer, and the saved data can be imported, and the data can be analyzed, processed and calibrated using MATLAB software. Finally, the support vector machine (SVM) algorithm is used to train the data set to achieve automatic recognition and diagnosis of different sensor signals.

[0101] In view of the existing working condition monitoring sensor solutions embedded in couplings on the market, the difference of the present invention is that different friction power sensor installation methods are used for different transmission system fault types, which has a significant gain effect on the sensor signal compared with the conventional solution. It can be understood that the present invention realizes friction power self-driven sensing through a special assembly method, and effectively improves the amplitude of the output signal, significantly enhances the ability to distinguish fault types, and combines machine learning algorithms to realize automatic diagnosis of real transmission system faults.

[0102] In summary, the sensor embedded in the coupling designed based on the principle of friction power generation in the present invention can react differently to normal operation, belt drive slippage, misalignment of the transmission shaft and other working conditions, output different signals, and automatically identify the corresponding fault type by processing and analyzing the output signal using a machine learning algorithm; through different friction power generation sensor installation combination experiments, a combination with obvious differences in characteristics such as strong output signal, sensitive fault response, and signal amplitude frequency is explored, which is very beneficial to the signal processing and identification at the back end, and can effectively improve the accuracy of corresponding fault identification. The present invention uses sheet-like friction power generation materials, which are embedded in the gap between the coupling elastomer and the steel claw. It is small in size and has no effect on the performance of the coupling itself. Compared with traditional probe sensors, it does not require additional power supply, is easy to install, and can be widely used in various transmission systems with couplings.

[0103] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or terminal including the element.

[0104] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A fault diagnosis method for a transmission system, characterized in that: The fault diagnosis method of the transmission system comprises: Install a friction-generating sensor on the coupling of the transmission system; Collecting the output signal of the friction power generation sensor of the transmission system under different working conditions; Performing feature extraction and calibration on the output signal to obtain a data set; Using the data set to train and test a support vector machine to obtain a fault diagnosis model; The target output signal of the friction power generation sensor in the transmission system to be diagnosed is collected in real time, the target output signal is input into the fault diagnosis model to perform transmission system fault diagnosis, and the fault type of the transmission system to be diagnosed is output.

2. The fault diagnosis method of the transmission system according to claim 1, characterized in that: The coupling comprises two steel claws and a plum blossom-shaped elastic body as an intermediate body connecting the two steel claws; The friction power generation sensor is installed on the coupling of the transmission system, specifically: The friction power generation sensor is embedded in the gap between the plum blossom-shaped elastomer and the steel claw, and the friction power generation sensor is installed on any surface of several extrusion surfaces of the plum blossom-shaped elastomer in a specific installation method; wherein, the specific installation method refers to an installation method that makes the amplitude of the output signal of the friction power generation sensor greater than a preset value and the difference in the output signal characteristics of the friction power generation sensor under different working conditions of the transmission system is greater than a threshold.

3. The fault diagnosis method of the transmission system according to claim 2, characterized in that: The friction power generation sensor consists of a three-layer structure of a friction power generation sheet, a conductive copper foil and a PI insulating tape. The friction power generation sensor is used to collect the vibration signal between the plum blossom-shaped elastomer and the steel claw, and convert the vibration signal into a voltage signal.

4. The fault diagnosis method of the transmission system according to claim 1, characterized in that: The collecting of the output signal of the friction power generation sensor of the transmission system under different working conditions specifically includes: Collecting the output signal of the friction power generation sensor when the transmission system is in a healthy state and using it as a healthy signal; Collecting the output signal of the friction power generation sensor under the simulated fault state of the transmission system and using it as a fault signal; The simulated fault conditions include belt drive slippage, drive shaft misalignment and gear wear.

5. The fault diagnosis method of the transmission system according to claim 4, characterized in that: The feature extraction and calibration of the output signal to obtain a data set specifically includes: Performing a fast Fourier transform on the output signal to obtain a frequency domain signal corresponding to the output signal; Respectively extracting a time domain eigenvalue of the output signal and a frequency domain eigenvalue of a frequency domain signal corresponding to the output signal, wherein both the time domain eigenvalue and the frequency domain eigenvalue include a standard deviation, a root mean square, a waveform factor, and a peak-to-peak value; Marking the healthy signal and the fault signal in the output signal to obtain the label of the output signal; A data set is constructed based on the feature values ​​and labels of the output signals.

6. The fault diagnosis method of a transmission system according to claim 1 or 5, characterized in that: The method of using the data set to train and test a support vector machine to obtain a fault diagnosis model specifically includes: Dividing the data set into a training set and a test set; Based on the improved particle swarm algorithm, determining the optimal penalty factor and the optimal kernel function parameter of the support vector machine; According to the optimal penalty factor and the optimal kernel function parameter, the support vector machine is trained using the training set to obtain a classification model of the support vector machine; The classification model of the support vector machine is tested using the test set to verify the performance of the classification model of the support vector machine, and the verified classification model of the support vector machine is used as a fault diagnosis model for transmission system fault diagnosis.

7. A transmission system fault diagnosis system, used to implement the steps in the transmission system fault diagnosis method according to any one of claims 1 to 6, characterized in that: The transmission system fault diagnosis system comprises a transmission system test bench and a friction power generation sensor; The transmission system test bench includes a base and a variable frequency motor, a speed sensor, a gear box, a torque sensor, a conveyor pulley, a conveyor belt, a magnetic powder brake and a control console arranged on the base. The motor shaft of the variable frequency motor is connected to the input shaft of the gear box through a coupling. The coupling includes two steel claws and a plum blossom-shaped elastomer as an intermediate connecting the two steel claws. The friction power generation sensor is embedded in the gap between the plum blossom-shaped elastomer and the steel claws. The friction power generation sensor is installed on any surface of several extrusion surfaces of the plum blossom-shaped elastomer in a specific installation manner.

8. The fault diagnosis system of the transmission system according to claim 7, characterized in that: The friction power generation sensor consists of a three-layer structure of a friction power generation sheet, a conductive copper foil and a PI insulating tape. The friction power generation sensor is used to collect the vibration signal between the plum blossom-shaped elastomer and the steel claw, and convert the vibration signal into a voltage signal.

9. The fault diagnosis system of the transmission system according to claim 7 or 8, characterized in that: The fault diagnosis system of the transmission system also includes a conductive slip ring, an oscilloscope and a host computer. The conductive slip ring is installed on the motor shaft where the coupling is located, the friction power generation sensor is connected to one end of the conductive slip ring, the other end of the conductive slip ring is connected to one end of the oscilloscope, and the other end of the oscilloscope is connected to the host computer.

10. The fault diagnosis system of the transmission system according to claim 9, characterized in that: The conductive slip ring is used to derive the output signal of the friction power generation sensor, the oscilloscope is used to collect the output signal, the host computer uses MATLAB software to extract features and calibrate the output signal to obtain a data set, and the host computer also uses a support vector machine algorithm to classify and identify the data set and output the fault type of the transmission system.

Images