A method for extracting motor fault features and diagnosing faults

Abstract

The present invention discloses a method for extracting and diagnosing motor fault features. The method comprises the following steps: Step 1: Acquire an MCAVFCNN training dataset, where motor fault types include two major categories, mechanical faults and electrical faults, with a total of nine subcategories; Step 2: Build an MCAVFCNN model, which includes a dynamic random erasure layer, an adaptive weighted acoustic vibration fusion layer, and an adaptive pooling layer. The model is trained using the dataset acquired in Step 1 until convergence; Step 3: Apply the trained MCAVFCNN model online. The method described in the present invention can accurately monitor the status and detect faults of a running motor in real time, ensuring safe motor operation while reducing unnecessary losses.

Description

Technical Field

[0001] The present invention relates to the field of motor state monitoring and fault identification, and in particular to a method for extracting motor fault features and diagnosing faults. Background Art

[0002] In the modern industrial field, motors, as key power equipment, are widely used in various production links, from the production lines of large factories to daily household appliances. The reliability and stability of their operation are directly related to the normal operation of the entire system.

[0003] With the continuous advancement of industrial automation, motors are facing increasingly complex and diverse operating environments, impacted by varying loads, temperatures, humidity, and power grid fluctuations. Traditional motor maintenance relies on regular inspections, but this approach has significant drawbacks. It consumes significant manpower and resources, and it is difficult to accurately predict motor failures, which can easily lead to over- or under-maintenance.

[0004] The rapid development of sensor technology, signal processing techniques, and artificial intelligence algorithms in recent years has brought new opportunities for motor condition monitoring and fault diagnosis. An increasing number of high-precision sensors, such as vibration sensors, temperature sensors, and current sensors, are being used to collect motor operating parameters, enabling real-time acquisition of multi-dimensional motor data. At the same time, fault diagnosis methods based on big data analysis and machine learning are gaining popularity. By training large amounts of historical data and real-time monitoring data, intelligent diagnostic models are constructed, which can more accurately identify the type and severity of motor faults.

[0005] Effective motor condition monitoring and fault diagnosis can proactively identify potential faults and schedule repairs, avoiding production interruptions caused by unexpected failures and reducing the significant economic losses associated with downtime. For motors involved in critical production processes, such as those in the chemical and power industries, timely fault diagnosis can even be crucial for both production and environmental safety. Furthermore, accurate condition monitoring helps optimize motor operation and management, improve energy efficiency, and extend motor life. This aligns with the modern industrial philosophy of high efficiency, energy conservation, and sustainable development, and is an essential component in promoting the intelligent upgrading of industry.

[0006] Through research on relevant literature and analysis of the current research status, the following two major problems exist in motor status monitoring and diagnosis: 1. The trained model is prone to overfitting and has poor generalization performance in actual application. This is mainly because the characteristics of the training data are relatively obvious, and no disturbance is added during model training, resulting in poor generalization of the final model; 2. The trained model is difficult to comprehensively diagnose both mechanical and electrical faults of the motor. This is mainly limited by the training data category. Mechanical faults have obvious characteristics in vibration data, while electrical faults have obvious characteristics in sound signals. Most of the current training data are single vibration signals and sound signals. How to fuse vibration signals and sound signals poses certain challenges. Summary of the Invention

[0007] In order to solve the above problems, the present invention proposes a motor fault feature extraction and fault diagnosis method based on the Convolutional Neural Network (CNN). In response to problem 1, this patent proposes a dynamic random erasure algorithm, which adopts a random idea and pays attention to signals with obvious features. By designing calculation formulas and reasonably formulating algorithm thresholds, the dynamic erasure of acoustic vibration signals is achieved, which brings randomness to the training of the model; at the same time, an adaptive POOLING layer is proposed, which enables the model to adaptively extract features for data with different features, and can also flexibly handle the diversity of data in online application scenarios, thereby improving the generalization of the model; in response to problem 2, this patent proposes an adaptive weighted acoustic vibration fusion algorithm, which introduces information entropy to calculate the weight coefficients of different channel data under different fault types, which can well highlight the signals with obvious features and suppress interference signals, and finally complete the fusion of acoustic vibration data and improve the model's ability to diagnose various motor faults. To achieve this purpose, the present invention provides a motor fault feature extraction and fault diagnosis method, the specific steps are as follows, and it is characterized in that:

[0008] Step 1: Collect the MCAVFCNN training dataset, in which the motor fault types include two major categories: mechanical faults and electrical faults, with a total of 9 subcategories. Mechanical faults include local overheating, vibration faults, and air gap eccentricity faults; electrical faults include inter-turn short circuit, broken bar, end ring cracking, internal discharge, local insulation damage, and winding open circuit. Then, create corresponding category labels for the collected data.

[0009] Step 2: Build the MCAVFCNN model, which includes a dynamic random erasure layer, an adaptive weighted acoustic vibration fusion layer, and an adaptive pooling layer. Use the data set collected in step 1 to train the model until convergence.

[0010] Step 3: Use the trained model online. Apply the trained MCAVFCNN model online to perform real-time status monitoring and early fault identification on the running motor to reduce unnecessary losses.

[0011] As a further improvement of the present invention, in step 1, data set collection is performed, wherein the sensors of the collection system include an SSF-VIB-Z300 three-axis vibration sensor and an INV9206 high-precision ICP sound pressure sensor, wherein the three-axis vibration sensor is installed just above the motor drive end, and then a three-dimensional coordinate system is established with the circular plane where the three-axis sensor and the motor are located, the origin of the coordinate system coincides with the center of the motor rotor, the distance between the installation point of the sound pressure sensor and the coordinate origin is the motor diameter, and the circular plane where the x-axis and y-axis are located coincides with the plane where the vibration sensor is located, and then the collection system is used to collect the acoustic vibration data of the x-axis, y-axis and z-axis under different fault states of the motor and make corresponding labels;

[0012] As a further improvement of the present invention, the specific steps of offline training of the MCAVFCNN model in step 2 are:

[0013] Step 2.1. A dynamic random erasure algorithm is proposed to erase the collected sound and vibration signals to improve the robustness of the model. The erasure rule is: if the data is selected for erasure, the value will be set to 0. A set of sampled data {x1, x2, ..., x N}, where N represents the dimension of the collected data, for the i-th data x i Perform erase calculation, where x i The corresponding erasure probability P(x i )The expression is designed to be:

[0014]

[0015] Among them, r represents a random number that satisfies uniform distribution, that is, r~U[0,1], x max and xave Respectively represent the data {x1,x2,...,x N}'s maximum and average values.

[0016] Step 2.2. Use the FFT algorithm to calculate the signal processed in step 2.1. The order of the multi-channel acoustic and vibroacoustic data is: x-direction vibration signal, x-direction sound signal, y-direction vibration signal, y-direction sound signal, z-direction vibration signal, and z-direction sound signal. After obtaining the six-channel acoustic and vibroacoustic signals, use the proposed adaptive weighted acoustic and vibroacoustic fusion algorithm to fuse the acoustic and vibroacoustic signals. The fusion calculation process is as follows:

[0017] Step 2.2.1. Multi-channel acoustic vibration signal matrix (M) corresponding to the jth type fault j , which can be expressed as:

[0018]

[0019] Wherein, q represents the total number of channels, and the value of q in this patent is 6.

[0020] Step 2.2.2. Calculate the information entropy of a single channel data, where the calculation formula is:

[0021]

[0022] Where, (E l ) k represents the information entropy corresponding to the collected data of the lth channel under the kth type fault, (M il ) k Represents the i-th data point of the l-th channel of the k-th type fault.

[0023] Step 2.2.3. Considering the differences in fault feature sensitivity in the X, Y, and Z directions, a weighted weighting algorithm is proposed to weight the data of each channel to highlight the signals with obvious features as much as possible and suppress the interference signals. The specific calculation formula is as follows:

[0024]

[0025] In the formula, (w l ) k Indicates the fusion weight corresponding to the lth channel data of the kth type fault.

[0026] Step 2.2.4. Based on the weight coefficients corresponding to each channel, solve the fusion result (f) of the k-th type of fault data. k , and its calculation formula is:

[0027]

[0028] Step 2.3. Build a convolutional neural network to extract and classify features from the data in step 2.2. The specific model architecture is: convolution layer 1 - adaptive pooling layer 1 - convolution layer 2 - dropout layer - convolution layer 3 - adaptive pooling layer 2 - fully connected layer - softmax layer. The algorithm calculation formula of the proposed adaptive pooling layer is as follows:

[0029]

[0030] Where, (AC il ) krepresents the pooling value calculated based on the starting point of the i-th data point of the l-th channel of the k-th fault, ξ represents the pooling adjustment factor, t represents the t-th pooling kernel, v represents the convolution kernel length, and b represents the bias coefficient.

[0031] Among them, the calculation formula of the pooling adjustment factor ξ is:

[0032]

[0033] In the formula, η represents the characteristic coefficient, δ represents the correction coefficient, and the value range is (0,1), c and v max They respectively represent the average value calculated after removing the maximum value in the pooling domain where the convolution kernel is located and the maximum value in the pooling domain.

[0034] The expression of the characteristic coefficient η is as follows:

[0035]

[0036] where n epoch Indicates the number of rounds of network training.

[0037] Step 2.4. Use the cross entropy loss function and Adam parameter optimization algorithm to train the MCAVFCNN model proposed in this patent until the model converges, where the iteration termination condition is set to the loss function less than 0.001.

[0038] The present invention provides a motor fault feature extraction and fault diagnosis method. Based on an acoustic and vibroacoustic data acquisition system, the system collects acoustic and vibroacoustic signals of different fault categories. The model is then constructed using an MCAVFCNN model. This model incorporates a dynamic random erasure layer, an adaptive weighted acoustic and vibroacoustic fusion layer, and an adaptive pooling layer. Through continuous training iterations, the model is ultimately developed to diagnose a variety of motor faults and demonstrate excellent generalization performance. The overall beneficial effects are:

[0039] 1. This patent proposes a dynamic random erasure algorithm that uses randomization while focusing on signals with distinct features. It dynamically erases acoustic vibration signals by designing a calculation formula and rationally setting algorithm thresholds, which introduces randomness into model training. It also proposes an adaptive pooling layer, which enables the model to adaptively extract features based on data with different characteristics and flexibly handle the diversity of data in online application scenarios, thereby improving the model's generalization.

[0040] 2. The adaptive weighted acoustic-vibration fusion algorithm proposed in this patent introduces information entropy to calculate the weight coefficients of different channel data under different fault types. This can well highlight the characteristic signals and suppress the interference signals, and finally complete the fusion of acoustic-vibration data and improve the model's ability to diagnose various motor faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the sensor layout of the acoustic vibration data acquisition system in this patent;

[0042] Figure 2 is a flow chart of the present invention;

[0043] Figure 3 Schematic diagram of weighted fusion of acoustic vibration signals;

[0044] Figure 4 This is the network architecture diagram of the MCAVFCNN model in this patent. DETAILED DESCRIPTION

[0045] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0046] The present invention proposes a motor fault feature extraction and fault diagnosis method, which aims to perform real-time and accurate status monitoring and fault detection on a running motor, thereby ensuring the safe operation of the motor and reducing unnecessary losses.

[0047] Figure 1 This is a schematic diagram of the sensor layout of the acoustic vibration data acquisition system in this patent, where the sensors of the acquisition system include the SSF-VIB-Z300 three-axis vibration sensor and the INV9206 high-precision ICP sound pressure sensor. The three-axis vibration sensor is installed directly above the motor drive end, and then a three-dimensional coordinate system is established using the circular plane where the three-axis sensor and the motor are located. The origin of the coordinate system coincides with the center of the motor rotor, the distance between the installation point of the sound pressure sensor and the coordinate origin is the motor diameter, and the circular plane where the x-axis and y-axis are located coincides with the plane where the vibration sensor is located. Then, the acquisition system is used to collect the acoustic vibration data of the x-axis, y-axis and z-axis under different fault states of the motor and make corresponding labels.

[0048] Figure 2 It is a flow chart of the present invention. As can be seen from the figure, the specific implementation steps of the method are:

[0049] Step 1: Collect the MCAVFCNN training dataset, in which the motor fault types include two major categories: mechanical faults and electrical faults, with a total of 9 subcategories. Mechanical faults include local overheating, vibration faults, and air gap eccentricity faults; electrical faults include inter-turn short circuit, broken bar, end ring cracking, internal discharge, local insulation damage, and winding open circuit. Then, create corresponding category labels for the collected data.

[0050] Step 2: Build the MCAVFCNN model, which includes a dynamic random erasure layer, an adaptive weighted acoustic vibration fusion layer, and an adaptive pooling layer. Use the data set collected in step 1 to train the model until convergence.

[0051] Step 3: Use the trained model online. Apply the trained MCAVFCNN model online to perform real-time status monitoring and early fault identification on the running motor to reduce unnecessary losses.

[0052] The specific steps for offline training of the MCAVFCNN model in step 2 are:

[0053] Step 2.1. A dynamic random erasure algorithm is proposed to erase the collected sound and vibration signals to improve the robustness of the model. The erasure rule is: if the data is selected for erasure, the value will be set to 0. A set of sampled data {x1, x2, ..., x N}, where N represents the dimension of the collected data, for the i-th data x i Perform erase calculation, where x i The corresponding erasure probability P(x i )The expression is designed to be:

[0054]

[0055] Among them, r represents a random number that satisfies uniform distribution, that is, r~U[0,1], x max and xave Respectively represent the data {x1,x2,...,x N}'s maximum and average values.

[0056] Step 2.2. Use the FFT algorithm to calculate the signal processed in step 2.1. The order of the multi-channel acoustic and vibroacoustic data is: x-direction vibration signal, x-direction sound signal, y-direction vibration signal, y-direction sound signal, z-direction vibration signal, and z-direction sound signal. After obtaining the six-channel acoustic and vibroacoustic signals, use the proposed adaptive weighted acoustic and vibroacoustic fusion algorithm to fuse the acoustic and vibroacoustic signals. The fusion calculation process is as follows:

[0057] Step 2.2.1. Multi-channel acoustic vibration signal matrix (M) corresponding to the jth type fault j , which can be expressed as:

[0058]

[0059] Wherein, q represents the total number of channels, and the value of q in this patent is 6.

[0060] Step 2.2.2. Calculate the information entropy of a single channel data, where the calculation formula is:

[0061]

[0062] Where, (E l ) krepresents the information entropy corresponding to the collected data of the lth channel under the kth type fault, (M il ) k Represents the i-th data point of the l-th channel of the k-th type fault.

[0063] Step 2.2.3. Considering the differences in fault feature sensitivity in the X, Y, and Z directions, a weighted weighting algorithm is proposed to weight the data of each channel to highlight the signals with obvious features as much as possible and suppress the interference signals. The specific calculation formula is as follows:

[0064]

[0065] In the formula, (w l ) k Indicates the fusion weight corresponding to the lth channel data of the kth type fault.

[0066] Step 2.2.4. Based on the weight coefficients corresponding to each channel, solve the fusion result (f) of the k-th type of fault data. k , and its calculation formula is:

[0067]

[0068] Step 2.3. Build a convolutional neural network to extract and classify features from the data in step 2.2. The specific model architecture is: convolution layer 1 - adaptive pooling layer 1 - convolution layer 2 - dropout layer - convolution layer 3 - adaptive pooling layer 2 - fully connected layer - softmax layer. The algorithm calculation formula of the proposed adaptive pooling layer is as follows:

[0069]

[0070] Where, (AC il ) k represents the pooling value calculated based on the starting point of the i-th data point of the l-th channel of the k-th fault, ξ represents the pooling adjustment factor, t represents the t-th pooling kernel, v represents the convolution kernel length, and b represents the bias coefficient.

[0071] Among them, the calculation formula of the pooling adjustment factor ξ is:

[0072]

[0073] In the formula, η represents the characteristic coefficient, δ represents the correction coefficient, and the value range is (0,1), c and v max They respectively represent the average value calculated after removing the maximum value in the pooling domain where the convolution kernel is located and the maximum value in the pooling domain.

[0074] The expression of the characteristic coefficient η is as follows:

[0075]

[0076] where n epoch Indicates the number of rounds of network training.

[0077] Step 2.4. Use the cross entropy loss function and Adam parameter optimization algorithm to train the MCAVFCNN model proposed in this patent until the model converges, where the iteration termination condition is set to the loss function less than 0.001.

[0078] Figure 3 This is a schematic diagram of the weighted fusion of acoustic and vibration signals. It can be clearly seen that for data from different channels, the fusion weight coefficient is calculated using information entropy to finally obtain the fused data. Through this operation, the characteristics of sound and vibration signals are fused without increasing the data dimension, which greatly enhances the fault characteristics of the training data.

[0079] Figure 4 This is the architecture diagram of MCAVFCNN in this patent. As can be seen from the figure, the specific network architecture is: acoustic vibration data acquisition layer - dynamic random erasure layer - FFT data processing layer - adaptive weighted acoustic vibration fusion layer - convolution layer 1 - adaptive POOLING layer 1 - convolution layer 2 - dropout layer - convolution layer 3 - adaptive POOLING layer 2 - fully connected layer - SOFTMAX layer. It can be seen that the collected acoustic vibration data, after dynamic random erasure and FFT transformation, uses the adaptive weighted fusion algorithm to fuse features, and then uses the convolution layer and pooling layer of the convolutional neural network to extract features, and incorporates adaptive ideas in the extraction process. Finally, the SOFTMAX layer realizes the identification and classification of different motor faults. Through the adjustment of the above model architecture, the classification ability and generalization performance of the training model are improved.

Claims

1. A method for extracting and diagnosing motor fault features, characterized in that: The specific steps are as follows: Step 1: Collect the MCAVFCNN training dataset, which contains nine categories of motor faults, including two major categories: mechanical faults and electrical faults. Mechanical faults include local overheating, vibration faults, and air gap eccentricity faults; electrical faults include inter-turn short circuits, broken bars, end ring cracks, internal discharges, local insulation damage, and winding open circuits. Then, create corresponding category labels for the collected data. Step 2: Build the MCAVFCNN model, which includes a dynamic random erasure layer, an adaptive weighted acoustic vibration fusion layer, and an adaptive pooling layer. Use the data set collected in step 1 to train the model until convergence. The specific steps for offline training of the MCAVFCNN model in step 2 are: Step 2.

1. A dynamic random erasure algorithm is proposed to erase the collected sound and vibration signals to improve the robustness of the model. The erasure rule is: if the data is selected for erasure, the corresponding value of the data will be set to 0. A set of sampled data , where N represents the dimension of the collected data, for the i-th data Perform erase calculation, where Corresponding erasure probability The expression is designed to be: ； in Represents a random number that satisfies a uniform distribution, that is , and Represents data separately The maximum and average values ​​of Step 2.

2. Use the FFT algorithm to calculate the signal processed in step 2.

1. The order of the multi-channel acoustic and vibroacoustic data is: x-direction vibration signal, x-direction sound signal, y-direction vibration signal, y-direction sound signal, z-direction vibration signal, and z-direction sound signal. After obtaining the six-channel acoustic and vibroacoustic signals, use the proposed adaptive weighted acoustic and vibroacoustic fusion algorithm to fuse the acoustic and vibroacoustic signals. The fusion calculation process is as follows: Step 2.2.

1. Multi-channel acoustic vibration signal matrix corresponding to the jth type fault , which can be expressed as: ； Where q represents the total number of channels, and in this patent the value of q is 6; Step 2.2.

2. Calculate the information entropy of a single channel data, where the calculation formula is: ； Where, represents the information entropy corresponding to the collected data of the lth channel under the kth type fault, represents the i-th data point of the l-th channel of the k-th type fault; Step 2.2.

3. Considering the differences in fault feature sensitivity in the X, Y, and Z directions, a weighted weighting algorithm is proposed to weight the data of each channel to highlight the signals with obvious features as much as possible and suppress the interference signals. The specific calculation formula is as follows: ； Where, represents the fusion weight corresponding to the lth channel data of the kth type fault; Step 2.2.

4. Based on the weight coefficients corresponding to each channel, solve the fusion result of the k-th type of fault data , and its calculation formula is: ； Step 2.

3. Build a convolutional neural network to extract and classify features from the data in step 2.

2. The specific model architecture is: convolution layer 1 - adaptive pooling layer 1 - convolution layer 2 - dropout layer - convolution layer 3 - adaptive pooling layer 2 - fully connected layer - softmax layer. The algorithm calculation formula of the proposed adaptive pooling layer is as follows: ； Where, represents the pooling value calculated from the starting point of the ith data point of the lth channel of the kth type fault, represents the pooling adjustment factor, t represents the tth pooling kernel, v represents the convolution kernel length, and b represents the bias coefficient; Among them, the pooling adjustment factor The calculation formula is: ； Where, represents the characteristic coefficient, Represents the correction coefficient, the value range is (0,1), c and They represent the average value calculated after removing the maximum value in the pooling domain where the convolution kernel is located and the maximum value in the pooling domain respectively; Characteristic coefficient The expression is as follows: ； in Indicates the number of rounds of network training; Step 2.

4. Use the cross entropy loss function and Adam parameter optimization algorithm to train the MCAVFCNN model proposed in this patent until the model converges, where the iteration termination condition is set to the loss function less than 0.001; Step 3: Use the trained model online and apply the trained MCAVFCNN model online to perform real-time status monitoring and early fault identification on the running motor.

2. A motor fault feature extraction and fault diagnosis method according to claim 1, wherein the algorithm model MCAVFCNN is characterized in that: in step 1, the data set is collected, wherein the sensors of the collection system include an SSF-VIB-Z300 three-axis vibration sensor and an INV9206 high-precision ICP sound pressure sensor, wherein the three-axis vibration sensor is installed just above the motor drive end, and then a three-dimensional coordinate system is established with the circular plane where the three-axis sensor and the motor are located, the origin of the coordinate system coincides with the center of the motor rotor, the distance between the installation point of the sound pressure sensor and the coordinate origin is the motor diameter, and the circular plane where the x-axis and y-axis are located coincides with the plane where the vibration sensor is located, and then the collection system is used to collect the acoustic and vibration data of the x-axis, y-axis and z-axis under different fault states of the motor and create corresponding labels.

Images