Cross-working-condition bearing residual life prediction method based on multi-cycle feature alignment

Abstract

The invention discloses a cross-working-condition bearing residual life prediction method based on multi-cycle feature alignment, and the method comprises the steps: firstly dividing a full-life-cycle vibration signal of a rolling bearing into signal samples with the same length, selecting a bearing signal under any working condition Wi as a source domain, and selecting a bearing signal under other working conditions as a target domain; secondly, periodically detecting each sample of a source domain and a target domain by adopting an autocorrelation function, clustering signal sample periods by using a clustering algorithm, and obtaining K period lengths to remodel the signal samples; and then constructing a convolutional neural network, and carrying out feature extraction and feature fusion on the remolded signal sample to obtain a multi-scale periodic degradation feature. And finally, constructing a regression network, predicting the remaining service life by using the multi-scale periodic degradation characteristics, constructing a correlation alignment loss function based on the multi-periodic characteristics, and carrying out training and testing. The method improves the extraction of the degradation characteristics of the bearing, and accurately predicts the residual life of the bearing.

Description

Technical Field

[0001] The present invention belongs to the field of rotating machinery health management, and in particular relates to a method for predicting the remaining life of a cross-operating-condition bearing based on multi-cycle feature alignment. Background Art

[0002] Rotating machinery is widely used in industrial manufacturing and often needs to operate for a long time under high intensity, high speed and variable working conditions, which accelerates the wear and aging of the equipment, thereby increasing the risk of failure. With the development of artificial intelligence technology, health management technology based on big data and artificial intelligence has gradually received attention, among which health management (Prognostics and Health Management, PHM) has gradually become a mainstream intelligent maintenance mode. PHM dynamically evaluates the health status of equipment by real-time monitoring and analysis of equipment operation data, and comprehensively uses advanced algorithms such as sensor technology, big data analysis and machine learning to predict potential failures and take targeted maintenance measures. The application of PHM is becoming more and more mature, which not only improves production efficiency and extends the service life of equipment, but also provides solid support for the transformation and upgrading of enterprises in the era of intelligent manufacturing. Especially for key mechanical equipment, real-time monitoring and prediction of its remaining useful life (Remaining Useful Life, RUL) is of great significance to ensure the safe and reliable operation of the equipment.

[0003] In recent years, domestic and foreign scholars have conducted extensive research on the life prediction of rolling bearings. The main methods are divided into two categories: physical model-driven and data-driven. Traditional physical models rely on precise mathematical or physical formulas. Although they can provide theoretically accurate predictions, they are difficult to apply to complex nonlinear systems and require high professional knowledge from researchers, which limits their widespread application. With the advent of the big data era, the advancement of sensor technology and data processing capabilities has promoted the rapid development of data-driven methods. This type of method reduces the dependence on professional knowledge by mining the potential information in the monitoring data and has stronger adaptability and portability. In particular, the application of deep learning technology has significantly improved the accuracy of life prediction and reduced costs by constructing deep neural networks to automatically extract fault features, which has promoted the development of rotating machinery life prediction towards intelligence and reliability. However, in actual industrial applications, the diversity of working conditions and the complexity of the operating environment make the data of the entire life cycle of bearings present different distribution characteristics. Factors such as working conditions, loads, and speed affect the consistency of data distribution, making it difficult for deep learning models to generalize between training sets and test sets, thereby reducing the performance of rolling bearing remaining service life prediction based on deep learning. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention proposes a cross-operating condition bearing remaining life prediction method based on multi-cycle feature alignment. This method introduces a multi-cycle-based convolutional neural network feature fusion technology to effectively extract multi-level fault features in the signal. At the same time, a multi-cycle feature correlation alignment strategy is introduced to align the covariance matrices of the source domain and target domain features at each cycle feature and fusion feature level, thereby significantly reducing the inconsistency of feature distribution under different operating conditions. This method not only enhances the generalization ability of the model, but also improves the prediction accuracy and reliability under variable operating conditions.

[0005] The technical solution adopted by the present invention to achieve the above-mentioned purpose is:

[0006] A method for predicting the remaining life of a bearing across working conditions based on multi-cycle feature alignment comprises the following steps:

[0007] S1. Divide the vibration signal of the rolling bearing throughout its life cycle into signal samples of the same length based on the sampling frequency. At the same time, according to different working conditions, W={W1,W2,…,W p}, where l is the number of working conditions, select any working condition W i The bearing signal under the condition W is taken as the source domain. j , and the bearing signal under j≠i is taken as the target domain.

[0008] S2. Use the autocorrelation function (ACF) to perform periodic detection on each sample in the source domain and the target domain to identify the signal sample period.

[0009] S3. Use the K-Means algorithm to cluster the signal sample periods identified in the previous step, obtain the K most significant period lengths by finding the cluster centers, and reshape the signal samples based on these significant signal sample periods to prepare for feature extraction by the convolutional neural network.

[0010] S4. Convolutional neural network (CNN) is constructed, including convolution layer, pooling layer and nonlinear activation function, to extract features of reshaped signal samples under different significant periods, and generate multi-scale period degradation features. At the same time, the average pooling layer is used to align the multi-scale period features and perform feature fusion to obtain multi-scale period degradation features.

[0011] S5. Construct a regression network and use multi-scale periodic degradation features to predict the remaining service life, completing the transformation process from feature extraction to prediction of the remaining service life of rolling bearings.

[0012] S6. Based on the multi-scale periodic degradation features extracted by convolutional neural networks at different periods, a multi-period feature correlation alignment loss function is constructed between the source domain and the target domain. By minimizing the loss function, the feature distribution of the source domain and the target domain is aligned, reducing the inconsistency of feature distribution under different working conditions.

[0013] S7. Train the model constructed in S4 and S5 on the training set using the loss function constructed in S6. Test the trained model on the test set to obtain the predicted life of the bearing.

[0014] Furthermore, the step S1 divides the horizontal vibration signal of the rolling bearing throughout its life cycle into samples of equal length and divides them into a source domain and a target domain according to different working conditions. Specifically, the source domain signal sample is The target domain signal sample is Where n is the number of source domain signal samples, and m is the number of target domain signal samples.

[0015] Furthermore, the step S2 performs periodic detection on each sample in the source domain and the target domain. First, the autocorrelation function is used to quantify the similarity of the signal at different lag times, and the formula is as follows:

[0016]

[0017] Among them, R(τ) represents the similarity of the signal x(t) at the lag time τ, τ represents the lag time, that is, the offset of the time series relative to itself, and N represents the length of the time series. Then, based on the significant peaks in the autocorrelation function, the period T = {T1, T2, …, T n+m}.

[0018] Furthermore, the step S3 obtains the significant period of the signal sample and performs signal reshaping. First, the K-means algorithm is used to reshape the sample period T = {T1, T2, ..., T n+m} to cluster and obtain K cluster centers C={C1,C2,…,C K} as a significant cycle. Then, the source domain sample S S Based on the significant cycles C={C1,C2,…,C K} is reshaped according to the following formula, for example, for any significant period C k,k＝1,2,,K , converting the source domain samples from length 1×l to length

[0019]

[0020] Similarly, the target domain samples It is also reshaped according to the cycle of significance.

[0021] Furthermore, the step S4 constructs a convolutional neural network (CNN) to extract degradation features of sample signals in a multi-cycle situation. The samples of K cycles reshaped in the source domain and the target domain are respectively subjected to feature extraction using the constructed convolutional neural network to obtain features {f1, f2, …, f k}, then the average pooling layer is used to align the feature sizes and vertically splice them, and the convolutional layer is used for feature fusion.

[0022] Furthermore, in step S5, a regression network is constructed using the fused multi-scale periodic degradation features to complete the transformation from feature extraction to life prediction. The constructed regression network includes a fully connected layer with 100 hidden nodes, a Dropout layer, and an output layer.

[0023] Furthermore, step S6 constructs a multi-cycle feature correlation alignment loss function between the source domain and the target domain. Specifically, the CORAL loss function achieves feature alignment by minimizing the difference in the covariance matrix between the source domain and the target domain, and its expression is as follows:

[0024]

[0025] where f s represents the learned features of the source domain, f s represents the learned features of the target domain, ‖·‖ F represents the Frobenius norm, which is used to calculate the sum of squares of matrix elements. d is the feature dimension. Dividing by 4d2 is to normalize the loss value to ensure that the loss magnitude is consistent under different feature dimensions. Based on the CORAL loss function, a loss function is constructed based on multi-period features. Its expression is as follows:

[0026]

[0027] Where MSE pred is the minimum mean square error, which is used to measure the predicted lifespan The difference between the actual lifespan value y is expressed as follows:

[0028]

[0029] is the source domain feature after periodic feature fusion and target domain features The loss between them is expressed as:

[0030]

[0031] α is the parameter that controls the contribution of this part of the loss;

[0032] For each period feature {f1,f2,…,f k The loss between the source domain features and the target domain features under} is expressed as follows:

[0033]

[0034] Among them, i=1,2,…,K, {λ1,λ2,…,λ K} is the coefficient for adjusting the weight of each loss item.

[0035] Furthermore, in step S7, a training set is constructed using the full life cycle data of multiple bearings under one working condition as the source domain and the full life cycle data of a bearing under another working condition as the target domain. The models constructed in S4 and S5 are trained on the training set using the loss function constructed in S6, wherein Adam is used as the most optimized method. The trained model is tested on multiple bearings under another working condition to obtain their predicted life.

[0036] The beneficial effects of the present invention are:

[0037] The present invention analyzes the periodicity of bearing vibration signals based on autocorrelation functions, extracts significant period lengths by combining K-Means clustering, and constructs multi-period extraction and fusion based on convolutional neural networks, thereby improving the extraction of bearing degradation features.

[0038] The present invention constructs a multi-period feature correlation alignment loss function, which aligns the feature distributions of the source domain and the target domain at different period scales, reduces the distribution differences between different working conditions, and improves the generalization ability of the model across working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of the method of the present invention;

[0040] Figure 2 It is a schematic diagram of the multi-scale periodic feature fusion method based on convolutional neural network of the present invention;

[0041] Figure 3 The invention provides a framework for a method for predicting the remaining life of a bearing across working conditions based on multi-cycle feature alignment. DETAILED DESCRIPTION

[0042] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways different from those described herein, and therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0043] like Figure 1 As shown, the present invention provides a cross-operating condition bearing remaining life prediction method based on multi-cycle feature alignment. First, the vibration signal of the rolling bearing throughout its life cycle is divided into equal-length samples based on the sampling frequency, and is divided into a source domain and a target domain according to different operating conditions. Next, the periodicity of the sample signal is analyzed using the autocorrelation function, and the K-Means algorithm is used to obtain the significant period length. Then, the signal samples are reshaped based on these significant periods to prepare for subsequent feature extraction. Subsequently, a convolutional neural network is constructed to extract features from the reshaped signals under multiple periods, and these features are aligned and fused through an average pooling layer. Finally, a regression network is constructed using the fused multi-scale periodic degradation features to achieve the prediction of the remaining service life of the rolling bearing. In order to ensure the consistency of feature distribution under different operating conditions, the present invention constructs a multi-cycle feature correlation alignment loss function between the source domain and the target domain, and trains and tests the model based on the loss function.

[0044] In order to facilitate understanding of the above technical solution of the present invention, the above technical solution of the present invention is described in detail below through the XJTU-SY rolling bearing full life cycle data set.

[0045] Example 1

[0046] A method for predicting the remaining life of a cross-operating bearing based on multi-cycle feature alignment includes the following steps:

[0047] S1. Example 1 uses the XJTU-SY data set for experimental verification. Five rolling bearings of model LDK UER204 are experimented under three different operating conditions. Acceleration sensors are placed in the vertical and horizontal directions of the bearing rotation plane to collect the vibration acceleration signals of the rolling bearing throughout its life cycle.

[0048] When the experimental platform is running, the sampling frequency is 25.6kHz, the sensor collects data every 1 minute, each sampling lasts 1.28 seconds, and 32768 data are obtained. In order to reduce redundancy and reduce computational complexity, the 32768 data collected each time are evenly sampled to 4096 data points. Taking Bearing1_1 as an example, a total of 4030464 data points are collected, and 123 signal samples are obtained after division by degradation unit time, and each sample length is 4096 points. The bearing data set under different working conditions is divided into source domain and target domain, as shown in Table 1.

[0049] Table 1 XJTU-SY cross-operating condition remaining service life prediction tasks

[0050]

[0051] Taking task T1 as an example, the source domains Bearing2-2 and Bearing2-3 have a total of 694 samples. 161 samples were obtained in the target domain Bearing1-2

[0052] S2. Use the autocorrelation function shown in formula (1) to perform periodicity detection on each sample in the source domain and the target domain. Taking task T1 as an example, the period T of each sample in the source domain and the target domain is T = {T1, T2, ..., T 855}.

[0053] S3, use K-means algorithm to analyze the sample period T of source domain and target domain = {T1, T2, …, T 855} for clustering, and obtain K = 3 cluster centers C = {15,216,91} as significant cycles. Then, based on the significant cycles, the source domain samples and the target domain samples are reshaped according to the following formula:

[0054]

[0055] S4 Figure 2 As shown, a three-layer convolutional neural network is constructed to extract the degradation features of the reshaped sample signals respectively. First, the specific structure of the constructed three-layer convolutional neural network is as follows:

[0056] The first layer includes a convolution layer, a maximum pooling layer, and a nonlinear activation function. The number of input channels is 1, the number of output channels is 16, the convolution kernel size is 2*2, the pooling window size is 2*2, and the activation function is ReLU;

[0057] The second layer includes a convolution layer, a maximum pooling layer, and a nonlinear activation function. The number of input channels is 16, the number of output channels is 32, the convolution kernel size is 2*2, the pooling window size is 2*2, and the activation function is ReLU;

[0058] The third layer includes a convolution layer, a maximum pooling layer, and a nonlinear activation function. The number of input channels is 32, the number of output channels is 64, the convolution kernel size is 2*2, the pooling window size is 2*2, and the activation function is ReLU.

[0059] The samples of the three cycles of source domain and target domain reshaping are respectively extracted using the constructed convolutional neural network to obtain the features {f1, f2, f3}, and then the average pooling layer with a window size of 4*9 is used to align the feature sizes and vertically splice them, and the convolution layer with a convolution kernel size of 2*2 is used for feature fusion.

[0060] S5. A regression network is constructed using the fused multi-cycle degradation features to complete the transformation from feature extraction to life prediction. The constructed regression network includes a fully connected layer with 100 hidden nodes, a Dropout layer, and an output layer.

[0061] S6. Based on the CORAL loss function, a multi-cycle feature correlation alignment loss function is constructed between the source domain and the target domain. Its expression is as follows:

[0062]

[0063] Among them, the weight of the fusion feature is α∈(0,1), and the weight of the feature in each cycle is MSE pred is the minimum mean square error, measuring the predicted life span The difference between the actual lifespan value y, Loss CORALfusion is the source domain feature after periodic feature fusion and target domain features The CORAL loss between For each period feature {f1,f2,…,f k}CORAL loss between source domain features and target domain features under .

[0064] S7, taking task T1 as an example, bearings Bearing2-2 and Bearing2-3 in one working condition are used as source domains, and bearings Bearing1-2 in another working condition are used as target domains to construct a training set, such as Figure 3 As shown in the figure, the source domain and the target domain are trained, and the CORAL loss function is used to align the source domain and target domain features. Finally, the trained model is tested on the Bearing 1-3 and Bearing 1-5 bearings to obtain their predicted life curves.

[0065] As shown in Table 2, the remaining service life prediction results of 6 test bearings in the XJTU-SY cross-operating condition remaining service life prediction task are shown. The root mean square error (RMSE) and mean absolute error (MAE) are used as evaluation indicators to compare the method proposed in the present invention with the convolutional neural network life prediction method without domain adaptation. The results in the table show that after feature alignment, the RMSE and MAE of the remaining service life prediction of all test bearings are reduced, indicating that the method of the present invention has excellent effects in reducing the feature differences between the source domain and the target domain and improving the prediction performance of the target domain. The effectiveness of this method in bearing life prediction is verified, which helps to enhance the generalization ability of the model under different working conditions.

[0066] Table 2 Prediction error of remaining service life of XJTU-SY test bearings

[0067]

[0068] At the same time, the proposed method is compared with the classic feature alignment methods MECA, MMD and CMMD. The prediction error comparison on the test bearings 1-5 is shown in Table 3. As can be seen from Table 3, the proposed method shows lower values ​​in the error indicators RMSE and MAE, which is better than feature alignment methods such as MECA, MMD and CMMD.

[0069] Table 3 Comparison results of different feature alignment methods for XJTU-SY

[0070]

Claims

1. A method for predicting the remaining life of a cross-operating bearing based on multi-cycle feature alignment, characterized in that: The following steps are involved: S1. Divide the vibration signal of the rolling bearing throughout its life cycle into signal samples of the same length, and select any working condition W i The bearing signal under the condition W is taken as the source domain. j , and the bearing signal under j≠i is taken as the target domain; S2, using the autocorrelation function to perform periodic detection on each sample in the source domain and the target domain to identify the signal sample period; S3. Use K-Means algorithm to cluster the signal sample period, obtain the K most significant period lengths by finding the cluster center, and reshape the signal sample based on the period length; S4. Construct a convolutional neural network to extract and fuse features of the reshaped signal samples with different period lengths to obtain multi-scale period degradation features; S5. Construct a regression network and use multi-scale periodic degradation features to predict the remaining service life; S6. Based on the multi-scale periodic degradation features extracted by convolutional neural networks at different periods, a multi-period feature correlation alignment loss function is constructed between the source domain and the target domain for training, and the predicted life of the bearing is obtained by testing on the test set.

2. The method for predicting the remaining life of a bearing across working conditions based on multi-cycle feature alignment according to claim 1 is characterized in that: The step S1 is specifically implemented as follows: based on the sampling frequency, the vibration signal of the rolling bearing throughout its life cycle is divided into signal samples of the same length, and at the same time, according to different working conditions W = {W1, W2, ..., W p },, where l is the number of working conditions, select any working condition W i The bearing signal under the condition W is taken as the source domain. j The bearing signal under j≠i is taken as the target domain; the source domain signal sample is obtained as The target domain signal sample is Where n is the number of source domain signal samples, and m is the number of target domain signal samples.

3. The method for predicting the remaining life of a cross-operating bearing based on multi-cycle feature alignment according to claim 2 is characterized in that: The specific process of performing periodic detection in step S2 is as follows: First, the autocorrelation function (ACF) is used to quantify the similarity of signals at different lag times, and its formula is as follows: Among them, R(τ) represents the similarity of the signal x(t) at the lag time τ, τ represents the lag time, that is, the offset of the time series relative to itself, and N represents the length of the time series; Then, based on the significant peaks in the autocorrelation function, the period T = {T1, T2, …, T n+m }.

4. The method for predicting the remaining life of a bearing across operating conditions based on multi-cycle feature alignment according to claim 3 is characterized in that: The specific implementation process of step S3 is as follows: First, the K-means algorithm is used to perform a multi-sample analysis on the source and target domains. n+m } to cluster and obtain K cluster centers C={C1,C2,…,C K } as a significant cycle; Then, the source domain sample S S Based on the significant cycles C={C1,C2,…,C K }Reshape according to the following formula: Reshape Similarly, the target domain samples It is also reshaped according to the cycle of significance.

5. The method for predicting the remaining life of a bearing across operating conditions based on multi-cycle feature alignment according to claim 4 is characterized in that: The specific process of step S4 is: the samples of K cycles reshaped in the source domain and the target domain are respectively subjected to feature extraction using the constructed convolutional neural network to obtain features {f1, f2, ..., f k }, then the average pooling layer is used to align the feature sizes and vertically splice them, and the convolutional layer is used for feature fusion.

6. The method for predicting the remaining life of a bearing across operating conditions based on multi-cycle feature alignment according to claim 5 is characterized in that: The regression network constructed in step S5 includes a fully connected layer, a Dropout layer and an output layer that are constructed sequentially.

7. The method for predicting the remaining life of a bearing across operating conditions based on multi-cycle feature alignment according to claim 6 is characterized in that: The specific process of constructing the multi-cycle feature correlation alignment loss function is as follows: The CORAL loss function achieves feature alignment by minimizing the difference in the covariance matrix between the source domain and the target domain. Its expression is as follows: where f s represents the learned features of the source domain, f s represents the learned features of the target domain, ||·|| F represents the Frobenius norm, d is the feature dimension, and based on the CORAL loss function, a loss function is constructed based on multi-period features, and its expression is as follows: Where MSE pred is the minimum mean square error, measuring the predicted life span The difference between the actual life span value y, is the source domain feature after periodic feature fusion and target domain features The CORAL loss between For each period feature {f1,f2,…,f k }, α is the parameter controlling the contribution of this part of the loss; {λ1,λ2,…,λ K } is the coefficient for adjusting the weight of each loss item.

Images