Mechanical equipment migration fault diagnosis method and system based on source domain black box model

Abstract

The present invention provides a mechanical equipment migration fault diagnosis method and system based on a source domain black box model, relating to the technical field of mechanical fault diagnosis. The method comprises: obtaining acceleration signals of mechanical equipment under various states and preprocessing them to obtain a preprocessed data set; constructing a source domain black box model and training it to obtain a trained source domain black box model; constructing a target domain model; obtaining the KL divergence loss between the output of the source domain black box model and the output of the target domain model; obtaining the consistency loss of the target domain model using an interpolation consistency regularization method; obtaining the nuclear norm loss of the target domain model; obtaining the supervised loss of the target domain model with an adversarial regularizer; optimizing the target domain model; and performing mechanical equipment fault diagnosis based on the optimized target domain model. The invention utilizes the output of the source domain black box model to extract and transfer source domain knowledge to the target domain model without requiring source domain data or model parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical fault diagnosis, and in particular to a mechanical equipment migration fault diagnosis method and system based on a source domain black box model. Background Art

[0002] With the rapid development of modern industry, the safe and stable operation of mechanical equipment has become increasingly important. In recent years, a variety of monitoring and diagnostic technologies have emerged, promoting the identification of equipment weaknesses and the development of operational and maintenance technologies. Among these, deep learning-based intelligent diagnosis and maintenance, as well as industrial big data analysis, have seen rapid growth. Furthermore, this technology breaks away from conventional diagnostic methods based on expert knowledge, establishing a more intelligent, end-to-end, data-driven diagnostic model with broad application prospects.

[0003] Although deep learning-based intelligent diagnostic technology has made remarkable progress, training deep learning-based diagnostic models typically requires massive amounts of labeled data. However, due to the difficulty of disassembling and assembling some equipment or safety restrictions, collecting sufficient labeled data for each diagnostic scenario or task is extremely expensive or even impractical. In addition, due to the independent and identically distributed assumptions of intelligent diagnostic models, data collected at different times exhibit different probability distributions under the influence of changes in working conditions and equipment status, making it difficult to directly apply previously trained diagnostic models across domains. Therefore, unsupervised domain adaptation technology is gaining increasing attention. This technology can effectively address the problem of domain probability distribution differences and solve the diagnostic task of unlabeled data in the target domain through the diagnostic knowledge of labeled source domain data.

[0004] However, regardless of the domain adaptation strategy employed, labeled source domain data is always required. Furthermore, due to the high-frequency sampling of monitoring data, the transmission and storage costs of massive source data are high, especially for portable or edge diagnostics. To address this issue, passive adaptive diagnostic methods have been proposed, which have access to well-trained source domain models but lack labeled source domain data. Specifically, during the learning process, model parameters are open and can be retrained to adapt to the target domain. Although this approach has achieved satisfactory results, it still carries the risk of exposing the details of the source domain model. Summary of the Invention

[0005] The purpose of the present invention is to provide a mechanical equipment migration fault diagnosis method and system based on a source domain black box model. The method only requires the output of the source domain black box model to extract and transfer the source domain knowledge to the target domain model without requiring source domain data and model parameters. At the same time, the method prevents the model from falling into local optimality and obtains a more accurate target domain model.

[0006] A mechanical equipment migration fault diagnosis method based on a source domain black box model, comprising:

[0007] S1, acquiring acceleration signals generated by stiffness fluctuations caused by the movement and contact of mechanical equipment parts in various states, preprocessing the acceleration signals to obtain a preprocessed dataset, and dividing the preprocessed dataset into a source domain dataset and a target domain dataset;

[0008] S2, constructing a source domain black box model, and training the source domain black box model based on the source domain dataset and combining a smoothed cross entropy loss function to obtain a trained source domain black box model;

[0009] S3, inputting any source domain data in the source domain data set into the trained source domain black box model to obtain an output acceleration signal of the source domain black box model;

[0010] S4, build the target domain model;

[0011] S5, inputting any target domain data in the target domain data set into the target domain model to obtain an output acceleration signal of the target domain model;

[0012] S6, obtaining the KL divergence loss between the acceleration signal output by the source domain black box model and the acceleration signal output by the target domain model; the KL divergence loss expression is:

[0013]

[0014] Where:

[0015] D kl (p s ||p t ) represents the KL divergence loss, p s Represents the acceleration signal output by the black box model in the source domain, p t Indicates the output acceleration signal of the target domain model, D kl (b s ||b t ) represents the KL divergence of the c-th category prediction probability between the source domain black box model and the target domain model, represents the KL divergence of the predicted probabilities of the source domain black box model and the target domain model excluding non-c classes, is the predicted probability of the cth class of the source domain black box model, represents the prediction probability of the source domain black box model excluding non-c classes, is the predicted probability of the c-th sample of the target domain model, represents the predicted probability of the target domain model excluding non-c classes, K represents the total number of fault categories, represents the predicted probability of the i-th category of the source domain black box model, represents the predicted probability of the target domain model class i, λ is the first hyperparameter, and class c is the fault class with the highest predicted probability;

[0016] S7, based on the target domain dataset, using an interpolation consistency regularization method to obtain the consistency loss of the target domain model;

[0017] The consistency loss expression is:

[0018]

[0019] Where: X t represents the target domain dataset, L mix represents the consistency loss, represents the qth target area acceleration signal data in the target area dataset, represents the j-th target area acceleration signal data in the target area dataset, The input of the target domain model is The output when The input of the target domain model is Output when Mix β (a, b) = β·a+(1-β)·b, β∈Beat(δ,δ) means β is sampled from the Beta distribution, δ is the parameter value,

[0020] for and The KL divergence between

[0021] S8, obtaining a nuclear norm loss of the target domain model based on the acceleration signal output by the target domain model;

[0022] S9, obtaining a supervised loss with an adversarial regularizer of the target domain model based on the acceleration signal output by the target domain model;

[0023] S10, optimizing the target domain model based on the KL divergence loss, the consistency loss, the nuclear norm loss, and the supervised loss with the adversarial regularizer;

[0024] S11, repeat S5-S10 to obtain the final target domain model;

[0025] S12: Based on the final target domain model, the real-time acceleration signal of the mechanical equipment is input into the final target domain model to perform fault diagnosis on the mechanical equipment.

[0026] Optionally, the smoothed cross entropy loss function expression is:

[0027]

[0028] Where: is the source domain dataset, is the label corresponding to the source domain dataset, represents the source domain black box model, represents the output of the source domain black box model when the input is x, represents the predicted probability of the kth class of the source domain black box model, is a smooth label, and When y=k, q k =1, otherwise q k =0, α is the smoothing parameter, and u(k)=1 / K is the uniform distribution of labels.

[0029] Optionally, the nuclear norm loss expression is:

[0030]

[0031] Where: L nm is the nuclear norm loss, B is the length of the qth target domain data in the target domain dataset, || || * represents the nuclear norm.

[0032] Optionally, the supervised loss expression with the adversarial regularizer is:

[0033]

[0034] Where: L ce is a supervised loss function with an adversarial regularizer; y e For easy sample X e Pseudo labels of θ t are the parameters of the target domain model, θ Ω is the parameter of Ω, Ω is the adversarial domain adaptation discriminator, For easy sample sets, is a difficult sample set, L sa is the supervised loss with adversarial regularizer, f t is the target domain model, f t (X h ) indicates that the input of the target domain model is X h The output when f t (X e ) indicates that the input of the target domain model is X e Output when .

[0035] Optionally, obtaining the self-entropy of each target domain data in the target domain data set;

[0036] The average value of each self-entropy is obtained, and the target domain data corresponding to the self-entropy smaller than the average value is used as the easy sample set, and the target domain data corresponding to the self-entropy greater than or equal to the average value is used as the difficult sample set.

[0037] Optionally, the expression of the self-entropy is:

[0038]

[0039] Where: express The self-entropy, Indicates that the target domain model number input is The predicted probability of the kth class when .

[0040] The present invention also provides a mechanical equipment migration fault diagnosis system based on a source domain black box model, which includes:

[0041] A data acquisition module is used to acquire acceleration signals of mechanical equipment in various states, preprocess the acceleration signals to obtain a preprocessed data set, and divide the preprocessed data set into a source domain data set and a target domain data set;

[0042] A first model building module is used to build a source domain black box model, and train the source domain black box model based on the source domain dataset in combination with a smoothed cross entropy loss function to obtain the trained source domain black box model;

[0043] A model training module, configured to input any source domain data in the source domain dataset into the trained source domain black box model to obtain a source domain black box model output;

[0044] The second model building module is used to build a target domain model;

[0045] A data input module, configured to input any target domain data in the target domain data set into the target domain model to obtain a target domain model output;

[0046] The KL divergence loss module is used to obtain the KL divergence loss between the output of the source domain black box model and the output of the target domain model. The KL divergence loss expression is:

[0047]

[0048] Where: D kl (p s ||p t) represents the KL divergence loss, p s represents the output of the source domain black box model, p t Denotes the target domain model output, D kl (b s ||b t ) represents the KL divergence of the c-th category prediction probability between the source domain black box model and the target domain model, It represents the KL divergence of the predicted probabilities of the source domain black box model and the target domain model excluding the c-th class. is the predicted probability of the cth class of the source domain black box model, represents the predicted probability of other classes excluding class c by the source domain black box model, is the predicted probability of the c-th sample of the target domain model, represents the prediction probability of other classes excluding class c in the target domain model, K represents the total number of fault categories, represents the predicted probability of the i-th category of the source domain black box model, represents the prediction probability of the target domain model for the i-th category, λ is the first hyperparameter, and category c is the fault category with the highest probability of accurate prediction;

[0049] A consistency loss module, configured to obtain the consistency loss of the target domain model using an interpolation consistency regularization method based on the target domain dataset;

[0050] The consistency loss expression is:

[0051]

[0052] Where: X t represents the target domain dataset, L mix represents the consistency loss, represents the qth target domain data in the target domain dataset, represents the jth target domain data in the target domain dataset, The input of the target domain model is The output when The input of the target domain model is Output when Mix β (a, b) = β·a+(1-β)·b, β∈Beat(δ,δ) means β is sampled from the Beta distribution, δ is the parameter value, for and The KL divergence between

[0053] A nuclear norm loss module, configured to obtain a nuclear norm loss of the target domain model based on the target domain model output;

[0054] A supervised loss with an adversarial regularizer, for obtaining a supervised loss with an adversarial regularizer of the target domain model based on the target domain model output;

[0055] A model optimization module, configured to optimize the target domain model based on the KL divergence loss, the consistency loss, the nuclear norm loss, and the supervised loss with the adversarial regularizer;

[0056] A repeated execution module, used for repeatedly executing the data input module to the model optimization module to obtain a final target domain model;

[0057] A fault diagnosis module is used to perform fault diagnosis on mechanical equipment based on the final target domain model.

[0058] The effects of the present invention are as follows:

[0059] The mechanical equipment migration fault diagnosis method of the present invention is based on the source domain black box model. The constructed target domain model does not require training set data, labels and parameters of the source model. It only relies on the predicted output obtained from the training model to construct the target domain model for unlabeled test data. On the one hand, it greatly ensures the privacy and security of the data. On the other hand, since there is no need to obtain the parameters of the source domain black box model, it greatly expands the applicability and convenience.

[0060] The present invention is a mechanical equipment migration fault diagnosis method based on the source domain black box model. The target domain model is optimized by obtaining the KL divergence loss of the source domain black box model output and the target domain model output. In order to avoid the local optimum of the model, the target domain model is optimized by combining consistency loss, nuclear norm loss and supervised loss with adversarial regularizer, thereby improving the diagnostic performance of the target domain model.

[0061] The present invention proposes a mechanical equipment migration fault diagnosis method based on the source domain black box model, establishes a learning method based on easy samples to guide difficult samples, and greatly improves the diagnostic performance of the target domain model. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of the mechanical equipment migration fault diagnosis method based on the source domain black box model of the present invention;

[0063] Figure 2 Schematic diagram of the structure of the source domain black box model of the present invention;

[0064] Figure 3 Schematic diagram of detailed structural parameters of the diagnostic network of the source domain black box model of the present invention;

[0065] Figure 4This is a structural diagram of a rolling bearing experimental device according to the present invention;

[0066] Figure 5 This is a structural schematic diagram of the nuclear power gearbox experimental device of the present invention.

[0067] In the figure: 1. Drive motor; 2. Coupling; 3. Planetary gearbox; 4. Parallel shaft gearbox; 5. Speed ​​and torque sensor; 6. Rolling bearing under test; 7. Acceleration sensor; 8. Magnetic powder brake.

[0068] 11. Drive motor; 12. Coupling; 13. Planetary gearbox under test; 14. Acceleration sensor; 15. Support bearing; 16. Bevel gear; 17. Gear pump. DETAILED DESCRIPTION

[0069] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0070] Figure 1 This is a flow chart of the mechanical equipment migration fault diagnosis method based on the source domain black box model of the present invention. Figure 1 As shown, the present invention provides a mechanical equipment migration fault diagnosis method based on the source domain black box model, especially for performing status detection and fault diagnosis on bearing-type mechanical equipment (such as motors, gearboxes and other rotating mechanical equipment), which specifically includes the following steps:

[0071] S1, obtains acceleration signals, i.e., vibration signals, generated by stiffness fluctuations caused by the moving contact of mechanical equipment parts in various states, and preprocesses the acceleration signals to obtain a preprocessed data set, and divides the preprocessed data set into a source domain data set and a target domain data set; specifically, the acceleration signals of the mechanical equipment in various health states are collected by an acceleration sensor, and several data sets are obtained by window sliding. For several data sets, normal standardization preprocessing is used to normalize the data amplitude to the [0,1] interval to ensure data consistency, thereby obtaining a preprocessed data set.

[0072] S2: Construct a source domain black box model. Based on the source domain dataset, the source domain black box model is trained with the smoothed cross entropy loss function to obtain a trained source domain black box model. The input data and parameters of the trained source domain black box model become unavailable. The smoothed cross entropy loss function is expressed as:

[0073]

[0074] Where: is the source domain dataset, is the label corresponding to the source domain dataset, represents the source domain black box model, represents the output of the source domain black box model when the input is x, represents the predicted probability of the kth class of the source domain black box model, is a smooth label, and When y=k, q k =1, otherwise q k =0, α is the smoothing parameter, α=0.1, u(k)=1 / K is the uniform distribution of labels.

[0075] S3, input any source domain data in the source domain dataset into the trained source domain black box model to obtain the source domain black box model output.

[0076] S4, build the target domain model.

[0077] S5, input any target domain data in the target domain dataset into the target domain model to obtain the target domain model output.

[0078] S6, obtain the KL divergence loss between the source domain black box model output and the target domain model output. The KL divergence loss expression is:

[0079]

[0080] Where: D kl (p s ||p t ) represents the KL divergence loss, p s represents the output of the source domain black box model, p t Denotes the target domain model output, D kl (b s ||b t ) represents the KL divergence of the c-th category prediction probability between the source domain black box model and the target domain model, It represents the KL divergence of the predicted probabilities of the source domain black box model and the target domain model excluding the c-th class. is the predicted probability of the cth class of the source domain black box model, represents the predicted probability of other classes excluding class c by the source domain black box model, is the predicted probability of the c-th sample of the target domain model, represents the prediction probability of other classes excluding class c in the target domain model, K represents the total number of fault categories, represents the predicted probability of the i-th category of the source domain black box model, represents the predicted probability of the target domain model for the i-th category, λ is the first hyperparameter, and category c is the fault category with the highest probability of accurate prediction.

[0081] The predicted probability of excluding the cth class from the source domain black box model output and the target domain model output is expressed as Then the predicted probability of the non-target class of class i (i≠c) is Therefore, the following expression can be obtained:

[0082]

[0083] This gives the KL divergence loss expression.

[0084] S7, based on the target domain dataset, uses the interpolation consistency regularization method to obtain the consistency loss of the target domain model. The consistency loss expression is:

[0085]

[0086] Where: X t represents the target domain dataset, L mix represents the consistency loss, represents the qth target domain data in the target domain dataset, represents the jth target domain data in the target domain dataset, The input of the target domain model is The output when The input of the target domain model is Output when Mix β (a, b) = β·a+(1-β)·b, β∈Beat(δ,δ) means β is sampled from Beta distribution, δ is the parameter value, δ = 0.3, for and The KL divergence between .

[0087] The consistency loss is implemented based on MixUp.

[0088] S8, based on the output of the target domain model, obtain the nuclear norm loss of the target domain model. The nuclear norm loss expression is:

[0089]

[0090] Where: L nm is the nuclear norm loss, B is the length of the qth target domain data in the target domain dataset, || || * represents the nuclear norm.

[0091] S9, based on the output of the target domain model, obtain the supervised loss of the target domain model with the adversarial regularizer. The supervised loss expression with the adversarial regularizer is:

[0092]

[0093] Where: Lce is a supervised loss function with an adversarial regularizer; y e For easy sample X e Pseudo labels of θ t are the parameters of the target domain model, θ Ω is the parameter of Ω, Ω is the adversarial domain adaptation discriminator, For easy sample sets, is a difficult sample set, L sa is the supervised loss with adversarial regularizer, f t is the target domain model, f t (X h ) indicates that the input of the target domain model is X h The output when f t (X e ) indicates that the input of the target domain model is X e Output when .

[0094] Preferably, for the target domain model that is prone to fall into the local optimal parameter space guided by easy samples, a difficult sample adaptive learning strategy guided by easy samples is designed, and the self-entropy of each target domain data in the target domain dataset is first obtained.

[0095] Furthermore, the average self-entropy is used as the division criterion to obtain the average value of each entropy. The target domain data corresponding to the self-entropy less than the average value is regarded as the easy sample set, and the target domain data corresponding to the self-entropy greater than or equal to the average value is regarded as the difficult sample set.

[0096] The expression of self-entropy is:

[0097]

[0098] Where: express The self-entropy, Indicates that the target domain model number input is The predicted probability of the kth class when .

[0099] By introducing the adversarial domain adaptation discriminator Ω, the distance between easy samples and difficult samples in the target sample is shortened in the high-dimensional space.

[0100] S10, optimizes the target domain model based on KL divergence loss, consistency loss, nuclear norm loss and supervised loss with adversarial regularizer.

[0101] S11, repeat S5-S10 to obtain the final target domain model.

[0102] S12, perform fault diagnosis on mechanical equipment based on the final target domain model.

[0103] Specifically, mechanical equipment such as gearboxes, motors and bearings are all suitable for fault diagnosis using the method of the present invention.

[0104] The present invention also provides a mechanical equipment migration fault diagnosis system based on a source domain black box model, which includes:

[0105] The data acquisition module is used to obtain the acceleration signals of mechanical equipment in various states, preprocess the acceleration signals to obtain a preprocessed data set, and divide the preprocessed data set into a source domain data set and a target domain data set.

[0106] The first model construction module is used to construct a source domain black box model. Based on the source domain dataset, the source domain black box model is trained in combination with the smoothed cross entropy loss function to obtain a trained source domain black box model.

[0107] The model training module is used to input any source domain data in the source domain dataset into the trained source domain black box model to obtain the source domain black box model output.

[0108] The second model building module is used to build a target domain model.

[0109] The data input module is used to input any target domain data in the target domain dataset into the target domain model to obtain the target domain model output.

[0110] The KL divergence loss module is used to obtain the KL divergence loss between the source domain black box model output and the target domain model output. The KL divergence loss expression is:

[0111]

[0112] Where: D kl (p s ||p t ) represents the KL divergence loss, p s represents the output of the source domain black box model, p t Denotes the target domain model output, D kl (b s ||b t ) represents the KL divergence of the c-th category prediction probability between the source domain black box model and the target domain model, It represents the KL divergence of the predicted probabilities of the source domain black box model and the target domain model excluding the c-th class. is the predicted probability of the cth class of the source domain black box model, represents the predicted probability of other classes excluding class c by the source domain black box model, is the predicted probability of the c-th sample of the target domain model, represents the prediction probability of other classes excluding class c in the target domain model, K represents the total number of fault categories, represents the predicted probability of the i-th category of the source domain black box model, represents the predicted probability of the target domain model for the i-th category, λ is the first hyperparameter, and category c is the fault category with the highest probability of accurate prediction.

[0113] The consistency loss module is used to obtain the consistency loss of the target domain model based on the target domain dataset using the interpolation consistency regularization method. The consistency loss expression is:

[0114]

[0115] Where: X t represents the target domain dataset, L mix represents the consistency loss, represents the qth target domain data in the target domain dataset, represents the jth target domain data in the target domain dataset, The input of the target domain model is The output when The input of the target domain model is Output when Mix β (a, b) = β·a+(1-β)·b, β∈Beat(δ,δ) means β is sampled from the Beta distribution, δ is the parameter value, for and The KL divergence between .

[0116] The nuclear norm loss module is used to obtain the nuclear norm loss of the target domain model based on the target domain model output.

[0117] Supervised loss with adversarial regularizer, used to obtain the supervised loss with adversarial regularizer of the target domain model based on the target domain model output.

[0118] The model optimization module is used to optimize the target domain model based on KL divergence loss, consistency loss, nuclear norm loss, and supervised loss with adversarial regularizer.

[0119] The repeated execution module is used to repeatedly execute the data input module to the model optimization module to obtain the final target domain model.

[0120] The fault diagnosis module is used to diagnose mechanical equipment faults based on the final target domain model.

[0121] The present invention also provides a rolling bearing migration fault diagnosis method based on a source domain black box model, which includes:

[0122] S10, obtain the acceleration signal of the rolling bearing under various states, and preprocess the acceleration signal to obtain a preprocessed data set, and divide the preprocessed data set into a source domain data set and a target domain data set. Specifically, the acceleration signal of the mechanical equipment under various health states is collected by an acceleration sensor, and several data sets are obtained by window sliding. For several data sets, normal standardization preprocessing is used to normalize the data amplitude to the [0,1] interval to ensure data consistency and obtain a preprocessed data set. The rolling bearing is any one of a deep groove ball bearing, a cylindrical roller bearing and a thrust bearing. The rolling bearing includes a bearing outer ring, a bearing inner ring, rolling elements and a cage. The rolling bearing is set on the bearing seat, and the acceleration sensor measures the acceleration signal of the bearing seat. The inner ring of the bearing cooperates with the shaft and rotates with the shaft; the outer ring of the bearing cooperates with the bearing seat and plays a supporting role; the rolling elements are evenly distributed between the inner ring and the outer ring of the bearing through the cage.

[0123] S20, construct a source domain black box model. Based on the source domain dataset, the source domain black box model is trained in combination with the smoothed cross entropy loss function to obtain a trained source domain black box model. The input data and parameters of the trained source domain black box model become unavailable. The smoothed cross entropy loss function is expressed as:

[0124]

[0125] Where: is the source domain dataset, is the label corresponding to the source domain dataset, represents the source domain black box model, represents the output of the source domain black box model when the input is x, represents the predicted probability of the kth class of the source domain black box model, is a smooth label, and When y=k, q k =1, otherwise q k =0, α is the smoothing parameter, α=0.1, u(k)=1 / K is the uniform distribution of labels.

[0126] S30: Input any source domain data in the source domain dataset into the trained source domain black box model to obtain the source domain black box model output.

[0127] S40, build the target domain model.

[0128] S50: Input any target domain data in the target domain data set into the target domain model to obtain the target domain model output.

[0129] S60: Obtain the KL divergence loss between the source domain black box model output and the target domain model output. The KL divergence loss is expressed as:

[0130]

[0131] Where:

[0132] D kl (p s ||p t ) represents the KL divergence loss, p s represents the output of the source domain black box model, p t Denotes the target domain model output, D kl (b s ||b t ) represents the KL divergence of the c-th category prediction probability between the source domain black box model and the target domain model, It represents the KL divergence of the predicted probabilities of the source domain black box model and the target domain model excluding the c-th class. is the predicted probability of the cth class of the source domain black box model, represents the predicted probability of other classes excluding class c by the source domain black box model, is the predicted probability of the c-th sample of the target domain model, represents the prediction probability of other classes excluding class c in the target domain model, K represents the total number of fault categories, represents the predicted probability of the i-th category of the source domain black box model, represents the predicted probability of the target domain model for the i-th category, λ is the first hyperparameter, and category c is the fault category with the highest probability of accurate prediction.

[0133] The predicted probability of excluding the cth class from the source domain black box model output and the target domain model output is expressed as Then the predicted probability of the non-target class of class i (i≠c) is Therefore, the following expression can be obtained:

[0134]

[0135] This gives the KL divergence loss expression.

[0136] S70, based on the target domain dataset, uses the interpolation consistency regularization method to obtain the consistency loss of the target domain model. The consistency loss expression is:

[0137]

[0138] Where: X t represents the target domain dataset, L mixrepresents the consistency loss, represents the qth target domain data in the target domain dataset, represents the jth target domain data in the target domain dataset, The input of the target domain model is The output when The input of the target domain model is Output when Mix β (a, b) = β·a+(1-β)·b, β∈Beat(δ,δ) means β is sampled from Beta distribution, δ is the parameter value, δ = 0.3, for and The KL divergence between .

[0139] The consistency loss is implemented based on MixUp.

[0140] S80, obtaining the nuclear norm loss of the target domain model based on the output of the target domain model. The nuclear norm loss expression is:

[0141]

[0142] Where: L nm is the nuclear norm loss, B is the length of the qth target domain data in the target domain dataset, || || * represents the nuclear norm.

[0143] S90, obtain the supervised loss of the target domain model with the adversarial regularizer based on the output of the target domain model. The supervised loss with the adversarial regularizer is expressed as:

[0144]

[0145] Where: L ce is a supervised loss function with an adversarial regularizer; y e For easy sample X e Pseudo labels of θ t are the parameters of the target domain model, θ Ω is the parameter of Ω, Ω is the adversarial domain adaptation discriminator, For easy sample sets, is a difficult sample set, L sa is the supervised loss with adversarial regularizer, f t is the target domain model, f t (X h ) indicates that the input of the target domain model is X h The output when f t (X e ) indicates that the input of the target domain model is Xe Output when .

[0146] Preferably, for the target domain model that is prone to fall into the local optimal parameter space guided by easy samples, a difficult sample adaptive learning strategy guided by easy samples is designed, and the self-entropy of each target domain data in the target domain dataset is first obtained.

[0147] Furthermore, the average self-entropy is used as the division criterion to obtain the average value of each entropy. The target domain data corresponding to the self-entropy less than the average value is regarded as the easy sample set, and the target domain data corresponding to the self-entropy greater than or equal to the average value is regarded as the difficult sample set.

[0148] The expression of self-entropy is:

[0149]

[0150] Where: express The self-entropy, Indicates that the target domain model number input is The predicted probability of the kth class when .

[0151] By introducing the adversarial domain adaptation discriminator Ω, the distance between easy samples and difficult samples in the target sample is shortened in the high-dimensional space.

[0152] S100 optimizes the target domain model based on KL divergence loss, consistency loss, nuclear norm loss, and supervised loss with adversarial regularizer.

[0153] S110, repeat S50-S100 to obtain the final target domain model.

[0154] S120, performing fault diagnosis on the rolling bearing based on the final target domain model.

[0155] Specifically, a rolling bearing is taken as an example for further explanation. The sampling frequency of the acceleration signal of the rolling bearing is 10 kHz.

[0156] The traditional method that can obtain source domain data and source model parameters and the diagnostic method that does not have passive domain data but can obtain source model parameters are taken as references. The comparison methods are as follows: 1) Directly testing the target data based on the source model trained with labeled data in the source domain (SourceOnly); 2) Traditional domain adaptation fault diagnosis method based on maximum mean discrepancy (MAD); 3) Traditional domain adaptation transfer fault diagnosis method based on correlation alignment (CAD); 4) Adversarial learning-based adaptation diagnosis (ALD); 5) Domain adaptation fault diagnosis method for passive data (Source hypothesis transfer, SHOT); 6) Improved domain adaptation fault diagnosis method for passive data (Source free adaptation diagnosis, SFAD); 7) Comparative fault diagnosis method constructed by replacing the knowledge decoupled self-distillation method of the present invention with the traditional knowledge distillation method (Decoupled self-distillation in KTDA, KTDArd). For domain-adapted intelligent diagnostic methods for transmissions, such as MAD, CAD, and ALD, 70% of the training and test data are used to train the source and target models, with the remaining target data used for final model performance verification testing. For SHOT, KTDArd, and the proposed SBAD method, source data is unavailable. Models must be trained using target data based on the source model, with 70% of the target data used for model training and the remaining data for model verification testing.

[0157] Specifically applied to rolling bearing fault diagnosis, the present invention uses transmission test bench bearing data for verification, the test bench schematic diagram is shown as follows Figure 4 The source domain black box model structure used is as follows: Figure 2 The specific process and parameters are as shown in Figure 3 shown.

[0158] The test bench includes a drive motor 1, a coupling 2, a speed and torque sensor 5, a planetary gearbox 3, a parallel shaft gearbox 4, a rolling bearing under test 6, a magnetic powder brake 8, and an acceleration sensor 7. The test bench is powered by the drive motor 1. The present invention uses vibration data with a sampling frequency of 10kHz for analysis. The rolling bearing under test has seven health states: normal, inner ring crack, inner ring wear, outer ring crack, outer ring wear, inner and outer ring cracks, and retainer failure. The test conditions include W11: speed 1020 rpm, load 0.7 Nm; W12: speed 1020 rpm, load 1.2 Nm; W13: speed 1020 rpm, load 2 Nm; W21: speed 780 rpm, load 0.7 Nm; and W22: speed 780 rpm, load 1.2 Nm. Fourteen migration diagnostic tasks were established using these five test conditions: six for speed variation and eight for load variation.

[0159] Based on a constructed acceleration signal dataset, after end-to-end training and testing using multiple methods, the final diagnostic classification accuracy results are shown in Table 1. "W11→W12" represents the migration diagnosis from the source domain condition W11 to the target domain condition W12, and the same applies to the others. Each diagnostic method was tested five times for each diagnostic task, and the average test accuracy and standard deviation were taken to more comprehensively verify the effectiveness of the present invention. Table 1 uses the √ and × expressions to describe whether the tested diagnostic method meets the requirements of the passive data and passive model parameter settings. As can be seen from the table: 1) Even without access to source data and source model parameters, the SBAD method of the present invention still achieves the best average diagnostic accuracy (98.71%); 2) Because the SourceOnly method cannot obtain the target data distribution, its average diagnostic accuracy as a conventional diagnostic method is only 83.53%; 3) Although the ALD, SHOT, and SFAD methods outperform SBAD in certain diagnostic tasks, the ALD, SHOT, and SFAD methods are able to obtain more prior knowledge, namely source data or source model parameters; 4) After replacing the decoupled self-distillation method, the diagnostic performance of the KTDArd method decreases, verifying the effectiveness of the target sample feature mining in the present invention; 5) The SBAD proposed in the present invention performs well across multiple diagnostic tasks, with a minimum diagnostic accuracy exceeding 93.08%. Therefore, despite the inability to obtain source data and source model parameters during the target diagnostic model training process, the SBAD proposed in the present invention still stands out among various mainstream diagnostic methods.

[0160]

[0161]

[0162] Specifically applied to gearbox fault diagnosis, another embodiment of the present invention uses nuclear power gearbox (scaled) test bench test data for verification. The test bench schematic diagram is shown as follows: Figure 4 The source domain black box model structure used is as follows: Figure 2 The specific process and parameters are as shown in Figure 3 shown.

[0163] The test bench includes a drive motor 1, a coupling 2, a planetary gearbox under test 3, a support bearing 5, a bevel gear 6, a gear pump 7, and an accelerometer 4. The test bench is powered by the drive motor 1. The present invention uses vibration data with a sampling frequency of 12.8kHz for analysis. The planetary gearbox under test has eight health states: normal, sun gear spalling, sun gear wear, sun gear cracks, planet gear pitting, planet gear wear, planet gear cracks, and planet gear severe cracks. The test conditions include T11: speed of 746 rpm, load of 25% of the rated load; T12: speed of 746 rpm, load of 50% of the rated load; T13: speed of 746 rpm, load of 75% of the rated load; T21: speed of 895 rpm, load of 25% of the rated load; T22: speed of 895 rpm, load of 50% of the rated load; and T23: speed of 895 rpm, load of 75% of the rated load. Using these six test conditions, 12 sets of migration diagnostic tasks with varying speeds and loads were established.

[0164] Based on a constructed acceleration signal dataset, after end-to-end training and testing using various methods, the final diagnostic classification accuracy results are shown in Table 1. "T11→T22" represents the transfer diagnosis from source domain condition T11 to target domain condition T22, and the same applies to the others. Table 2 uses √ and × to indicate whether the tested diagnostic methods meet the requirements for passive data and passive model parameters. The table shows that: 1) excluding the Sourceonly method, the average diagnostic accuracy of the other methods exceeds 94%; 2) the average diagnostic accuracy of the KTDA method reaches 99.82%; 3) the traditional ALD method, which requires source domain data, a model, and unlabeled data in the target domain, also achieves an average diagnostic accuracy of 99.33%, but still lags 0.49% behind the proposed SBAD method; 4) while the SFAD method achieves the best average diagnostic accuracy, it requires the acquisition of source domain diagnostic model parameters, which is difficult to achieve in some practical industrial scenarios. Therefore, the SBAD method proposed in this invention has excellent results in terms of diagnostic performance and applicability, standing out among various mainstream diagnostic methods, and can handle more mechanical equipment diagnostic tasks under the constraints of passive field data and source model parameters.

[0165]

[0166]

[0167] The embodiments described above are merely descriptions of preferred implementations of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should fall within the scope of protection determined by the claims of the present invention.

Claims

1. A mechanical equipment migration fault diagnosis method based on a source domain black box model, characterized in that: It includes: S1, acquiring acceleration signals generated by stiffness fluctuations caused by the motion contact of mechanical equipment parts in various states, preprocessing the acceleration signals to obtain a preprocessed dataset, and dividing the preprocessed dataset into a source domain dataset and a target domain dataset; S2, constructing a source domain black box model, and training the source domain black box model based on the source domain dataset and combining a smoothed cross entropy loss function to obtain a trained source domain black box model; S3, inputting any source domain data in the source domain data set into the trained source domain black box model to obtain an output acceleration signal of the source domain black box model; S4, build the target domain model; S5, inputting any target domain data in the target domain data set into the target domain model to obtain an output acceleration signal of the target domain model; S6, obtaining the KL divergence loss between the acceleration signal output by the source domain black box model and the acceleration signal output by the target domain model; the KL divergence loss expression is: Where: D kl (p s ||p t ) represents the KL divergence loss, p s Represents the acceleration signal output by the black box model in the source domain, p t Indicates the output acceleration signal of the target domain model, D kl (b s ||b t ) represents the KL divergence of the c-th category prediction probability between the source domain black box model and the target domain model, represents the KL divergence of the predicted probabilities of the source domain black box model and the target domain model excluding non-c classes, is the predicted probability of the cth class of the source domain black box model, represents the prediction probability of the source domain black box model excluding non-c classes, is the predicted probability of the c-th sample of the target domain model, represents the predicted probability of the target domain model excluding non-c classes, K represents the total number of fault categories, represents the predicted probability of the i-th category of the source domain black box model, represents the prediction probability of the target domain model for the i-th category, λ is the first hyperparameter, and category c is the fault category with the highest probability of accurate prediction; S7, based on the target domain dataset, using an interpolation consistency regularization method to obtain the consistency loss of the target domain model; The consistency loss expression is: Where: X t represents the target domain dataset, L mix represents the consistency loss, represents the qth target area acceleration signal data in the target area dataset, represents the j-th target area acceleration signal data in the target area dataset, The input of the target domain model is The output when The input of the target domain model is Output when Mix β (a, b) = β·a+(1-β)·b, β∈Beat(δ,δ) means β is sampled from the Beta distribution, δ is the parameter value, for and The KL divergence between S8, obtaining a nuclear norm loss of the target domain model based on the acceleration signal output by the target domain model; S9, obtaining a supervised loss with an adversarial regularizer of the target domain model based on the acceleration signal output by the target domain model; S10, optimizing the target domain model based on the KL divergence loss, the consistency loss, the nuclear norm loss, and the supervised loss with the adversarial regularizer; S11, repeat S5-S10 to obtain the final target domain model; S12: Based on the final target domain model, the real-time acceleration signal of the mechanical equipment is input into the final target domain model to perform fault diagnosis on the mechanical equipment.

2. The mechanical equipment migration fault diagnosis method based on the source domain black box model according to claim 1 is characterized in that: The smoothed cross entropy loss function expression is: Where: X s is the source domain dataset, Y s is the label corresponding to the source domain dataset, represents the source domain black box model, represents the output of the source domain black box model when the input is x, represents the predicted probability of the kth class of the source domain black box model, is a smooth label, and When y=k, q k =1, otherwise q k =0, α is the smoothing parameter, and u(k)=1 / K is the uniform distribution of labels.

3. The mechanical equipment migration fault diagnosis method based on the source domain black box model according to claim 1 is characterized in that: The nuclear norm loss expression is: Where: L nm is the nuclear norm loss, B is the length of the qth target domain data in the target domain dataset, || || * represents the nuclear norm.

4. The mechanical equipment migration fault diagnosis method based on the source domain black box model according to claim 1 is characterized in that: The supervised loss expression with adversarial regularizer is: Where: L ce is a supervised loss function with an adversarial regularizer; y e For easy sample X e Pseudo labels of θ t are the parameters of the target domain model, θ Ω is the parameter of Ω, Ω is the adversarial domain adaptation discriminator, For easy sample sets, is a difficult sample set, L sa is the supervised loss with adversarial regularizer, f t is the target domain model, f t (X h ) indicates that the input of the target domain model is X h The output when f t (X e ) indicates that the input of the target domain model is X e Output when .

5. The mechanical equipment migration fault diagnosis method based on the source domain black box model according to claim 4 is characterized in that: Obtaining the self-entropy of each target domain data in the target domain data set; The average value of each self-entropy is obtained, and the target domain data corresponding to the self-entropy smaller than the average value is used as the easy sample set, and the target domain data corresponding to the self-entropy greater than or equal to the average value is used as the difficult sample set.

6. The mechanical equipment migration fault diagnosis method based on the source domain black box model according to claim 5 is characterized in that: The expression of the self-entropy is: Where: express The self-entropy, Indicates that the target domain model number input is The predicted probability of the kth class when .

7. A mechanical equipment migration fault diagnosis system based on the source domain black box model, characterized in that: It includes: A data acquisition module is used to acquire acceleration signals of mechanical equipment in various states, preprocess the acceleration signals to obtain a preprocessed data set, and divide the preprocessed data set into a source domain data set and a target domain data set; A first model building module is used to build a source domain black box model, and train the source domain black box model based on the source domain dataset in combination with a smoothed cross entropy loss function to obtain the trained source domain black box model; A model training module, configured to input any source domain data in the source domain dataset into the trained source domain black box model to obtain a source domain black box model output; The second model building module is used to build a target domain model; A data input module, configured to input any target domain data in the target domain data set into the target domain model to obtain a target domain model output; The KL divergence loss module is used to obtain the KL divergence loss between the output of the source domain black box model and the output of the target domain model. The KL divergence loss expression is: Where: D kl (p s ||p t ) represents the KL divergence loss, p s represents the output of the source domain black box model, p t Denotes the target domain model output, D kl (b s ||b t ) represents the KL divergence of the c-th category prediction probability between the source domain black box model and the target domain model, It represents the KL divergence of the predicted probabilities of the source domain black box model and the target domain model excluding the c-th class. is the predicted probability of the cth class of the source domain black box model, represents the predicted probability of other classes excluding class c by the source domain black box model, is the predicted probability of the c-th sample of the target domain model, represents the prediction probability of other classes excluding class c in the target domain model, K represents the total number of fault categories, represents the predicted probability of the i-th category of the source domain black box model, represents the prediction probability of the target domain model for the i-th category, λ is the first hyperparameter, and category c is the fault category with the highest probability of accurate prediction; A consistency loss module, configured to obtain the consistency loss of the target domain model using an interpolation consistency regularization method based on the target domain dataset; The consistency loss expression is: Where: X t represents the target domain dataset, L mix represents the consistency loss, represents the qth target domain data in the target domain dataset, represents the jth target domain data in the target domain dataset, The input of the target domain model is The output when The input of the target domain model is Output when Mix β (a, b) = β·a+(1-β)·b, β∈Beat(δ,δ) means β is sampled from the Beta distribution, δ is the parameter value, for and The KL divergence between A nuclear norm loss module, configured to obtain a nuclear norm loss of the target domain model based on the target domain model output; A supervised loss with an adversarial regularizer, for obtaining a supervised loss with an adversarial regularizer of the target domain model based on the target domain model output; A model optimization module, configured to optimize the target domain model based on the KL divergence loss, the consistency loss, the nuclear norm loss, and the supervised loss with the adversarial regularizer; A repeated execution module, used for repeatedly executing the data input module to the model optimization module to obtain a final target domain model; A fault diagnosis module is used to perform fault diagnosis on mechanical equipment based on the final target domain model.

Images