Search

CN-116204781-B - Rotary machine fault migration diagnosis method and system

CN116204781BCN 116204781 BCN116204781 BCN 116204781BCN-116204781-B

Abstract

The invention provides a rotating machinery fault migration diagnosis method and system, comprising the following steps of respectively obtaining a marked monitoring data set and a non-marked monitoring data set of a rotating part under different running conditions, carrying out the same data preprocessing on a source domain data set and a target domain data set, selecting a deep migration learning algorithm, constructing a deep neural network model, inputting source domain data with a label to obtain an output prediction label, inputting source domain data and target domain data, calculating the loss between the characteristics of the two domain data, calculating the deviation between a prediction result and a real result, carrying out gradient derivation and optimization on the loss, repeating the steps to obtain a trained neural network model, and carrying out the data preprocessing of the step 2 on new monitoring data to obtain the prediction label of a data sample. The invention realizes the integration of the self-adaptive methods in different fields on the algorithm level by constructing a self-adaptive integration framework.

Inventors

  • HE QINGBO
  • HU KUI

Assignees

  • 上海交通大学

Dates

Publication Date
20260505
Application Date
20221109

Claims (9)

  1. 1. A rotary machine fault migration diagnostic method, comprising the steps of: Step 1, respectively acquiring a marked monitoring data set and a non-marked monitoring data set of a rotating component under different running conditions to form a source domain data set and a target domain data set; step2, carrying out the same data preprocessing on the source domain data set and the target domain data set to obtain two sample sets which can be identified by an integrated network algorithm; Step 3, selecting different deep migration learning algorithms, and constructing a deep neural network model according to an integrated migration framework; Step 4, inputting source domain data with labels, obtaining output prediction labels through forward propagation in a constructed network, and comparing the prediction labels with real labels to obtain a loss function; Step 5, inputting source domain data and target domain data, and calculating loss between the features of the two domain data by using a selected deep migration learning algorithm; Step 6, calculating the deviation between the predicted result and the real result of different migration methods, and calculating the corresponding optimization self-adaptive factors of different deep migration learning methods in the integrated framework according to the deviation; Step 7, gradient derivation and optimization are carried out on the loss; Step 8, repeating the iteration steps 4-7 until reaching the convergence condition, and stopping training to obtain a trained neural network model; step 9, inputting new monitoring data into the trained deep neural network model after the data preprocessing in the step 2 to obtain a prediction label of the data sample; the step 6 specifically comprises the following steps: Step 6.1, calculating A-distance after using different migration technologies, wherein the calculation formula is as follows: Wherein, the Representing the j-th sample of the sample, Is a classifier For samples Is provided with an output of (a), The indication function is represented by a representation of the indication function, For two different domain data sets And Is used for the measurement of the sample size of (a), Data representing two different fields in the i-th field adaptive method in classifier Errors in the above; to use the ith domain adaptation method and the last two domains And A-distance between; step 6.2, for the used i-th field self-adapting method, respectively calculating the corresponding optimized self-adapting factors The calculation formula is as follows: Wherein, the Is the A-distance parameter of the adaptive method in the ith field Sum of all A-distance parameters For representing the weight occupied by the migration technique, The weight occupied by the j-th domain self-adaptive method; step 6.3. The obtained optimized adaptive factor Loss function of field adaptive method selected in step 5 In combination, the total loss function of the migration method is obtained, expressed as: 。
  2. 2. The rotary machine fault migration diagnostic method of claim 1, wherein in step 1, the source domain data set is represented as The target domain dataset is represented as , wherein, 、 Respectively the monitoring data of the ith sample in the source domain data set and the corresponding health condition mark thereof, For the ith sample in the target domain dataset, n is the minimum number of training samples for the batch.
  3. 3. The rotary machine fault migration diagnosis method according to claim 1, wherein in the step 6, the deviation between the predicted result and the true result of the different migration methods is calculated using the a-distance.
  4. 4. A method of diagnosing a fault migration of a rotary machine according to claim 3, wherein in step 7, the loss is gradient derived and optimized by a back propagation algorithm.
  5. 5. The rotary machine fault migration diagnosis method according to claim 1, wherein the step 2 specifically comprises the steps of: Step 2.1, removing abnormal values of an original vibration signal of the rotating component by using Laida rule, wherein the formula is as follows: Wherein, the Represents the average value of the signal segments, The value of the ith sample point in the signal sample; is the standard deviation, n is the total number of sampling points in the signal segment; Step 2.2, slicing the time sequence vibration signals with the abnormal values eliminated, wherein each segment comprises 4096 vibration signal sampling points, and obtaining a wavelet time-frequency pattern book through continuous wavelet transformation as the input of a network model, wherein the continuous wavelet transformation formula is as follows: Wherein, the Is a wavelet mother-of-wave function, Is the time shift coefficient and, Is a scale factor, and 。
  6. 6. The rotary machine fault migration diagnostic method of claim 5, wherein in step 3, the deep neural network model is DDAE-based neural network DDAENN; The DDAE-based neural network DDAENN includes a feature extractor G f , a domain discriminator G d , and a label predictor G l ; the DDAE-based neural network DDAENN employs domain countermeasure techniques and MMD-based domain adaptation techniques.
  7. 7. The method according to claim 6, wherein in the step4, the classification accuracy of the source domain labeled data by the deep neural network is trained, and the cross entropy loss is used, and the loss function is as follows: Wherein, the And Is a matrix vector pair obtained by linear transformation by the label predictor G l , f represents the feature extracted by the feature extractor G f , The softmax function is represented by a graph, Representing an indication function, k being the ith sample of the source domain A corresponding real sample tag; In the step 5, for domain countermeasure technology, the loss function is expressed as: Wherein, the The function of sigmod is represented as such, And Is a pair of linear transformation matrix vectors for a domain classifier, The domain label representing the ith training sample, And Representing the source domain and target domain features extracted by feature extractor G f , respectively, m being the total number of samples within a training batch; For MMD-based domain adaptation techniques, its loss function is expressed as: Wherein, the And Representing the number of batch training samples from the source domain and the target domain respectively, And Representing domain invariant features of the two domains of the depth feature extractor output, Is a regenerative core Hilbert space with k characteristic cores.
  8. 8. The fault migration diagnosis method of rotary machine according to claim 1, wherein the step 9 specifically comprises the steps of: Step 9.1, collecting on-line monitoring data of the actual rotary machine through a sensor, and regulating the monitoring data into a proper wavelet time-frequency diagram sample after the data pretreatment in the step 2; And 9.2, sequentially inputting the obtained time-frequency pattern book into the trained neural network model obtained in the step 8 to obtain a prediction label of a corresponding sample, and evaluating the running state of the current equipment according to the prediction result of the label.
  9. 9. A rotary machine fault migration diagnosis system, characterized by employing the rotary machine fault migration diagnosis method according to any one of claims 1 to 8, comprising the following modules: The method comprises the steps of obtaining a marked monitoring data set and a label-free monitoring data set of a rotating component under different running conditions respectively to form a source domain data set and a target domain data set, wherein the monitoring data of an ith sample in the source domain data set and a corresponding health condition mark thereof are respectively, the ith sample in the target domain data set is obtained, and n is the number of training samples in the minimum batch; the module M2 performs the same data preprocessing on the source domain data set and the target domain data set to obtain two sample sets which can be identified by an integrated network algorithm; selecting different deep migration learning algorithms, and constructing a deep neural network model according to an integrated migration framework; The module M4 is used for inputting source domain data with labels, obtaining output prediction labels through forward propagation in a constructed network, and comparing the prediction labels with real labels to obtain a loss function; The module M5 is used for inputting source domain data and target domain data and calculating the loss between the features of the two domain data by utilizing a selected deep migration learning algorithm; the module M6 is used for calculating the deviation between the predicted results and the real results of different migration methods by utilizing the A-distance, and calculating the optimization self-adaptive factors corresponding to different deep migration learning methods in the integrated framework according to the deviation; the module M7 is used for carrying out gradient derivation and optimization on the loss by using a back propagation algorithm; the module M8 is used for repeatedly triggering the modules M4-M7 to execute iteration until reaching convergence conditions, and stopping training to obtain a trained neural network model; and the module M9 is used for preprocessing the new monitoring data through the data of the module 2, and inputting the new monitoring data into the trained deep neural network model to obtain a prediction label of the data sample.

Description

Rotary machine fault migration diagnosis method and system Technical Field The invention relates to the technical field of mechanical state monitoring and fault diagnosis, in particular to a rotary machine fault migration diagnosis method and system, and especially relates to a rotary machine fault migration diagnosis method and system based on dynamic field self-adaptive integration. Background Rotary machines are widely used as one of the core devices in manufacturing. The rotary machine fault diagnosis research is developed, the safe and reliable operation of the rotary machine is ensured, and the rotary machine fault diagnosis method has great practical significance for improving the production benefit of enterprises and ensuring the safety of national economy. The traditional rotary machine fault diagnosis method is often based on a time-frequency domain analysis technology to extract a mode of typical fault characteristics, and most of the methods are complex in operation and have high professional requirements on detection personnel. In the current large background of mass monitoring data, the rapid, real-time and efficient diagnosis and analysis requirements are difficult to adapt. Intelligent diagnosis methods based on artificial intelligence techniques such as deep learning have been widely used since the advent of the artificial intelligence era. These intelligent diagnostic methods have continually made new research progress in the field of fault diagnosis by virtue of powerful feature extraction and fitting capabilities. While these deep smart diagnostic methods have many advantages, there are two reasons that limit their application in complex real-world scenarios. First, these deep learning models require training and testing data to follow the same data distribution, but in practical applications, the collected monitoring data often encompasses different operating conditions and even different mechanical devices. In addition, the monitoring signal data does not have a fault tag, and the cost of manual labeling is quite high, so that the fault data with the tag is seriously deficient. With the continuous progress of the related technology, the transfer learning hopefully relieves the requirement of data acquisition, and provides a possible solution to the above challenges. The purpose of the migration learning is to find a method to combine the source domain containing rich information data and the target domain lacking information, and to use the knowledge learned by the source domain to apply to the related tasks in the target domain. The patent document with publication number CN113076834B discloses a rotary machine fault information processing method, a processing system, a processing terminal and a medium, a neural network model comprising a depth feature extractor, a domain classifier and a state predictor is constructed, migration fault features of rotary part monitoring data from laboratory simulation data and actual engineering equipment are automatically extracted by the neural network model through the depth feature extractor, the difference between two data distributions is shortened by the domain classifier, the state predictor is utilized, domain adaptation constraint is introduced, a fault diagnosis model based on a depth domain self-adaptation countermeasure network is formed, and intelligent fault diagnosis of the rotary machine is realized by the model. However, the patent document still has the defect that monitoring signal data does not have a fault label, subject to the same data distribution. Disclosure of Invention In view of the defects in the prior art, the invention aims to provide a fault migration diagnosis method for rotary machinery. The invention provides a fault migration diagnosis method for rotary machinery, which comprises the following steps: Step 1, respectively acquiring a marked monitoring data set and a non-marked monitoring data set of a rotating component under different running conditions to form a source domain data set and a target domain data set; step2, carrying out the same data preprocessing on the source domain data set and the target domain data set to obtain two sample sets which can be identified by an integrated network algorithm; Step 3, selecting different deep migration learning algorithms, and constructing a deep neural network model according to an integrated migration framework; Step 4, inputting source domain data with labels, obtaining output prediction labels through forward propagation in a constructed network, and comparing the prediction labels with real labels to obtain a loss function; Step 5, inputting source domain data and target domain data, and calculating loss between the features of the two domain data by using a selected deep migration learning algorithm; Step 6, calculating the deviation between the predicted result and the real result of different migration methods, and calculating the corresponding optimization sel