Rotating equipment cross-working-condition open set fault diagnosis method and device

By building and training deep neural networks, using feature extractors, probability neural networks and domain discriminators, the problem of low accuracy in fault diagnosis under different operating conditions is solved, and the accurate identification and diagnosis of fault categories is achieved.

CN118965097AActive Publication Date: 2024-11-15CHINA THREE GORGES CORPORATION

Patent Information

Application Number
CN202410959370.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-15
Estimated Expiration
2044-07-17

AI Technical Summary

Technical Problem

The vibration signal data distribution of rotating equipment varies greatly under different operating conditions, resulting in low accuracy in fault diagnosis and lack of ability to identify fault categories.

Method used

Build source domain and target domain data sets, use deep neural networks for training, including feature extractors, probability neural networks, domain discriminators and unknown detectors, optimize the objective function through transfer learning and weighted domain adversarial training, and achieve cross-work conditions Identification of fault categories.

Benefits of technology

It improves the accuracy of fault diagnosis of rotating equipment, can accurately identify fault categories under different operating conditions, enhances the ability to identify unknown categories, and reduces the dependence on traditional threshold settings.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965097A_ABST
    Figure CN118965097A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of rotating equipment fault diagnosis, and discloses a rotating equipment cross-working-condition open-set fault diagnosis method and device, and the method comprises the steps: constructing a source domain data set and a target domain data set; wherein the source domain data set comprises rotating equipment vibration data with a fault state label under a first operation condition, and the target domain data set comprises rotating equipment vibration data under a second operation condition; training a deep neural network by using the source domain data set and the target domain data set to obtain a trained deep neural network; the deep neural network comprises a feature extractor, a probabilistic neural network, a domain discriminator and an unknown class detector; and performing open set classification on the target domain data set by using the trained deep neural network to obtain a rotating equipment cross-working-condition open set fault category. According to the invention, the fault types of the rotating equipment under different working conditions are accurately identified, and the accuracy of fault diagnosis of the rotating equipment is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of rotating equipment fault diagnosis, and in particular to a method and device for diagnosing rotating equipment faults across operating conditions. Background Art

[0002] Rotating equipment is the core equipment of intelligent industry. Once a failure occurs, it will affect production at the least, causing unplanned maintenance and thus economic losses, or even cause personal injury or death. The fault diagnosis technology of rotating equipment is the core issue of the condition monitoring system.

[0003] Deep learning and vibration signals are used to diagnose faults of rotating equipment. However, rotating equipment is in different operating conditions for a long time. The differences in data distribution of vibration signals under different conditions increase the difficulty of fault identification, and the lack of ability to identify fault categories leads to low accuracy in fault diagnosis of rotating equipment. Summary of the invention

[0004] In view of this, the present invention provides a method and device for open-set fault diagnosis of rotating equipment across operating conditions, so as to solve the problems of high difficulty in identifying rotating equipment faults under different operating conditions and lack of ability to identify fault categories.

[0005] In a first aspect, the present invention provides a method for cross-operating condition open set fault diagnosis of a rotating equipment, the method comprising:

[0006] Constructing a source domain data set and a target domain data set; wherein the source domain data set includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain data set includes vibration data of rotating equipment under a second operating condition;

[0007] The deep neural network is trained using the source domain data set and the target domain data set to obtain a trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector;

[0008] The trained deep neural network is used to perform open set classification on the target domain dataset to obtain the open set fault categories of rotating equipment across working conditions.

[0009] The present embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment. A deep neural network is trained by using vibration data of rotating equipment with fault status labels and vibration data of rotating equipment without fault status labels under different operating conditions. The trained deep neural network is then used to perform open-set classification on a target domain data set to obtain cross-operating condition open-set fault categories of rotating equipment. Transfer learning is used to transfer fault-related knowledge from one operating condition to another operating condition. Furthermore, there are fault category samples that have not appeared in the other operating condition. This achieves accurate identification of rotating equipment fault categories under different operating conditions and improves the accuracy of fault diagnosis of rotating equipment.

[0010] In an optional implementation, a deep neural network is trained using a source domain dataset and a target domain dataset to obtain a trained deep neural network, including:

[0011] Based on the source domain data set, the feature extractor and probabilistic neural network are used to construct the classification training optimization objective function of the source domain data;

[0012] Based on the source domain dataset and the target domain dataset, a weighted domain adversarial training optimization objective function is constructed using feature extractors and domain discriminators.

[0013] Using probabilistic neural networks and unknown class detectors to build weighted classification training optimization objective function;

[0014] The total optimization objective function is constructed based on the classification training optimization objective function of the source domain data, the weighted domain adversarial training optimization objective function and the weighted classification training optimization objective function;

[0015] The deep neural network is trained using the total optimization objective function until the deep neural network converges to obtain the trained deep neural network.

[0016] The present embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment. Based on a source domain data set and a target domain data set, an overall optimization objective function is constructed through a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector. The overall optimization objective function is used to train a deep neural network, so that the training process of the deep neural network aims at accurately identifying fault category samples under different operating conditions, thereby ensuring that the trained deep neural network can accurately identify and diagnose the fault categories of rotating equipment under different operating conditions.

[0017] In an optional implementation, based on the source domain data set, a feature extractor and a probabilistic neural network are used to construct a classification training optimization objective function for the source domain data, including:

[0018] Mapping the vibration data of the rotating equipment with the fault state label under the first operating condition in the source domain data set to the feature space through the feature extractor to obtain the source domain data features;

[0019] Based on the source domain data features, the probability distribution of the source domain data features is obtained using a probabilistic neural network;

[0020] The classification loss of the source domain data set is calculated based on the probability distribution of the source domain data features, and the classification training optimization objective function of the source domain data is constructed based on the classification loss of the source domain data set.

[0021] The present embodiment provides an open-set fault diagnosis method for rotating equipment across operating conditions, which introduces a probabilistic neural network to participate in the training of a deep neural network. The probabilistic neural network is used to learn the probability distribution of known category features in a source domain data set, and calculate the classification loss of the source domain data set based on the probability distribution of the source domain data features. Then, based on the classification loss of the source domain data set, a classification training optimization objective function of the source domain data is constructed, thereby achieving accurate classification of the source domain data.

[0022] In an optional implementation, based on a source domain dataset and a target domain dataset, a feature extractor and a domain discriminator are used to construct a weighted domain adversarial training optimization objective function, including:

[0023] The vibration data of the rotating equipment under the second operating condition in the target domain data set is mapped to the feature space by a feature extractor to obtain the target domain data features;

[0024] Based on the target domain data characteristics, a probabilistic neural network is used to obtain multiple probability likelihood values, and the maximum probability likelihood value is used as the similarity indicator value;

[0025] Normalizing the similarity indicator value to obtain a similarity value;

[0026] Based on the target domain data features and similarity values, the domain discriminator is used to obtain the weighted domain adversarial training loss value;

[0027] The weighted domain adversarial training optimization objective function is constructed based on the weighted domain adversarial training loss value.

[0028] This embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment, which uses the output of a probabilistic neural network to determine the similarity of target domain samples, that is, the similarity value. The strong feature distribution fitting ability of the probabilistic neural network can more accurately calculate the similarity of the target domain samples, thereby improving the classification accuracy of the target domain samples.

[0029] In an optional implementation, a weighted classification training optimization objective function is constructed using a probabilistic neural network and an unknown class detector, including:

[0030] Based on the source domain data characteristics, the unknown class detector is used to obtain the category classification loss value;

[0031] The target domain data features are sorted based on the similarity value, and the category features are divided based on the sorting result, and the weighted classification loss value is calculated based on the category feature division result;

[0032] The total classification loss value is calculated based on the category classification loss value and the weighted classification loss value, and the weighted classification training optimization objective function is constructed based on the total classification loss value.

[0033] The present embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment, which utilizes an unknown class detector to identify known classes and unknown classes, sorts target domain data features based on similarity values, and then divides class features based on the sorting results. In the process of deep neural network training, a dynamic division form is adopted to classify source domain data and part of target domain data into known classes, and part of target domain data into unknown classes. The similarity value is used in the training of the unknown class detector, which makes the training method more reasonable and further improves the accuracy of fault diagnosis of rotating equipment.

[0034] In an optional implementation, the trained deep neural network is used to perform open set classification on the target domain data set to obtain the cross-operating condition open set fault categories of the rotating equipment, including:

[0035] Based on the target domain samples in the target domain dataset, the target domain sample features are obtained through a feature extractor;

[0036] Based on the sample characteristics of the target domain, the output likelihood probability is obtained through a probabilistic neural network;

[0037] Based on the characteristics of the target domain samples, the unknown class detector is used to determine whether the target domain samples belong to a known category;

[0038] If the target domain samples are of known categories, the open set fault categories of rotating equipment across operating conditions are determined based on the output likelihood probability.

[0039] The present embodiment provides a method for diagnosing cross-operating-condition open-set faults of rotating equipment. Based on the characteristics of target domain samples, an unknown class detector is used to determine whether the target domain samples are of a known class. There is no need to set an empirical threshold. The unknown class detector can directly output whether the sample is of an unknown class. The testing process is more direct and efficient. When the target domain samples are of a known class, the cross-operating-condition open-set fault class of the rotating equipment is determined based on the output likelihood probability, thereby achieving accurate diagnosis of cross-operating-condition open-set faults of the rotating equipment.

[0040] In a second aspect, the present invention provides a device for cross-operating condition open-set fault diagnosis of rotating equipment, the device comprising:

[0041] A construction module is used to construct a source domain data set and a target domain data set; wherein the source domain data set includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain data set includes vibration data of rotating equipment under a second operating condition;

[0042] A training module, used to train a deep neural network using a source domain dataset and a target domain dataset to obtain a trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector;

[0043] The classification module is used to perform open-set classification on the target domain data set using the trained deep neural network to obtain the open-set fault categories of the rotating equipment across working conditions.

[0044] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for cross-operating condition open-set fault diagnosis of rotating equipment of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0045] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for cross-operating condition open-set fault diagnosis of rotating equipment of the first aspect or any corresponding embodiment thereof.

[0046] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the method for cross-operating condition open-set fault diagnosis of rotating equipment of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 It is a flow chart of a method for cross-operating condition open set fault diagnosis of rotating equipment according to an embodiment of the present invention;

[0049] Figure 2 is a flow chart of another method for cross-operating condition open set fault diagnosis of rotating equipment according to an embodiment of the present invention;

[0050] Figure 3is a schematic diagram of a process of deep neural network training according to an embodiment of the present invention;

[0051] Figure 4 is a flow chart of another method for cross-operating condition open set fault diagnosis of rotating equipment according to an embodiment of the present invention;

[0052] Figure 5 is a schematic diagram of a flow chart of cross-operating condition open set fault category judgment of a rotating equipment according to an embodiment of the present invention;

[0053] Figure 6 is a structural block diagram of a cross-operating-condition open-set fault diagnosis device for rotating equipment according to an embodiment of the present invention;

[0054] Figure 7 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0056] Vibration is an inherent characteristic of rotating equipment. The vibration signal contains a lot of information about the equipment's operating status. Analyzing the vibration signal of rotating equipment can identify the health status of the equipment. Deep learning technology can automatically extract fault-related features without manual feature engineering, and is a very promising fault diagnosis tool.

[0057] However, the rotating equipment fault diagnosis technology based on deep learning and vibration signals faces two major challenges. The first is cross-operating condition fault diagnosis. Rotating equipment is in different operating conditions for a long time, such as different speeds, different loads, etc. The data distribution differences (i.e., domain differences) of vibration signals under different operating conditions are very large, which greatly increases the difficulty of fault identification of deep learning models. Another challenge is open set fault diagnosis. The training data cannot exhaust all fault categories. The deployed model often encounters fault categories that did not appear in the training phase. Therefore, the model should have both known class classification and unknown class recognition functions, i.e., open set recognition capabilities. The training data and test data are collected under different operating conditions, and there are a large number of fault categories in the test data that do not exist in the training data. The goal of the open set domain adaptation technology is to classify the known categories in the test data and identify the unknown categories at the same time. It can well solve the above two challenges.

[0058] For open set recognition technology that does not consider domain influence, the flow model in the probabilistic neural network can be used to fit the marginal probability distribution of data features, and the output likelihood probability of the flow model can be used to identify samples of unknown categories; or, the variational autoencoder in the probabilistic neural network can be improved to a conditional variational autoencoder to learn the conditional distribution of data features, and the output likelihood probability of the model can also be used to identify samples of unknown categories.

[0059] An embodiment of the present invention provides an open-set fault diagnosis method for rotating equipment across working conditions, which is applied to a mobile terminal. By introducing a probabilistic neural network into open-set domain adaptive fault diagnosis, a unified probabilistic neural network training paradigm is proposed for learning the conditional distribution of data features. Cross-domain fault diagnosis is considered, and a similarity calculation method for target domain samples is proposed based on the output likelihood probability of the probabilistic neural network. An unknown category detector is added for directly identifying abnormal categories, and a new method is used for training, so that the classification process no longer needs to rely on the traditional method of manually setting thresholds.

[0060] According to an embodiment of the present invention, an embodiment of an open-set fault diagnosis method for a rotating equipment across operating conditions is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0061] In this embodiment, a method for cross-operating condition open set fault diagnosis of rotating equipment is provided, which can be used for the above-mentioned mobile terminal, such as a fault diagnosis device connected to the rotating equipment. Figure 1 is a flow chart of a method for cross-operating condition open set fault diagnosis of rotating equipment according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0062] Step S101, constructing a source domain dataset and a target domain dataset; wherein the source domain dataset includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain dataset includes vibration data of rotating equipment under a second operating condition.

[0063] Specifically, the vibration data of the rotating equipment from one or more vibration sensors on the rotating equipment under the first operating condition are collected, and then the vibration data of the rotating equipment is preprocessed, and the fault state labels are manually marked, which are recorded as the source domain data set. The source domain data set can be expressed as:

[0064]

[0065] in, represents the preprocessed source domain sample data (i.e., rotating equipment vibration data), Representation sample Corresponding fault status label, n s Represents the number of samples in the source domain dataset, where the label Y s represents the source domain label set, totaling |Y s | categories.

[0066] Furthermore, the vibration data of the rotating equipment under another operating condition (i.e., the second operating condition) is collected and preprocessed, and recorded as a target domain data set. The target domain data set can be expressed as:

[0067]

[0068] in, represents the target domain sample data after preprocessing, n t represents the number of samples in the target domain dataset, sample Corresponding unknown label Assume that the target domain label set is Y t , total |Y t | categories, That is, in addition to the known categories in the source domain, there are also unknown categories in the target domain.

[0069] Furthermore, the preprocessing methods of rotating equipment vibration data include but are not limited to denoising methods such as low-pass filtering and wavelet transform, feature extraction methods such as fast Fourier transform and variational mode decomposition, data enhancement methods such as multi-channel data stacking and one-dimensional vibration conversion to two-dimensional image, etc.

[0070] Step S102, using the source domain data set and the target domain data set to train the deep neural network to obtain a trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator and an unknown class detector.

[0071] Specifically, the deep neural network includes a feature extractor E, |Y s |Probabilistic Neural Network The algorithm consists of four parts: domain discriminator D, unknown class detector O; the feature extractor maps samples to feature space; the probabilistic neural network is used to learn the probability distribution of each category feature; the domain discriminator is used for weighted adversarial training to align the known classes of the source domain and the target domain in the feature space, and its output dimension is 2, which is the probability vector of the sample belonging to the source domain and the target domain; the unknown class detector is used to classify known categories and unknown categories, and its output dimension is 2, which is the probability vector of the sample belonging to known categories and unknown categories.

[0072] Furthermore, probabilistic neural networks include but are not limited to variational autoencoders, probability flow models, and other neural networks that can be used for data probability density estimation.

[0073] Step S103, using the trained deep neural network to perform open set classification on the target domain data set to obtain open set fault categories of rotating equipment across working conditions.

[0074] Specifically, the task of cross-operating condition open set fault diagnosis is to use the source domain dataset D s For the target domain dataset D t The samples can be classified correctly. t Known faults in and identify D t Unknown fault in .

[0075] Furthermore, corresponding maintenance actions are performed based on the open-set fault diagnosis output of the rotating equipment by the trained deep neural network; when the trained deep neural network determines that the rotating equipment is in a normal state, the maintenance strategy is not executed; when the trained deep neural network determines that the rotating equipment is a known fault, the corresponding maintenance strategy is executed; when the network determines that the rotating equipment is an unknown fault, the subsequent maintenance strategy should be comprehensively determined based on other data.

[0076] The present embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment. A deep neural network is trained by using vibration data of rotating equipment with fault status labels and vibration data of rotating equipment without fault status labels under different operating conditions. The trained deep neural network is then used to perform open-set classification on a target domain data set to obtain cross-operating condition open-set fault categories of rotating equipment. Transfer learning is used to transfer fault-related knowledge from one operating condition to another operating condition. Furthermore, there are fault category samples that have not appeared in the other operating condition. This achieves accurate identification of rotating equipment fault categories under different operating conditions and improves the accuracy of fault diagnosis of rotating equipment.

[0077] In this embodiment, a method for cross-operating condition open set fault diagnosis of rotating equipment is provided, which can be used for the above-mentioned mobile terminal, such as a fault diagnosis device connected to the rotating equipment. Figure 2 is a flow chart of a method for cross-operating condition open set fault diagnosis of rotating equipment according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0078] Step S201, constructing a source domain dataset and a target domain dataset; wherein the source domain dataset includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain dataset includes vibration data of rotating equipment under a second operating condition. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0079] Step S202, training a deep neural network using the source domain data set and the target domain data set to obtain a trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector.

[0080] Specifically, Figure 3 As shown, the above step S202 includes:

[0081] Step S2021, based on the source domain data set, using a feature extractor and a probabilistic neural network to construct a classification training optimization objective function for the source domain data.

[0082] In some optional implementations, the above step S2021 includes:

[0083] Step a1: Map the vibration data of the rotating equipment with a fault state label under the first operating condition in the source domain data set to the feature space through a feature extractor to obtain the source domain data features.

[0084] Specifically, for the source domain data sample The source domain data features extracted by feature extractor E for:

[0085]

[0086] Step a2: Based on the source domain data features, a probabilistic neural network is used to obtain the probability distribution of the source domain data features.

[0087] Specifically, the source domain data features Input the kth probabilistic neural network V k , output log-likelihood probability for:

[0088]

[0089] Among them, P Z Represents the probability density function of the random variable Z.

[0090] Furthermore, the method by which the kth probabilistic neural network calculates the log-likelihood probability is determined by the type of probabilistic neural network selected. For example, the variational autoencoder performs approximate log-likelihood probability calculation through the encoding-decoding architecture and the variational model, and the flow model calculates the log-likelihood probability through the Jacobian determinant derived from the reversible neural network.

[0091] Step a3: calculate the classification loss of the source domain data set based on the probability distribution of the source domain data features, and construct a classification training optimization objective function for the source domain data based on the classification loss of the source domain data set.

[0092] Specifically, the classification loss L of the source domain dataset isclf The calculation formula is as follows:

[0093]

[0094] Among them, L ce represents the cross entropy loss function with SoftMax function (normalized exponential function), γ and λ are hyperparameters, and the first term Can ensure that the network is reliable for source domain data samples The classification is correct, the second Can guarantee the Probabilistic Neural Network Fully learn features The probability distribution of .

[0095] Furthermore, the expression of the classification training optimization objective function of the source domain data is as follows:

[0096]

[0097] The present embodiment provides an open-set fault diagnosis method for rotating equipment across operating conditions, which introduces a probabilistic neural network to participate in the training of a deep neural network. The probabilistic neural network is used to learn the probability distribution of known category features in a source domain data set, and calculate the classification loss of the source domain data set based on the probability distribution of the source domain data features. Then, based on the classification loss of the source domain data set, a classification training optimization objective function of the source domain data is constructed, thereby achieving accurate classification of the source domain data.

[0098] Step S2022: Based on the source domain dataset and the target domain dataset, a feature extractor and a domain discriminator are used to construct a weighted domain adversarial training optimization objective function.

[0099] In some optional implementations, the above step S2022 includes:

[0100] Step b1, mapping the vibration data of the rotating equipment under the second operating condition in the target domain data set to the feature space through a feature extractor to obtain the target domain data features.

[0101] Specifically, for the target domain data sample The target domain data features extracted by feature extractor E for:

[0102]

[0103] Step b2, based on the target domain data characteristics, a probabilistic neural network is used to obtain multiple probability likelihood values, and the maximum probability likelihood value is used as the similarity indicator value.

[0104] Specifically, the target domain data features Input the kth probabilistic neural network V k , output probability likelihood value for:

[0105]

[0106] Furthermore, |Y s |Probabilistic neural network for features The output maximum probability likelihood value is used as the similarity indicator value, and the similarity indicator value ω j ′ can be expressed as:

[0107]

[0108] Step b3, normalizing the similarity indicator value to obtain a similarity value.

[0109] Specifically, within a training batch of the target domain dataset, the similarity indicator value ω j ′ is normalized to obtain the similarity value, similarity value ω j The calculation formula is as follows:

[0110]

[0111] Among them, ω′ max and ω′ min are the maximum similarity indicator value and the minimum similarity indicator value in the same training batch, respectively. ε is a very small positive value, which is usually selected to be no greater than 10. -5 Positive value.

[0112] Step b4, based on the target domain data features and similarity values, the domain discriminator is used to obtain the weighted domain adversarial training loss value.

[0113] Specifically, the domain discriminator is used to align the known classes in the source domain dataset and the target domain dataset in the feature space. The calculation formula of its weighted domain adversarial training loss value is as follows:

[0114]

[0115] Among them, d s and d t Represent the domain labels of the source domain dataset and the target domain dataset respectively, ω j Represents the target domain sample Samples from the source domain The similarity of samples The probability of belonging to a known class.

[0116] Step b5: construct a weighted domain adversarial training optimization objective function based on the weighted domain adversarial training loss value.

[0117] Specifically, the expression of the weighted domain adversarial training optimization objective function is as follows:

[0118] min E -L adv (12)

[0119] min D L adv (13)

[0120] This embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment, which uses the output of a probabilistic neural network to determine the similarity of target domain samples, that is, the similarity value. The strong feature distribution fitting ability of the probabilistic neural network can more accurately calculate the similarity of the target domain samples, thereby improving the classification accuracy of the target domain samples.

[0121] Step S2023, using the probabilistic neural network and the unknown class detector to construct a weighted classification training optimization objective function.

[0122] In some optional implementations, the above step S2022 includes:

[0123] Step c1: Based on the source domain data features, the unknown class detector is used to obtain the category classification loss value.

[0124] Specifically, for the source domain data features The unknown class detector O classifies it as a known class, and the class classification loss value The calculation formula is as follows:

[0125]

[0126] Among them, d k Represents known category labels.

[0127] Step c2, sorting the target domain data features based on the similarity value, and classifying the features based on the sorting result, and calculating the weighted classification loss value based on the class feature division result.

[0128] Specifically, for the target domain data features According to ω j In order from small to large Sorting, the first ρ% of the target domain data features are taken as the unknown class feature group, denoted as The number is [n t ρ%]; the latter ρ% of the target domain data features are taken as the known class feature group, denoted as The number is [n t ρ%]; the target domain data features in the middle part are taken as the pending group, where [*] represents the integer part of *; ρ is usually selected as a positive value not greater than 30.

[0129] Furthermore, the weighted classification loss The calculation formula is as follows:

[0130]

[0131] Among them, d uk Represents unknown class labels.

[0132] Step c3, calculating the total classification loss value based on the category classification loss value and the weighted classification loss value, and constructing the weighted classification training optimization objective function based on the total classification loss value.

[0133] Specifically, the total classification loss value L o The calculation formula is as follows:

[0134]

[0135] Furthermore, the expression of the weighted classification training optimization objective function is as follows:

[0136] min E,O L o (17)

[0137] The present embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment, which utilizes an unknown class detector to identify known classes and unknown classes, sorts target domain data features based on similarity values, and then divides class features based on the sorting results. In the process of deep neural network training, a dynamic division form is adopted to classify source domain data and part of target domain data into known classes, and part of target domain data into unknown classes. The similarity value is used in the training of the unknown class detector, which makes the training method more reasonable and further improves the accuracy of fault diagnosis of rotating equipment.

[0138] Step S2024, constructing a total optimization objective function based on the classification training optimization objective function of the source domain data, the weighted domain adversarial training optimization objective function and the weighted classification training optimization objective function.

[0139] Specifically, the expression of the overall optimization objective function is as follows:

[0140]

[0141] min D αL adv (19)

[0142] Among them, α and β are adjustable balance coefficients.

[0143] Step S2025, training the deep neural network using the total optimization objective function until the deep neural network converges to obtain a trained deep neural network.

[0144] Step S203, use the trained deep neural network to perform open set classification on the target domain data set to obtain the cross-operating open set fault categories of the rotating equipment. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0145] The present embodiment provides a method for cross-operating condition open-set fault diagnosis of rotating equipment. Based on a source domain data set and a target domain data set, an overall optimization objective function is constructed through a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector. The overall optimization objective function is used to train a deep neural network, so that the training process of the deep neural network aims at accurately identifying fault category samples under different operating conditions, thereby ensuring that the trained deep neural network can accurately identify and diagnose the fault categories of rotating equipment under different operating conditions.

[0146] In this embodiment, a method for cross-operating condition open set fault diagnosis of a rotating device is provided, which can be used for the above-mentioned mobile terminal, such as a fault diagnosis device connected to the rotating device. Figure 4 is a flow chart of a method for cross-operating condition open set fault diagnosis of rotating equipment according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0147] Step S401, constructing a source domain dataset and a target domain dataset; wherein the source domain dataset includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain dataset includes vibration data of rotating equipment under a second operating condition. Figure 2 Step S201 of the illustrated embodiment will not be described in detail here.

[0148] Step S402: train the deep neural network using the source domain dataset and the target domain dataset to obtain a trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector. Figure 2 Step S202 of the illustrated embodiment will not be described in detail here.

[0149] Step S403: Use the trained deep neural network to perform open set classification on the target domain data set to obtain open set fault categories of rotating equipment across working conditions.

[0150] Specifically, Figure 5 As shown, the above step S403 includes:

[0151] Step S4031, based on the target domain samples in the target domain data set, the target domain sample features are obtained through a feature extractor.

[0152] Specifically, the feature extractor E extracts the target domain data samples Perform feature extraction to obtain features Its calculation formula is the same as the above formula (7).

[0153] Step S4032, based on the target domain sample characteristics, obtain the output likelihood probability through a probabilistic neural network.

[0154] Specifically, probabilistic neural networks Features The output likelihood probability is

[0155] Step S4033: Based on the target domain sample features, determine whether the target domain sample is a known category through an unknown class detector.

[0156] Step S4034: if the target domain sample is a known category, determine the cross-operating condition open set fault category of the rotating equipment based on the output likelihood probability.

[0157] Specifically, when the unknown class detector O transforms the target domain sample features When classified as an unknown class, identify the sample is an unknown fault; otherwise, when O takes the target domain sample feature When classified as a known class, the sample is identified For known categories Identification features For Category The classification formula is expressed as:

[0158]

[0159] The present embodiment provides a method for diagnosing cross-operating-condition open-set faults of rotating equipment. Based on the characteristics of target domain samples, an unknown class detector is used to determine whether the target domain samples are of a known class. There is no need to set an empirical threshold. The unknown class detector can directly output whether the sample is of an unknown class. The testing process is more direct and efficient. When the target domain samples are of a known class, the cross-operating-condition open-set fault class of the rotating equipment is determined based on the output likelihood probability, thereby achieving accurate diagnosis of cross-operating-condition open-set faults of the rotating equipment.

[0160] In this embodiment, a rotating equipment cross-operating condition open-set fault diagnosis device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0161] This embodiment provides a rotating equipment cross-operating condition open-collection fault diagnosis device, such as Figure 6 As shown, including:

[0162] A construction module 601 is used to construct a source domain data set and a target domain data set; wherein the source domain data set includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain data set includes vibration data of rotating equipment under a second operating condition;

[0163] A training module 602 is used to train a deep neural network using a source domain data set and a target domain data set to obtain a trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector;

[0164] The classification module 603 is used to perform open set classification on the target domain data set using the trained deep neural network to obtain the open set fault categories of the rotating equipment across working conditions.

[0165] In some optional implementations, the training module 602 includes:

[0166] The first construction unit is used to construct a classification training optimization objective function of source domain data based on the source domain data set by using a feature extractor and a probabilistic neural network;

[0167] The second construction unit is used to construct a weighted domain adversarial training optimization objective function based on the source domain dataset and the target domain dataset using a feature extractor and a domain discriminator;

[0168] A third construction unit is used to construct a weighted classification training optimization objective function using a probabilistic neural network and an unknown class detector;

[0169] A fourth construction unit is used to construct a total optimization objective function based on the classification training optimization objective function of source domain data, the weighted domain adversarial training optimization objective function and the weighted classification training optimization objective function;

[0170] The training unit is used to train the deep neural network using the total optimization objective function until the deep neural network converges to obtain the trained deep neural network.

[0171] In some optional embodiments, the first building block includes:

[0172] A first mapping subunit is used to map the vibration data of the rotating equipment with a fault state label under the first operating condition in the source domain data set to a feature space through a feature extractor to obtain source domain data features;

[0173] A first computing subunit is used to obtain a probability distribution of source domain data features by using a probabilistic neural network based on source domain data features;

[0174] The first construction subunit is used to calculate the classification loss of the source domain data set based on the probability distribution of the source domain data features, and to construct a classification training optimization objective function of the source domain data based on the classification loss of the source domain data set.

[0175] In some optional embodiments, the second building block includes:

[0176] A second mapping subunit is used to map the vibration data of the rotating equipment under the second operating condition in the target domain data set to the feature space through a feature extractor to obtain the target domain data features;

[0177] A second calculation subunit is used to obtain multiple probability likelihood values ​​by using a probabilistic neural network based on the target domain data characteristics, and use the maximum probability likelihood value as a similarity indication value;

[0178] A normalization processing subunit, used for normalizing the similarity indication value to obtain a similarity value;

[0179] A third computing subunit is used to obtain a weighted domain adversarial training loss value using a domain discriminator based on target domain data features and similarity values;

[0180] The second construction subunit is used to construct a weighted domain adversarial training optimization objective function based on the weighted domain adversarial training loss value.

[0181] In some optional embodiments, the third building block includes:

[0182] A fourth computing subunit, configured to obtain a category classification loss value by using an unknown class detector based on source domain data features;

[0183] A sorting subunit, used to sort the target domain data features based on the similarity value, and to divide the category features based on the sorting result, and to calculate the weighted classification loss value based on the category feature division result;

[0184] The third construction subunit is used to calculate the total classification loss value based on the category classification loss value and the weighted classification loss value, and to construct a weighted classification training optimization objective function based on the total classification loss value.

[0185] In some optional implementations, the classification module 603 includes:

[0186] An extraction unit, used for obtaining target domain sample features through a feature extractor based on the target domain samples in the target domain data set;

[0187] A computing unit, used to obtain output likelihood probability through a probabilistic neural network based on the sample features of the target domain;

[0188] A judgment unit, used for judging whether the target domain sample is a known category through an unknown class detector based on the target domain sample characteristics;

[0189] A determination unit is used to determine the cross-operating condition open set fault category of the rotating equipment based on the output likelihood probability if the target domain sample is a known category.

[0190] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0191] In this embodiment, a rotating equipment cross-operating condition open-set fault diagnosis device is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above-mentioned functions.

[0192] The embodiment of the present invention also provides a computer device having the above Figure 6 A rotating equipment cross-operating condition open-collection fault diagnosis device is shown.

[0193] See also Figure 7 , Figure 7 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 7 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 7 A processor 10 is taken as an example.

[0194] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0195] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0196] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0197] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0198] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 7 The example of connecting through bus is taken in the following.

[0199] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0200] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0201] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0202] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for cross-operating condition open set fault diagnosis of rotating equipment, characterized in that: The method comprises: Constructing a source domain data set and a target domain data set; wherein the source domain data set includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain data set includes vibration data of rotating equipment under a second operating condition; Using the source domain data set and the target domain data set to train a deep neural network to obtain the trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator, and an unknown class detector; The trained deep neural network is used to perform open set classification on the target domain data set to obtain open set fault categories of rotating equipment across working conditions.

2. The method according to claim 1, characterized in that The step of training a deep neural network using the source domain dataset and the target domain dataset to obtain the trained deep neural network includes: Based on the source domain data set, constructing a classification training optimization objective function for source domain data using the feature extractor and the probabilistic neural network; Based on the source domain dataset and the target domain dataset, constructing a weighted domain adversarial training optimization objective function using the feature extractor and the domain discriminator; Using the probabilistic neural network and the unknown class detector to construct a weighted classification training optimization objective function; Constructing a total optimization objective function based on the classification training optimization objective function of the source domain data, the weighted domain adversarial training optimization objective function and the weighted classification training optimization objective function; The deep neural network is trained using the total optimization objective function until the deep neural network converges to obtain the trained deep neural network.

3. The method according to claim 2, characterized in that The step of constructing a classification training optimization objective function for source domain data based on the source domain data set by using the feature extractor and the probabilistic neural network includes: Mapping the vibration data of the rotating equipment with a fault state label under the first operating condition in the source domain data set to a feature space through a feature extractor to obtain source domain data features; Based on the source domain data features, using the probabilistic neural network to obtain a probability distribution of the source domain data features; The classification loss of the source domain data set is calculated based on the probability distribution of the source domain data features, and a classification training optimization objective function of the source domain data is constructed based on the classification loss of the source domain data set.

4. The method according to claim 3, characterized in that The step of constructing a weighted domain adversarial training optimization objective function based on the source domain dataset and the target domain dataset using the feature extractor and the domain discriminator includes: Mapping the vibration data of the rotating equipment under the second operating condition in the target domain data set to the feature space through a feature extractor to obtain target domain data features; Based on the target domain data features, the probabilistic neural network is used to obtain multiple probability likelihood values, and the maximum probability likelihood value is used as a similarity indicator value; Normalizing the similarity indicator value to obtain a similarity value; Based on the target domain data features and the similarity value, using the domain discriminator to obtain a weighted domain adversarial training loss value; A weighted domain adversarial training optimization objective function is constructed based on the weighted domain adversarial training loss value.

5. The method according to claim 4, characterized in that The method of using the probabilistic neural network and the unknown class detector to construct a weighted classification training optimization objective function includes: Based on the source domain data features, using the unknown class detector to obtain a category classification loss value; Sorting the target domain data features based on the similarity values, performing category feature division based on the sorting results, and calculating a weighted classification loss value based on the category feature division results; A total classification loss value is calculated based on the category classification loss value and the weighted classification loss value, and a weighted classification training optimization objective function is constructed based on the total classification loss value.

6. The method according to claim 1, characterized in that The open set classification of the target domain data set using the trained deep neural network to obtain the cross-operating condition open set fault categories of the rotating equipment includes: Based on the target domain samples in the target domain data set, obtaining target domain sample features through the feature extractor; Based on the target domain sample features, obtaining output likelihood probability through the probabilistic neural network; Based on the target domain sample features, determining whether the target domain sample is a known category by the unknown class detector; If the target domain sample is of a known category, the cross-operating condition open set fault category of the rotating equipment is determined based on the output likelihood probability.

7. A rotating equipment cross-operating condition open-collection fault diagnosis device, characterized in that: The device comprises: A construction module, used to construct a source domain data set and a target domain data set; wherein the source domain data set includes vibration data of rotating equipment with a fault state label under a first operating condition, and the target domain data set includes vibration data of rotating equipment under a second operating condition; A training module, used to train a deep neural network using the source domain data set and the target domain data set to obtain the trained deep neural network; the deep neural network includes a feature extractor, a probabilistic neural network, a domain discriminator and an unknown class detector; A classification module is used to perform open set classification on the target domain data set using the trained deep neural network to obtain open set fault categories of rotating equipment across working conditions.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the cross-operating condition open-set fault diagnosis method for rotating equipment according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the cross-operating condition open-set fault diagnosis method for rotating equipment according to any one of claims 1 to 6.

10. A computer program product, characterized in that It comprises computer instructions, and the computer instructions are used to enable a computer to execute the cross-operating condition open-set fault diagnosis method for rotating equipment according to any one of claims 1 to 6.

Citation Information

Patent Citations

  • Intelligent fault diagnosis method based on deep adversarial domain self-adaption

    CN111898634A

  • Cross-working-condition fault diagnosis method based on open set joint transfer learning

    CN112784872A

  • Open set fault diagnosis method for bearing of high-speed motor train unit

    CN113375941A

  • Deep online migration lithium ion energy storage battery fault diagnosis method

    CN116577667A

  • Rolling bearing variable working condition fault diagnosis method and device, terminal and storage medium

    CN117591950A

Cited By

  • Open set cross-domain fault diagnosis model training method, device and equipment

    CN120011989A

  • Training method, device and equipment of open set cross-domain fault diagnosis model

    CN120011989B