A mechanical heterogeneous fault diagnosis method and device based on multi-model collaborative neural architecture search

Abstract

The present invention belongs to the technical field related to mechanical fault diagnosis, and discloses a mechanical heterogeneous fault diagnosis method and device based on multi-model collaborative neural architecture search, which comprises the following steps: (1) processing vibration acceleration signals to form a training set and a test set; (2) constructing a multi-model collaborative network, and using training samples to train the multi-model collaborative network and optimize the weight of each round of collaborative network; after each round of collaborative network training is completed, using the test set to calculate the reward obtained by the current collaborative network, and determining whether to update the structure of the current collaborative network based on the difference between the rewards of the current collaborative network and the previous round of collaborative network, and using a neural architecture search framework based on an avoidance strategy to generate candidate structures for the next round of collaborative network; (3) the multi-model collaborative network identifies the task scenario based on the sensor parameters of the multi-channel sensor, and then performs fault diagnosis. The present invention realizes heterogeneous fault diagnosis of mechanical equipment while improving the accuracy of diagnosis.

Description

Technical Field

[0001] The present invention belongs to the technical field related to mechanical fault diagnosis, and more specifically, relates to a mechanical heterogeneous fault diagnosis method and equipment based on multi-model collaborative neural architecture search. Background Art

[0002] In modern industrial production, mechanical equipment plays a vital role across a wide range of industries. Their intelligent operation and maintenance has attracted considerable attention from scholars and engineers in recent years. With the development of artificial intelligence (AI), deep learning-based intelligent fault diagnosis methods for machinery have made significant progress. Numerous laboratory datasets on rotating machinery, such as gears and bearings, have been thoroughly analyzed. While these data-driven fault diagnosis methods continue to improve diagnostic accuracy, they still have some limitations.

[0003] Most methods focus on diagnostic objects with complete data acquisition, such as bearing fault simulation platforms. This means that the input to deep learning models always has a fixed data shape and channel dimension. Few studies consider scenarios with diverse data across multiple devices. However, in industry, raw data can be incomplete due to communication anomalies, sensor failures, and differences in diagnostic capabilities among downstream companies. Deep learning models with only a single input type are unable to handle heterogeneous fault diagnosis tasks with varying signal dimensions. This often requires constructing new deep learning model architectures to adapt to varying data structures and distributions, significantly increasing time and computing resource consumption.

[0004] While deep learning-based fault diagnosis methods avoid manual feature extraction, selecting model structures and hyperparameters typically requires extensive expertise and domain experience, coupled with continuous structural design and trial-and-error experiments. Deep learning models are highly sensitive to changes in datasets, even to vibration signals from different parts of a machine captured by sensors. Currently, researchers lack a consensus on how to select the appropriate model for specific diagnostic scenarios and signals within different network structures.

[0005] Neural architecture search is an automated method for designing neural network models. It optimizes candidate networks through search strategies and verification feedback to obtain high-performance models suitable for specific tasks. However, as the search space expands and the number of model layers increases, the computing resources consumed by this method increase exponentially, significantly increasing the cost of intelligent operations and maintenance for some small and medium-sized enterprises. Furthermore, current mainstream neural architecture search methods are typically task-specific, performing detailed searches and hyperparameter optimization on a single neural network structure. This method often lacks advantages when faced with multiple tasks, and companies using network search can incur high costs to resolve a single machine failure. In summary, these issues hinder the promotion of fault diagnosis based on deep learning methods in industrial applications. Summary of the Invention

[0006] In response to the above-mentioned defects or improvement needs of the prior art, the present invention provides a mechanical heterogeneous fault diagnosis method and device based on multi-model collaborative neural architecture search, which aims to realize heterogeneous fault diagnosis of mechanical equipment while improving the accuracy of diagnosis.

[0007] To achieve the above objectives, according to one aspect of the present invention, a method for mechanical heterogeneous fault diagnosis based on multi-model collaborative neural architecture search is provided, which includes the following steps:

[0008] (1) The vibration acceleration signals of the mechanical equipment to be diagnosed collected by the multi-channel sensor are normalized, and the normalized vibration acceleration signals are resampled using a sliding window to obtain a series of time window samples. Labels are added to the time window samples to form a training set and a test set;

[0009] (2) Use multiple models to construct a multi-model collaborative network, and use training samples to train the multi-model collaborative network and optimize the weight of each round of collaborative network; wherein, after each round of collaborative network training is completed, the reward obtained by the current collaborative network is calculated using the test set, and the difference between the rewards of the current collaborative network and the previous round of collaborative network is used to determine whether to update the structure of the current collaborative network, and a neural architecture search framework based on an avoidance strategy is used to generate candidate structures for the next round of collaborative network;

[0010] (3) Multi-model collaborative network based on sensor parameters of multi-channel sensors After identifying the task scenario, the corresponding sub-model is activated to perform fault diagnosis.

[0011] Furthermore, the training set includes sensor data of all channels, and the test set includes sensor data of some or all channels.

[0012] Furthermore, the full-channel training samples are divided into multiple one-dimensional single-channel data, and each sub-model extracts features through the single-channel data; all sub-models are connected to the same decision network.

[0013] Furthermore, for each sub-model, the multi-label based binary classification loss is calculated:

[0014]

[0015] in To use the bth model for n c The predicted value of the i-th sample of the category; in the process of collaborative training, the final inter-ring loss L in Affected by each sub-model, it is expressed as:

[0016]

[0017] where N i and B are the number of batch samples and sub-models participating in each round of collaborative training.

[0018] Furthermore, the training of the multi-model collaborative network includes inner-loop learning and outer-loop learning based on neural architecture search with avoidance strategy; the outer-loop learning includes the performance evaluation of the trained multi-model collaborative network in the current e-th epoch and the generation of candidate multi-model collaborative networks in the (e+1)-th epoch.

[0019] Furthermore, each state of the multi-model collaborative network architecture comes from the search space of the neural network, which contains n potential combinations of states S = {S1, S2, ..., S n}, each state space Contains two aspects of information, namely the branch index to be updated The index of the candidate submodel in model space and model structure

[0020] Furthermore, based on the task information p j Reconstruction state is S e In the multi-model collaborative network, the devices corresponding to the test set samples do not overlap with the devices corresponding to the training set.

[0021] Furthermore, the calculation formula for the average cross-individual accuracy is:

[0022]

[0023] in is the index parameter of the sensor, N p is the total number of heterogeneous tasks, and Evl(·) represents the classic neural network evaluation function.

[0024] The present invention also provides a mechanical heterogeneous fault diagnosis system based on multi-model collaborative neural architecture search, the fault diagnosis system includes a memory and a processor, the memory stores a computer program, and the processor executes the mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search as described above when executing the computer program.

[0025] The present invention also provides a computer-readable storage medium, which stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions prompt the processor to implement the mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search as described above.

[0026] In general, compared with the prior art, the above technical solutions conceived by the present invention provide a method and device for mechanical heterogeneous fault diagnosis based on multi-model collaborative neural architecture search, which has the following beneficial effects:

[0027] 1. The present invention uses the collected multi-channel vibration acceleration signals as the basis for multi-model training and testing, and selects a separate model for the vibration acceleration signal of each channel for decision-making, reducing the risk of overfitting of a single model to specific data, significantly improving the accuracy of mechanical equipment fault diagnosis, and being able to simultaneously diagnose heterogeneous mechanical faults.

[0028] 2. A neural architecture search framework based on an avoidance strategy is used to generate candidate structures for the next round of collaborative networks. This allows for end-to-end diagnosis without the need for complex expert knowledge, greatly reducing search time and computational overhead. This eliminates the need to refine the model from scratch and prevents the model from falling into local optimality, improving search efficiency.

[0029] 3. The devices corresponding to the test set samples do not overlap with the devices corresponding to the training set. The model can perform fault diagnosis on any unknown machine of the same model using known machines, which greatly promotes the application of deep learning-based fault diagnosis in industrial scenarios.

[0030] 4. This invention directly uses multiple high-level model structures as the basic elements of the search, which significantly reduces the search space and computational cost while fully leveraging knowledge and achievements in vertical fields. Furthermore, the inner-loop learning only trains and updates the modified sub-model structures and weights, while the remaining branches are frozen to further reduce computational cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flowchart of a mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search provided by the present invention;

[0032] Figure 2 This is the T-SNE plot of the original data;

[0033] Figure 3 It is a schematic diagram of data collection and segmentation;

[0034] Figure 4 Schematic diagram of the iterative optimization process of outer-loop neural architecture search. DETAILED DESCRIPTION

[0035] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0036] First, let’s define the problem. For multiple motors on an industrial production line, each motor contains multi-channel acceleration and vibration sensor signals. The motor data of the training set is represented as M represents the number of motors, n i represents the number of samples, Represents the i-th short-time data sample obtained from the m-th motor, L represents the length of the sample, and C represents the width of the sample, which is determined by the number of sensors on a motor. Indicates the corresponding label.

[0037] Heterogeneous faults can be viewed as a multi-task scenario characterized by differences in data dimensionality between the test set and the training set. This heterogeneity arises from changes in device state or sudden data loss, such as communication problems, system upgrades or modifications, etc. Therefore, the target device signal of the test set is defined as c is an integer between 1 and C, and the additional parameter Used to encode the sensor position.

[0038] According to general research in the field of fault diagnosis, the distribution differences of signal data collected from different machines are significant, namely:

[0039]

[0040] The main goal of this invention is to establish a universal classification model suitable for multi-task processing to handle heterogeneous signal inputs and improve diagnostic accuracy. Furthermore, it is required to search for the optimal collaborative neural network structure for heterogeneous tasks as efficiently as possible in the model-level space.

[0041] See also Figure 1 、 Figure 2 、 Figure 3 and Figure 4 The present invention provides a mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search, which mainly includes the following steps:

[0042] In step 1, the vibration acceleration signal of the mechanical equipment to be diagnosed collected by the multi-channel sensor is normalized, and the normalized vibration acceleration signal is resampled using a sliding window to obtain a series of time window samples. Labels are added to the time window samples to form a training set and a test set.

[0043] Normalize the vibration acceleration signal to initially reduce inherent differences and transient impact effects at different locations. Use a traditional sliding window to construct a batch of time window samples, creating composite fault labels for subsequent training and testing. The training set contains sensor data from all channels, while the test set contains sensor data from some or all channels.

[0044] Step 2: Use multiple models to construct a multi-model collaborative network, and use training samples to train the multi-model collaborative network and optimize the weights of each round of collaborative network. After each round of collaborative network training is completed, the test set is used to calculate the reward obtained by the current collaborative network, and the difference between the rewards of the current collaborative network and the previous round of collaborative network is used to determine whether to update the structure of the current collaborative network. A neural architecture search framework based on an avoidance strategy is used to generate candidate structures for the next round of collaborative network, and this cycle is repeated until the predetermined number of training times is reached.

[0045] This step primarily involves inner-loop learning of a multi-model collaborative network and outer-loop learning using a neural architecture search based on an avoidance strategy. Unlike traditional single models, multiple models are used for collaborative training and fusion decision-making, forming a multi-model collaborative network (MMCN). The MMCN structure is optimized using a neural architecture search framework based on an avoidance strategy. Training samples are used to train and optimize the weights of the collaborative network in each round. Test samples are used to calculate the rewards earned by the current collaborative network and guide whether to update the collaborative network structure.

[0046] The multiple models used in this implementation are mainly some classic deep learning models and advanced domain generalization models:

[0047] 1) WDCNN: A classic and widely compared fault diagnosis model based on convolutional neural networks, used as the initial model structure of each branch.

[0048] 2) MSMCNN: A multi-scale hybrid convolutional neural network for multi-channel signal fault diagnosis of industrial robots.

[0049] 3) MSRFCNN: A multi-stage residual fusion convolutional neural network for in-situ fault detection of CNC machine tool spindle motors.

[0050] 4) MSRCN: A cross-machine fault diagnosis model for complex faults in industrial motors.

[0051] 5) CFSPT: A lightweight Transformer model for cross-machine fault diagnosis of electric motors.

[0052] 6) MSPT: Multi-Scale Pruned Transformer Model for Cross-Machine Fault Diagnosis.

[0053] Inner-loop learning of multi-model collaborative networks: In inner-loop learning, unlike traditional single-model models, multiple models participate in training and decision output. To accommodate various channel dimensions, the full-channel training samples are split into multiple one-dimensional single-channel data, and each sub-model extracts features from the single-channel data. All sub-models are connected to the same decision network, which means that the weight updates and outputs of each sub-model are synchronized. For each sub-model, the multi-label binary classification loss is calculated:

[0054]

[0055] in To use the bth model for n c The predicted value of the i-th sample of the category. During the collaborative training process, the final inter-ring loss L in Affected by each sub-model, it can be expressed as:

[0056]

[0057] where N i and B are the number of batches and sub-models participating in each round of collaborative training. After the outer loop structure is updated, only the modified sub-models participate in training and weight updates, further reducing computational overhead. Multiple deep learning models participate in the training process, forming a multi-model collaborative network (MMCN), which undergoes multiple iterations based on the classic backpropagation algorithm.

[0058] Outer loop neural architecture search learning based on avoidance strategy: This part mainly includes the performance evaluation of the trained MMCN in the current e-th period and the generation of candidate MMCN in the (e+1)th period. Each MMCN architecture state comes from the search space of the neural network, which contains the potential combination of n states S = {S1, S2, ..., S n}, each state space Contains two aspects of information, namely the branch index to be updated The index of the candidate submodel in model space and model structure Unlike traditional neural architecture searches that focus on detailed basic element objects (such as the number of nodes and layers), this diagnostic method directly uses multiple high-level model structures as the basic elements of the search, which greatly reduces the search space and computational cost while making full use of the knowledge and achievements in vertical fields.

[0059] At the beginning of each outer loop, the two indexes generated by the avoidance strategy and like Figure 4 As shown. The algorithm receives the e-th round and As the search starting point, it receives a fixed-length stack I S ∈R 2×K , this stack provides short-term storage for tracking the model search history within the last K attempts. An avoidance-based diversity search strategy (ADS) is used to randomly generate new candidate values ​​and perform a first check to see if they are the same as the input value:

[0060]

[0061]

[0062] Where I represents the index space, and the two generated values ​​need to be different from the most recent K combinations to further enhance the diversity of the search.

[0063] Then execute the second judgment:

[0064]

[0065] once Through these two judgments, it is considered to be a valid and diverse candidate indicator group. At the same time, the oldest group in the stack is removed to maintain the stack size of K.

[0066] In order to complete the multi-task performance evaluation, based on the task information p j Reconstruction state is S e MMCN, where the devices corresponding to the test set samples do not overlap with the devices corresponding to the training set to ensure generalization performance, calculate the average cross-individual accuracy (CIA):

[0067]

[0068] in is the index parameter of the sensor, which can guide the selection of the corresponding sub-model, N pis the total number of heterogeneous tasks, and Evl(●) represents the classic neural network evaluation function. Get the status as S e+1 The candidate MMCN is trained in the inner loop to obtain a CIA similar to the above formula e+1 Note that the inner loop learning only trains and updates the modified sub-model structure and weights, and the remaining branches are frozen to further reduce the computational cost. The reward R of the (e+1)th search e+1 It can be calculated as:

[0069] R e+1 =CIA e+1 -CIA e

[0070] Whether MMCN updates the structure in the (e+1)th search of the outer loop is based on the reward R e+1 Among them, E M Represents the model-level search space. After N rounds of iteration, the outer loop automatically searches and outputs the optimal MMCN structure. and the corresponding weights.

[0071]

[0072] Step 3: Multi-model collaborative network based on sensor parameters of multi-channel sensors After identifying the task scenario, the corresponding sub-model is activated to perform fault diagnosis.

[0073] Based on sensor parameters Able to identify the corresponding sensor and the channel and position of the sensor.

[0074] The purpose of this step is to diagnose signals of various input dimensions. The multi-model collaborative network receives sensor parameters to recognize the task scenario; the corresponding neural networks are activated and reassembled into a matching collaborative network for final diagnosis.

[0075] The present invention also provides a mechanical heterogeneous fault diagnosis system based on multi-model collaborative neural architecture search, the fault diagnosis system includes a memory and a processor, the memory stores a computer program, and the processor executes the mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search as described above when executing the computer program.

[0076] The present invention also provides a computer-readable storage medium, which stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions prompt the processor to implement the mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search as described above.

[0077] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search, characterized by: The method comprises the following steps: (1) The vibration acceleration signals of the mechanical equipment to be diagnosed collected by the multi-channel sensor are normalized, and the normalized vibration acceleration signals are resampled using a sliding window to obtain a series of time window samples. Labels are added to the time window samples to form a training set and a test set; (2) Use multiple models to construct a multi-model collaborative network, and use training samples to train the multi-model collaborative network and optimize the weight of each round of collaborative network; wherein, after each round of collaborative network training is completed, the test set is used to calculate the reward obtained by the current collaborative network, and the difference between the rewards of the current collaborative network and the previous round of collaborative network is used to determine whether to update the structure of the current collaborative network, and a neural architecture search framework based on an avoidance strategy is used to generate the candidate structure of the next round of collaborative network; the training of the multi-model collaborative network includes inner-loop learning and outer-loop learning of a neural architecture search framework based on an avoidance strategy; the outer-loop learning includes the performance evaluation of the trained multi-model collaborative network in the current e-th period and the generation of candidate multi-model collaborative networks in the (e+1)th period; each multi-model collaborative network architecture state comes from the search space of the neural network, which contains a potential combination of n states S = {S1, S2, ..., S n }, each state space Contains two aspects of information, namely the branch index to be updated The index of the candidate submodel in model space and model structure (3) Multi-model collaborative network based on sensor parameters of multi-channel sensors After identifying the task scenario, the corresponding sub-model is activated to perform fault diagnosis.

2. The method for mechanical heterogeneous fault diagnosis based on multi-model collaborative neural architecture search according to claim 1, characterized in that: The training set includes sensor data of all channels, and the test set includes sensor data of some or all channels.

3. The method for mechanical heterogeneous fault diagnosis based on multi-model collaborative neural architecture search according to claim 1, characterized in that: The full-channel training samples are divided into multiple one-dimensional single-channel data, and each sub-model extracts features through the single-channel data; all sub-models are connected to the same decision network.

4. The method for mechanical heterogeneous fault diagnosis based on multi-model collaborative neural architecture search according to claim 3, characterized in that: For each sub-model, calculate the multi-label based binary classification loss: in To use the bth model for n c The predicted value of the i-th sample of the category; in the process of collaborative training, the final inter-ring loss L in Affected by each sub-model, it is expressed as: where N i and B are the number of batch samples and sub-models participating in each round of collaborative training.

5. The method for mechanical heterogeneous fault diagnosis based on multi-model collaborative neural architecture search according to claim 1, characterized in that: Based on task information p j Reconstruction state is S e In the multi-model collaborative network, the devices corresponding to the test set samples do not overlap with the devices corresponding to the training set.

6. The method for mechanical heterogeneous fault diagnosis based on multi-model collaborative neural architecture search according to claim 1, characterized in that: The calculation formula for the average cross-individual accuracy is: in is the index parameter of the sensor, N p is the total number of heterogeneous tasks, and Evl(·) represents the classic neural network evaluation function.

7. A mechanical heterogeneous fault diagnosis system based on multi-model collaborative neural architecture search, characterized by: The fault diagnosis system includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, it executes the mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by the processor, the machine-executable instructions prompt the processor to implement the mechanical heterogeneous fault diagnosis method based on multi-model collaborative neural architecture search as described in any one of claims 1-6.

Images