Motorcycle engine fault detection method based on vibration effect monitoring

Abstract

The invention relates to the technical field of fault detection, in particular to a motorcycle engine fault detection method based on vibration effect monitoring. First, first vibration characteristic data, state characteristic data and first position data of an engine are obtained from a sensor group; inputting the first position data and the first vibration characteristic data into a position adjusting module, and updating the first position data according to an output position adjusting coefficient to obtain second position data; then, inputting the state characteristic data and second vibration characteristic data corresponding to the second position data into a fault detection influence weight prediction module to obtain a fault detection influence weight; thirdly, classifying the second vibration characteristic data by using a multi-task fault diagnosis module to obtain fault characteristics and non-fault characteristics, and inputting the fault characteristics and the non-fault characteristics and a fault detection influence weight into the module for analysis to obtain a fault detection value for fault judgment; the method can improve the accuracy and effectiveness of motorcycle engine fault detection.

Description

Technical Field

[0001] The present invention relates to the technical field of fault detection, and particularly to a method for detecting motorcycle engine faults based on vibration effect monitoring. Background Technique

[0002] With the development of artificial intelligence, a corresponding relationship between vibration characteristics and engine state characteristics is established by using machine learning algorithms, and then an RVM classifier is used for fault diagnosis of samples to be measured; aiming at the interference problem during driving, the required fault analysis strategy is selected according to the fuzzy category of the time-domain waveform image segment.

[0003] Currently, the existing technologies for motorcycle engine fault detection methods still have deficiencies; on the one hand, the existing technologies do not consider that the sensor position will affect the complexity of interference characteristics in the data, resulting in a large deviation between the actual value and the reference value obtained, thus affecting fault analysis and evaluation; on the other hand, the existing technologies do not take monitoring parameters such as ambient temperature, rotational speed, and load as interference factors in the fault analysis process, thereby reducing the accuracy and effectiveness of motorcycle engine fault detection.

[0004] Therefore, a method for detecting motorcycle engine faults based on vibration effect monitoring is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for detecting motorcycle engine faults based on vibration effect monitoring. First, the first vibration characteristic data, state characteristic data, and first position data of the engine are obtained from the sensor group; the first position data and the first vibration characteristic data are input into the position adjustment module, and the first position data is updated according to the output position adjustment coefficient to obtain the second position data; then, the state characteristic data and the second vibration characteristic data corresponding to the second position data are input into the fault detection influence weight prediction module to obtain the fault detection influence weight; then, the multi-task fault diagnosis module is used to classify the second vibration characteristic data to obtain fault characteristics and non-fault characteristics, and they are input into the module together with the fault detection influence weight for analysis to obtain the fault detection value for fault judgment; this method can improve the accuracy and effectiveness of motorcycle engine fault detection.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: A method for detecting motorcycle engine faults based on vibration effect monitoring, including: Obtaining the first vibration characteristic data, state characteristic data, and first position data of the engine from the sensor group; Input the first position data and the corresponding first vibration characteristic data into a position adjustment module to obtain a position adjustment coefficient; use the position adjustment coefficient to update the first position data to obtain second position data; Construct a fault detection influence weight prediction module, and input the state characteristic data and the second vibration characteristic data corresponding to the second position data into the fault detection influence weight prediction module to obtain a fault detection influence weight; Use a multi-task fault diagnosis module to classify the second vibration characteristic data to obtain fault characteristics and non-fault characteristics; then, input the fault characteristics, the non-fault characteristics and the fault detection influence weight into the multi-task fault diagnosis module for analysis to obtain a fault detection value; the fault detection value is used for fault judgment; Compare the fault detection value with a fault detection threshold to obtain a fault detection result.

[0007] Further, the first vibration characteristic data and the second vibration characteristic data include fault vibration characteristic data and non-fault vibration characteristic data; the state characteristic data includes: rotational speed data, load data and temperature data.

[0008] Further, the process of inputting the first position data and the corresponding first vibration characteristic data into a position adjustment module to obtain a position adjustment coefficient; using the position adjustment coefficient to update the first position data to obtain second position data includes: Train the position adjustment module using historical position data and historical vibration characteristic data to obtain a pre-trained position adjustment module; Wherein, a position penalty term is introduced into the loss function of the position adjustment module to punish position data that exceeds the limit range; Input the first position data and the corresponding first vibration characteristic data into the pre-trained position adjustment module for prediction to obtain the position adjustment coefficient; Use the position adjustment coefficient to adjust the first position data to obtain the second position data.

[0009] Further, the process of inputting the state characteristic data and the second vibration characteristic data corresponding to the second position data into the fault detection influence weight prediction module to obtain a fault detection influence weight is: Construct the fault detection influence weight prediction module, including: an input processing layer, a feature extraction layer, an influence weight prediction layer and a fusion output layer; The input processing layer is used to receive the state feature data and the second vibration feature data, and perform frequency domain transformation and wavelet transformation on the second vibration feature data to obtain time domain feature data, frequency domain feature data, and wavelet feature data; The state feature data, the time domain feature data, the frequency domain feature data, and the wavelet feature data are input into the feature extraction layer for feature extraction to obtain state features, time domain features, frequency domain features, and wavelet features; The time domain features, the frequency domain features, and the wavelet features are respectively combined with the state features and input into corresponding sub-networks in the influence weight prediction layer for weight prediction to obtain time domain prediction weights, frequency domain prediction weights, and wavelet prediction weights; The fusion output layer is used to process the time domain prediction weights, the frequency domain prediction weights, and the wavelet prediction weights to obtain the fault detection influence weights.

[0010] Further, the multi-task fault diagnosis module includes: an input layer, a feature extraction and classification layer, a fault analysis layer, a weight correction layer, and an output layer; The input layer is used to transform the input data into data features; The feature extraction and classification layer is used to perform feature extraction and feature classification on the data features; The fault analysis layer is used to output a fault analysis result in combination with the classification features; The weight correction layer is used to correct the fault analysis result by using the fault detection influence weights; The output layer is used to map the corrected fault analysis result to the fault detection value.

[0011] Further, the process of using the multi-task fault diagnosis module to classify the second vibration feature data to obtain fault features and non-fault features; then, inputting the fault features, the non-fault features, and the fault detection influence weights into the multi-task fault diagnosis module for analysis to obtain the fault detection value includes: Using the input layer and the feature extraction and classification layer of the multi-task fault diagnosis module to process the second vibration feature data to obtain the fault features and the non-fault features; Inputting the fault features and the non-fault features into the fault analysis layer of the multi-task fault diagnosis module for processing to obtain fault analysis features; Inputting the fault analysis features and the fault detection influence weights into the weight correction layer of the multi-task fault diagnosis module for processing to obtain fault analysis corrected features; Transform the fault analysis and correction features using the output layer of the multi-task fault diagnosis module to obtain the fault detection value; wherein, the fault detection value is used for fault judgment.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention proposes a position adjustment method for adjusting the position of a sensor; this method learns the relationship between the sensor position and the complexity of vibration characteristic data in historical data by using a position adjustment module, and at the same time introduces a position penalty term in the loss function to limit the actual range of the output position data; the position adjustment coefficient obtained by the position adjustment module is used to adjust the real-time position, which can increase the proportion of fault vibration characteristic data in a complex environment, reduce interference, and is beneficial to subsequent data classification, thereby improving the accuracy and effectiveness of motorcycle engine fault detection.

[0013] 2. The present invention proposes a fault detection impact prediction method for obtaining the fault detection impact weight; this method uses a fault detection impact weight prediction module to divide the input data into time domain, frequency domain, and time-frequency domain features, and then uses an impact weight prediction layer to predict the impact degree of the real-time state on the vibration characteristics in different domains and fuse them to obtain the fault detection impact weight; the fault detection impact weight can effectively reflect the impact of the operating state on the detection during the fault detection process, so combining this weight can improve the accuracy and effectiveness of motorcycle engine fault detection.

[0014] 3. The present invention proposes a fault detection method for fault judgment; this method first classifies the vibration characteristic data into fault features and non-fault features, and then processes them using the fault analysis layer of the multi-task fault diagnosis module to obtain fault analysis features; the fault analysis features are corrected by combining the fault analysis features and the fault detection impact weight to obtain the fault detection value; this fault detection value can reflect the fault degree under the actual operating state, and using this value for fault judgment is beneficial to improving the accuracy and effectiveness of motorcycle engine fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic flow chart of a motorcycle engine fault detection method based on vibration effect monitoring according to the present invention; Figure 2 is a schematic structural diagram of the fault detection impact weight prediction module of the present invention; Figure 3 is a schematic flow chart of the multi-task fault diagnosis module of the present invention for obtaining the fault detection value. DETAILED DESCRIPTION OF THE INVENTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Please refer to Figures 1 to 3 , the present invention provides a method for detecting motorcycle engine faults based on vibration effect monitoring, and the technical solution is as follows: Embodiment 1: In order to reduce the operation and maintenance costs of a motorcycle engine, a certain company uses a method for detecting motorcycle engine faults based on vibration effect monitoring proposed by the present invention. The flow schematic of this method is as Figure 1 shown, and specifically includes: Obtain the first vibration characteristic data, state characteristic data, and first position data of the engine from the sensor group; Further, the first vibration characteristic data and the second vibration characteristic data include fault vibration characteristic data and non-fault vibration characteristic data; the state characteristic data includes: rotational speed data, load data, and temperature data; Further, the fault vibration characteristic data is generated by faults in components such as pistons, crankshafts, and valves; the non-fault vibration characteristic data includes data caused by periodic motions (such as piston reciprocation, crankshaft rotation, etc.) and combustion shocks under normal conditions, as well as interference data generated by the external environment (such as road conditions, aerodynamic noise, external machinery, etc.); Further, the position data is the three-dimensional coordinate data of the sensors installed on the device.

[0018] By adopting multi-dimensional data of the sensors, including: position data, state characteristic data, and vibration characteristic data, a data basis is provided for subsequent sensor position adjustment and fault detection, thereby ensuring the accuracy and effectiveness of motorcycle engine fault detection.

[0019] Input the first position data and the corresponding first vibration characteristic data into the position adjustment module to obtain a position adjustment coefficient; use the position adjustment coefficient to update the first position data to obtain second position data; Further, the process of inputting the first position data and the corresponding first vibration characteristic data into the position adjustment module to obtain a position adjustment coefficient and using the position adjustment coefficient to update the first position data to obtain second position data includes: Train the position adjustment module using historical position data and historical vibration characteristic data to obtain a pre-trained position adjustment module; Among them, a position penalty term is introduced into the loss function of the position adjustment module to penalize the position data that exceeds the limit range; Input the first position data and the corresponding first vibration feature data into the pre-trained position adjustment module for prediction to obtain the position adjustment coefficient; Adjust the first position data using the position adjustment coefficient to obtain the second position data; Furthermore, the position adjustment module adopts a GNN model, and the implementation process of this module is as follows: Each sensor position and each vibration feature are respectively regarded as a node; Connections are established between the sensor position nodes and the vibration feature nodes, between the vibration feature nodes, and between the sensor position nodes; The constructed graph structure is input into the GNN model to obtain the position adjustment coefficient; Furthermore, if the sensor position is a three-dimensional coordinate (x, y, z), then each position node has three coordinate features; Furthermore, the GNN model includes: graph convolutional network, graph attention network, gated graph neural network, etc.; Furthermore, the loss function of the position adjustment module includes: maximizing the proportion of fault feature energy, minimizing the information entropy, and the position penalty term; The position penalty term can be expressed as: ; Among them, is the position penalty term; represents the number of training samples of the position data; represents the maximization operation; represents the coordinate value of the j-axis of the i-th sample; represents the maximum coordinate value of the j-axis of the i-th sample; represents the minimum coordinate value of the j-axis of the i-th sample.

[0020] By using the position adjustment coefficient obtained by the position adjustment module to adjust the real-time position, the proportion of fault vibration feature data can be increased in a complex environment, interference can be reduced, which is beneficial to subsequent data classification, thereby improving the accuracy and effectiveness of motorcycle engine fault detection.

[0021] Construct a fault detection influence weight prediction module, input the state feature data and the second vibration feature data corresponding to the second position data into the fault detection influence weight prediction module to obtain the fault detection influence weight; Furthermore, the process of inputting the state feature data and the second vibration feature data corresponding to the second position data into the fault detection influence weight prediction module to obtain the fault detection influence weight is as follows: Construct a fault detection influence weight prediction module, and the structure of this module is as Figure 2As shown in the figure, it includes: an input processing layer, a feature extraction layer, an influence weight prediction layer, and a fusion output layer; The input processing layer is used to receive the state feature data and the second vibration feature data, and perform frequency domain transformation and wavelet transformation on the second vibration feature data to obtain time domain feature data, frequency domain feature data, and wavelet feature data; The state feature data, time domain feature data, frequency domain feature data, and wavelet feature data are input into the feature extraction layer for feature extraction to obtain state features, time domain features, frequency domain features, and wavelet features; The time domain features, frequency domain features, and wavelet features are respectively combined with the state features and input into the corresponding sub-networks in the influence weight prediction layer for weight prediction to obtain time domain prediction weights, frequency domain prediction weights, and wavelet prediction weights; The fusion output layer is used to process the time domain prediction weights, frequency domain prediction weights, and wavelet prediction weights to obtain the fault detection influence weights; Furthermore, the input processing layer directly extracts from the original data, performs frequency domain transformation using the fast Fourier transform, and performs decomposition using the continuous wavelet transform to obtain time domain feature data, frequency domain feature data, and wavelet feature data respectively; Furthermore, the feature extraction layer uses MLP to extract features from the state feature data, and uses autoencoders to extract features from the time domain feature data, frequency domain feature data, and wavelet feature data respectively; Furthermore, the influence weight prediction layer includes three sub-networks. The sub-networks implement fusing the state features with the time domain features, frequency domain features, and wavelet features respectively and performing weight prediction on the fused features of each; among them, the Transformer layer is used for feature fusion, and the MLP network is used for weight prediction; Furthermore, the fusion output layer uses the gating mechanism and the normalization function to output the expression of the fault detection influence weight as: ; Among them, is the fault detection influence weight; represents the softmax function; , and respectively represent the gating coefficients of the time domain, frequency domain, and wavelet, and the value range is between (0, 1); , and respectively represent the prediction weights of the time domain, frequency domain, and wavelet; Furthermore, the gating coefficients are obtained by successively using MLP and the Sigmoid function to process the prediction weights.

[0022] By using the fault detection impact weight prediction module to divide the input data into time-domain, frequency-domain, and time-frequency-domain features, and using the impact weight prediction layer to predict the impact degree of the real-time state on the vibration features in different domains and fuse them, the fault detection impact weight is obtained. This can effectively reflect the impact of the operating state on the detection during the fault detection process, and thus, combining this weight can improve the accuracy and effectiveness of motorcycle engine fault detection.

[0023] Use the multi-task fault diagnosis module to classify the second vibration feature data to obtain fault features and non-fault features. Then, input the fault features, non-fault features, and the fault detection impact weight into the multi-task fault diagnosis module for analysis to obtain the fault detection value. The fault detection value is used for fault judgment. Furthermore, the multi-task fault diagnosis module includes: an input layer, a feature extraction and classification layer, a fault analysis layer, a weight correction layer, and an output layer. The input layer is used to transform the input data into data features using a convolution kernel. The feature extraction and classification layer is used to perform feature extraction and feature classification on the data features. Furthermore, the feature extraction and classification layer adopts two parallel CNN classifiers, namely a fault classifier and a non-fault classifier, to identify fault features and non-fault features. The fault analysis layer is used to output a fault analysis result by combining the classification features. Furthermore, the fault analysis layer uses a ConvLSTM network to analyze the fault features, and at the same time uses the non-fault features as guiding features to assist the network in fault analysis. The weight correction layer is used to correct the fault analysis result using the fault detection impact weight. Furthermore, the weight correction layer weights the output features of the fault analysis layer using the fault detection impact weight, and performs a residual connection between the output features of the fault analysis layer and the weighted features. The output layer is used to map the corrected fault analysis result to a fault detection value.

[0024] By using the multi-task fault diagnosis module, tasks such as feature classification, feature analysis, and feature analysis can be achieved. Combining the fault analysis features and the fault detection impact weight to correct the fault analysis features, a fault detection value is obtained. This module can obtain the fault situation that conforms to the actual operating state, thereby improving the accuracy and effectiveness of motorcycle engine fault detection.

[0025] Furthermore, the process for the multi-task fault diagnosis module to obtain the fault detection value is as Figure 3 shown, specifically as follows: Use the input layer and the feature extraction and classification layer of the multi-task fault diagnosis module to process the second vibration feature data to obtain fault features and non-fault features. Input the fault features and non - fault features into the fault analysis layer of the multi - task fault diagnosis module for processing to obtain fault analysis features; Input the fault analysis features and the fault detection impact weights into the weight correction layer of the multi - task fault diagnosis module for processing to obtain fault analysis corrected features; Use the output layer of the multi - task fault diagnosis module to transform the fault analysis corrected features to obtain a fault detection value; wherein, the fault detection value is used for fault judgment.

[0026] Compare the fault detection value with the fault detection threshold to obtain a fault detection result.

[0027] To illustrate the fault detection value proposed by the present invention, three groups of data from different device models are randomly selected for fault detection tests, denoted as Test One, Test Two, and Test Three; the data includes vibration feature data and state feature data collected by sensors after adjusting the position; send each group of data into the pre - trained fault detection impact weight prediction module to obtain their respective fault detection impact weights; combine the specific process of the multi - task fault diagnosis module to obtain the fault detection value; set the fault detection threshold to 0.75, compare the fault detection values of each group with the fault detection threshold to obtain the fault detection test results, as shown in Table 1.

[0028] Table 1 Fault Detection Test Results Test Fault detection value Whether there is a fault Test 1 0.59 No Test 2 0.78 Yes Test 3 0.61 No By using the fault detection value obtained by the multi - task fault diagnosis module and combining with the fault detection impact weights, it can reflect the fault degree under the actual operating state. Using this value for fault judgment is beneficial to improving the accuracy and effectiveness of motorcycle engine fault detection.

[0029] This embodiment proposes a motorcycle engine fault detection method based on vibration effect monitoring. The method first obtains the first vibration feature data, state feature data, and first position data of the engine from the sensor group; input the first position data and the first vibration feature data into the position adjustment module, and update the first position data according to the output position adjustment coefficient to obtain the second position data; then, input the state feature data and the second vibration feature data corresponding to the second position data into the fault detection impact weight prediction module to obtain the fault detection impact weights; next, use the multi - task fault diagnosis module to classify the second vibration feature data to obtain fault features and non - fault features, and input them into the module for analysis together with the fault detection impact weights to obtain a fault detection value for fault judgment; this method can improve the accuracy and effectiveness of motorcycle engine fault detection.

[0030] Example 2: The present invention proposes a method for detecting motorcycle engine faults based on vibration effect monitoring. In order to further verify the effectiveness of the fault detection influence weight prediction module and the fault detection value acquisition process proposed by the present invention, the present invention conducts module ablation tests and process comparison tests for different modules and different acquisition processes respectively. The present invention selects two enterprises, A and B, to conduct the above two groups of tests respectively.

[0031] The present invention selects the historical data of enterprise A in the past 3 years as the data set of the module, where the data in the first and second years are used as the training set of the model, and the data in the third year are used as the verification set; among them, the historical data includes historical state data and historical vibration characteristic data; the sampling of the data set of the model refers to the following rules: taking a week as a unit, extracting 7 days of data per week; among them, 2 groups of data are extracted every morning, noon and evening every day; all data come from the same type of motorcycle engine.

[0032] The present invention inputs the training sets collected from enterprise A into different fault detection influence weight prediction modules for training respectively to obtain their respective pre-trained modules; then inputs the verification sets into each pre-trained module to obtain the fault detection influence weights of each module; then, compares the fault detection influence weights with the actual situation through manual verification to obtain the proportion of the fault detection influence weights of each module within a reasonable range.

[0033] Each test module is respectively: the fault detection influence weight prediction module proposed by the present invention, denoted as module one; the [description is missing here], denoted as model two; removing the gating mechanism in the fusion output layer and directly combining and normalizing the prediction weights of time domain, frequency domain and wavelet, denoted as module two; only retaining the time domain data for processing, that is, both the feature extraction layer and the influence weight prediction layer have only one network and the output layer has no fusion part, denoted as module three.

[0034] The test results of module effectiveness are shown in Table 2.

[0035] Table 2 Test results of module effectiveness Test module Proportion within a reasonable range Model 1 91.08% Model 2 89.37% Model 3 85.91% It can be seen from the results in Table 2 that the fault detection influence weight prediction module proposed by the present invention has better test results than those of other modules in terms of the effectiveness test results; thus, it can be shown that the module proposed by the present invention combines multi-domain feature prediction and gating fusion to be able to output accurate fault detection influence weights, further improving the accuracy and effectiveness of motorcycle engine fault detection.

[0036] To further test the effectiveness of the fault detection value acquisition process, this embodiment collected the historical data of Company B in the past year, according to the same data sampling rules, and then divided the sampled data set into three groups of data for testing, which were respectively recorded as: Test Sample One, Test Sample Two, and Test Sample Three; the test data of each group were processed according to Process One and Process Two respectively to obtain their respective fault detection values; among them, Process One is the process of using the multi-task fault diagnosis module proposed by the present invention to obtain the fault detection value; Process Two is to remove the feature classification operation and the weight correction layer, and directly input the vibration feature data into the fault analysis layer for output; finally, the rationality of the fault detection values obtained by each process was verified through manual verification; the test results of the process effectiveness are shown in Table 3; Table 3 Test Results of Process Effectiveness Test data Proportion within a reasonable range (Process 1) Proportion within a reasonable range (Process 2) Test sample 1 90.71% 88.64% Test sample 2 89.88% 87.43% Test sample 3 90.29% 87.98% It can be seen from the results in Table 3 that the effectiveness test results obtained by using Process One, that is, the fault detection value acquisition scheme proposed by the present invention, are better than those obtained by using Process Two, which shows the necessity of correcting the fault analysis by combining the fault detection influence weight, and can improve the accuracy and effectiveness of motorcycle engine fault detection.

[0037] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A motorcycle engine fault detection method based on vibration effect monitoring, characterized in that: include: Acquire first vibration characteristic data, state characteristic data and first position data of the engine from the sensor group; Inputting the first position data and the corresponding first vibration characteristic data into a position adjustment module to obtain a position adjustment coefficient; updating the first position data using the position adjustment coefficient to obtain second position data; Constructing a fault detection influence weight prediction module, inputting the state feature data and the second vibration feature data corresponding to the second position data into the fault detection influence weight prediction module to obtain a fault detection influence weight; The second vibration characteristic data is classified by using a multi-task fault diagnosis module to obtain fault characteristics and non-fault characteristics; then, the fault characteristics, the non-fault characteristics and the fault detection influence weight are input into the multi-task fault diagnosis module for analysis to obtain a fault detection value; the fault detection value is used for fault judgment; The fault detection value is compared with a fault detection threshold to obtain a fault detection result.

2. A motorcycle engine fault detection method based on vibration effect monitoring according to claim 1, characterized in that: The first vibration characteristic data and the second vibration characteristic data include fault vibration characteristic data and non-fault vibration characteristic data; The state characteristic data includes: rotation speed data, load data and temperature data.

3. A motorcycle engine fault detection method based on vibration effect monitoring according to claim 1, characterized in that: The process of inputting the first position data and the corresponding first vibration characteristic data into a position adjustment module to obtain a position adjustment coefficient; and updating the first position data using the position adjustment coefficient to obtain second position data includes: The position adjustment module is trained using the historical position data and the historical vibration characteristic data to obtain a pre-trained position adjustment module; Wherein, a position penalty term is introduced into the loss function of the position adjustment module to penalize the position data exceeding the limit range; Inputting the first position data and the corresponding first vibration characteristic data into the pre-trained position adjustment module for prediction to obtain the position adjustment coefficient; The first position data is adjusted using the position adjustment coefficient to obtain the second position data.

4. A motorcycle engine fault detection method based on vibration effect monitoring according to claim 1, characterized in that: The state feature data and the second vibration feature data corresponding to the second position data are input into the fault detection influence weight prediction module to obtain the fault detection influence weight as follows: Constructing the fault detection impact weight prediction module, including: an input processing layer, a feature extraction layer, an impact weight prediction layer and a fusion output layer; Using the input processing layer to receive the state characteristic data and the second vibration characteristic data, and performing frequency domain transformation and wavelet transformation on the second vibration characteristic data to obtain time domain characteristic data, frequency domain characteristic data and wavelet characteristic data; Inputting the state feature data, the time domain feature data, the frequency domain feature data and the wavelet feature data into the feature extraction layer for feature extraction to obtain state features, time domain features, frequency domain features and wavelet features; The time domain feature, the frequency domain feature and the wavelet feature are respectively combined with the state feature, and are respectively input into the corresponding sub-network in the influence weight prediction layer for weight prediction to obtain the time domain prediction weight, the frequency domain prediction weight and the wavelet prediction weight; The time domain prediction weight, the frequency domain prediction weight and the wavelet prediction weight are processed by using the fusion output layer to obtain the fault detection influence weight.

5. A motorcycle engine fault detection method based on vibration effect monitoring according to claim 1, characterized in that: The multi-task fault diagnosis module includes: an input layer, a feature extraction and classification layer, a fault analysis layer, a weight correction layer and an output layer; The input layer is used to transform input data into data features; The feature extraction and classification layer is used to extract and classify the data features; The fault analysis layer is used to output fault analysis results in combination with classification features; The weight correction layer is used to correct the fault analysis result using the fault detection impact weight; The output layer is used to map the corrected fault analysis result into the fault detection value.

6. A motorcycle engine fault detection method based on vibration effect monitoring according to claim 1, characterized in that: The process of classifying the second vibration characteristic data by using a multi-task fault diagnosis module to obtain fault characteristics and non-fault characteristics; then, inputting the fault characteristics, the non-fault characteristics and the fault detection influence weight into the multi-task fault diagnosis module for analysis to obtain the fault detection value includes: Processing the second vibration feature data using the input layer and feature extraction and classification layer of the multi-task fault diagnosis module to obtain the fault feature and the non-fault feature; Inputting the fault features and the non-fault features into the fault analysis layer of the multi-task fault diagnosis module for processing to obtain fault analysis features; Inputting the fault analysis feature and the fault detection influence weight into the weight correction layer of the multi-task fault diagnosis module for processing to obtain a fault analysis correction feature; The fault analysis correction feature is transformed using the output layer of the multi-task fault diagnosis module to obtain the fault detection value; wherein the fault detection value is used for fault judgment.

Images