Internal combustion engine fault diagnosis method, device, equipment and medium
Through multi-source signal fusion and image conversion technology, combined with Markov transfer field and generative adversarial network, the problem of low accuracy and efficiency of traditional fault diagnosis methods in complex environments is solved, and high accuracy of internal combustion engine failures is achieved and Efficient diagnosis.
Patent Information
- Application Number
- CN202411067590.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-22
AI Technical Summary
Traditional internal combustion engine fault diagnosis methods perform poorly in complex and diverse fault environments, and the fault diagnosis accuracy and efficiency are low, making it difficult to identify and deal with abnormal vibration problems of internal combustion engines in a timely manner.
By acquiring a multi-source vibration signal set, signal fusion and image conversion are performed, fault type identification is used using a trained fault diagnosis model, combined with Markov transfer field and generation adversarial network technology, the time dependence and pattern of vibration signals are captured, and the time dependence and pattern of vibration signals are improved. Accuracy and efficiency of fault diagnosis.
It improves the comprehensiveness and accuracy of fault diagnosis, is suitable for complex and changeable fault scenarios, reduces the uncertainty brought by a single signal source, enhances the ability to identify and diagnose faults, and can effectively identify and classify before the fault.
Smart Images

Figure CN119004206A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of internal combustion engine fault diagnosis, and in particular to an internal combustion engine fault diagnosis method, device, equipment and medium. Background Art
[0002] Internal combustion engines are critical industrial equipment that often face challenges due to a variety of fault problems. During its operation, unbalanced external forces, wear of key components, bearing failures, and damage to seals may cause abnormal vibrations in the internal combustion engine, which will shorten the service life of the internal combustion engine. Therefore, timely identification and handling of these faults, real-time monitoring and early warning of multi-source vibration signals are important strategies to reduce downtime, reduce maintenance costs, and improve equipment operation reliability and production efficiency. Traditional fault diagnosis methods may perform poorly in complex and diverse fault environments, and have low fault diagnosis accuracy and efficiency. Summary of the invention
[0003] The purpose of this application is to provide an internal combustion engine fault diagnosis method, device, equipment and medium, which can improve the accuracy and efficiency of fault diagnosis.
[0004] To achieve the above objectives, this application provides the following solutions:
[0005] In a first aspect, the present application provides a method for diagnosing a fault in an internal combustion engine, comprising:
[0006] Acquire a multi-source vibration signal set of a target internal combustion engine; the multi-source vibration signal set includes a plurality of original vibration signals; the original vibration signals in the multi-source vibration signal set are obtained by measuring the target internal combustion engine with different sensors;
[0007] fusing the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal;
[0008] Performing image conversion on the fused vibration signal to obtain a target two-dimensional image;
[0009] The target two-dimensional image is input into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine; the trained fault diagnosis model is a model trained with the sample two-dimensional image as input and the fault type corresponding to the sample two-dimensional image as a label.
[0010] In a second aspect, the present application provides an internal combustion engine fault diagnosis device, comprising:
[0011] A signal acquisition module is used to: acquire a multi-source vibration signal set of a target internal combustion engine; the multi-source vibration signal set includes a plurality of original vibration signals; the original vibration signals in the multi-source vibration signal set are obtained by measuring the target internal combustion engine with different sensors;
[0012] A fusion module is used to: fuse the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal;
[0013] An image conversion module is used to: perform image conversion on the fused vibration signal to obtain a target two-dimensional image;
[0014] The diagnostic module is used to: input the target two-dimensional image into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine; the trained fault diagnosis model is a model trained with the sample two-dimensional image as input and the fault type corresponding to the sample two-dimensional image as a label.
[0015] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned internal combustion engine fault diagnosis method.
[0016] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned internal combustion engine fault diagnosis method when executed by a processor.
[0017] According to the specific embodiments provided in this application, this application discloses the following technical effects:
[0018] The present application provides a method, apparatus, equipment and medium for diagnosing faults in an internal combustion engine. By fusing multi-source signals, the information provided by different sensors can be fully utilized to obtain richer feature descriptions, thereby improving the comprehensiveness of fault diagnosis and reducing the uncertainty caused by a single signal source; the original vibration signals in the multi-source vibration signal set are fused to obtain a fused vibration signal, which is converted into a target two-dimensional image by fusing the vibration signal to capture the time dependency and pattern in the vibration signal sequence, and finally a trained fault diagnosis model is used to perform fault diagnosis on the target two-dimensional image containing rich features and time dependency, thereby improving the accuracy and efficiency of fault diagnosis and being suitable for complex and changeable fault scenarios. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0020] Figure 1 This is an application environment diagram of an internal combustion engine fault diagnosis method in one embodiment of the present application;
[0021] Figure 2 A schematic diagram of a flow chart of a method for diagnosing a fault in an internal combustion engine provided in one embodiment of the present application;
[0022] Figure 3 A schematic diagram of a model training framework provided in one embodiment of the present application;
[0023] Figure 4 A schematic diagram of bearing fault types provided in an embodiment of the present application;
[0024] Figure 5 A schematic diagram of the accuracy visualization process provided by an embodiment of the present application;
[0025] Figure 6 A schematic diagram of a multi-source signal data fusion process provided in an embodiment of the present application;
[0026] Figure 7 A schematic diagram of a two-dimensional image conversion process provided by an embodiment of the present application;
[0027] Figure 8 A schematic diagram of a fault diagnosis model training process provided in one embodiment of the present application;
[0028] Fig. 9 A schematic diagram of functional modules of an internal combustion engine fault diagnosis device provided in one embodiment of the present application;
[0029] Fig.10 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION
[0030] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0031] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0032] The internal combustion engine fault diagnosis method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, or it can be integrated on the server 104, or it can be placed on the cloud or other servers. The terminal 102 can send a multi-source vibration signal set to the server 104. After the server 104 receives the multi-source vibration signal set, for the multi-source vibration signal set, the server 104 fuses the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal, performs image conversion on the fused vibration signal to obtain a target two-dimensional image, and inputs the target two-dimensional image into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine. The server 104 can feedback the obtained fault type of the target internal combustion engine to the terminal 102. In addition, in some embodiments, the internal combustion engine fault diagnosis method can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform internal combustion engine fault diagnosis on the multi-source vibration signal set, or the server 104 can obtain the multi-source vibration signal set from the data storage system and perform internal combustion engine fault diagnosis on the multi-source vibration signal set.
[0033] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, and tablet computers. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.
[0034] In an exemplary embodiment, Figure 2 As shown, a method for diagnosing internal combustion engine faults is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate the method, which includes the following steps 201 to 204. Among them:
[0035] Step 201: Acquire a multi-source vibration signal set of a target internal combustion engine; the multi-source vibration signal set includes a plurality of original vibration signals; the original vibration signals in the multi-source vibration signal set are obtained by measuring the target internal combustion engine with different sensors.
[0036] Step 202: fusing the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal.
[0037] Step 203: performing image conversion on the fused vibration signal to obtain a target two-dimensional image.
[0038] Step 204: Input the target two-dimensional image into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine; the trained fault diagnosis model is a model trained with the sample two-dimensional image as input and the sample fault type corresponding to the sample two-dimensional image as a label.
[0039] By implementing the above steps 201 to 204, the information provided by different sensors can be fully utilized by fusing multi-source signals to obtain richer feature descriptions, thereby improving the comprehensiveness of fault diagnosis and reducing the uncertainty caused by a single signal source; the original vibration signals in the multi-source vibration signal set are fused to obtain a fused vibration signal, which is converted into a target two-dimensional image by fusing the vibration signal to capture the time dependency and pattern in the vibration signal sequence, and finally the target two-dimensional image containing rich features and time dependency is fault diagnosed by a trained fault diagnosis model, thereby improving the accuracy and efficiency of fault diagnosis. It is suitable for complex and changeable fault scenarios. The multi-source signal fusion, Markov transition field (MTF) and generative adversarial network (GAN) technology are combined and applied to the internal combustion engine fault diagnosis. First, the vibration signal data from multiple signal sources are collected, and then multi-source signal fusion is performed to reduce the uncertainty caused by a single signal source and provide a more reliable decision-making basis. Specifically, the vibration signal data is collected from multiple sensors, features are sorted and extracted, and feature data sets are organized and fused to finally form a feature vector containing all sensor features; then the Markov transition field is used to convert the time series vibration signal data into a two-dimensional image output to capture the transition relationship between different states in the vibration signal data, thereby enhancing the ability of pattern recognition and fault diagnosis, and dividing it into a training set and a test set; next, a GAN model is constructed and trained with the training set. During the training process, some optimization algorithms can be used to adjust the parameters of the model to improve the prediction effect. Finally, the trained model is used to predict the test set, and the accuracy and reliability of the prediction results are evaluated. Based on the above method, accurate identification and classification of faults at different locations in the field of internal combustion engines can be achieved. Finally, the classification results are visualized using the confusion matrix. The internal combustion engine fault diagnosis method provided by the present application can be applied to fault identification and diagnosis at different positions of the internal combustion engine. It has certain universality, practicality and promotion value in the field of internal combustion engines, and can be widely used in engineering practices in the field of internal combustion engines. In addition, the present application uses the principal component analysis method to fuse the vibration signals of multiple sensors, and combines the Markov transfer field and the generative adversarial network model to perform fault identification and diagnosis on the relevant components of the internal combustion engine, which improves the accuracy and efficiency of fault identification and diagnosis, effectively solves the uncertainty and error of a single signal, and has certain engineering practice value; the present application is more sensitive to fault data, and it is easier to discover and identify faults in the early stage of faults, and can effectively detect the working state of the internal combustion engine, thereby effectively avoiding losses caused by processing errors, improper operation and other factors.
[0040] Step 201 obtains original vibration signals of multiple sensors at different fault locations of a component of an internal combustion engine.
[0041] In another exemplary embodiment of the present application, step 202 specifically includes: using the principal component analysis (PCA) method to fuse the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal, such as Figure 6 As shown, the above steps are replaced by the following steps 301 to 303:
[0042] Step 301: performing standardization processing on each original vibration signal to obtain standard vibration data corresponding to each original vibration signal.
[0043] Step 302: Calculate the covariance matrix based on the standard vibration data corresponding to all original vibration signals.
[0044] Step 303: Perform eigenvalue decomposition on the covariance matrix to obtain a number of eigenvalues and an eigenvector corresponding to each eigenvalue; the eigenvalue is used to represent the variance of the data in the direction of the corresponding eigenvector; the eigenvector is used to represent the direction of the new coordinate axis.
[0045] Step 304: Arrange the eigenvalues in descending order, select the first k eigenvalues and the eigenvectors corresponding to each eigenvalue as the principal components, and construct a feature matrix with the selected eigenvectors; k is less than the total amount of eigenvalue data.
[0046] Step 305: Based on the feature matrix, perform data conversion on the original vibration signal to obtain a fused vibration signal.
[0047] Multi-source signal data are fused using principal component analysis. PCA is a statistical method for dimensionality reduction and feature extraction. It converts data from the original coordinate system to a new coordinate system through linear transformation so that the variance of the data in the new coordinate system is maximized. Each axis in the new coordinate system (i.e., principal component) is a linear combination of the original coordinates, and these axes are orthogonal.
[0048] Step 301 is to ensure that each variable has the same weight. The data needs to be standardized, that is, the mean is subtracted and then divided by the standard deviation, so that the mean of each variable is 0 and the standard deviation is 1. The formula is as follows:
[0049]
[0050] Among them, X is the original vibration signal, X standardized is the standard vibration data corresponding to the original vibration signal, μ is the mean of the data, and σ is the standard deviation of the data.
[0051] Step 302 calculates the covariance matrix C of step 301, and the formula is as follows:
[0052]
[0053] Here, n is the number of samples.
[0054] Step 303 performs eigenvalue decomposition on the covariance matrix C obtained in step 302 to obtain eigenvalues and eigenvectors. The eigenvalue represents the variance of the data in the direction of the corresponding eigenvector. The eigenvector is the direction of the new coordinate axis (i.e., the principal component), and the formula is as follows:
[0055] Cv i =λ i v i (3);
[0056] Among them, λ i is the ith eigenvalue, v i is the eigenvector corresponding to the i-th eigenvalue.
[0057] Step 304 arranges the eigenvalues in descending order, selects the first k largest eigenvalues and their corresponding eigenvectors as principal components, and records the characteristic matrix composed of the selected eigenvectors as V.
[0058] Step 305 projects the original vibration signal X onto the new coordinate system to obtain a fused vibration signal, i.e., the dimensionally reduced data Z:
[0059] Z=XV (4).
[0060] In another exemplary embodiment of the present application, step 203 specifically includes: using a Markov transfer field to perform image conversion on the fused vibration signal to obtain a target two-dimensional image.
[0061] MTF is a technique for converting time series data into images, aiming to capture the time dependencies and patterns in the sequence. It uses the state transition matrix of the Markov chain to represent the transition probabilities between each moment in the time series, and then represents these probability information in the form of a two-dimensional image. The specific steps of MTF are as follows:
[0062] Time series discretization: discretize time series data into a finite number of states. Common methods include equally spaced discretization or clustering-based discretization.
[0063] Calculate the transfer matrix: Based on the discretized state, calculate the transfer matrix of the Markov chain, which represents the probability of transitioning from one state to another.
[0064] Generate transition image: Map the transition probabilities in the transition matrix onto a two-dimensional plane to form an image. The value of each pixel represents the probability of transition between specific time steps.
[0065] Smoothing: In order to eliminate noise, the generated image can be smoothed to enhance the visualization effect of the image.
[0066] After converting the time series data into images, deep learning technology is used to extract features to improve the accuracy of classification.
[0067] like Figure 7 As shown, the above step of "using the Markov transfer field to perform image conversion on the fused vibration signal to obtain a target two-dimensional image" is replaced by the following steps 401 to 404:
[0068] Step 401: read the fused vibration signal in step 305, and set initialization parameters, including data size, window overlap rate, number of bins, number of categories, etc. The parameters can be specifically set as follows: data size is set to 802; window overlap rate is set to 0; number of bins is set to 5; number of categories is set to 13.
[0069] Step 402: Segment the fused vibration signal to obtain a plurality of window data. Specifically, segment the data into a plurality of small windows according to a set window size and overlap rate, and store the data in each window using a cell array.
[0070] Step 403: Perform image conversion on each window data to obtain a two-dimensional image corresponding to the window data.
[0071] Step 404: After obtaining the two-dimensional image corresponding to the window data, the method further includes: cutting the two-dimensional image corresponding to the window data to obtain a cut image, and saving the cut image to a target directory.
[0072] Each of the above cropped images is used as a target two-dimensional image, and the obtained target two-dimensional images are respectively input into the trained fault diagnosis model to obtain the internal combustion engine fault type corresponding to each target two-dimensional image.
[0073] The entire model training framework of this application is as follows Figure 3As shown in FIG. 1 , it specifically includes: data collection, obtaining vibration signal data of multiple sensors at different fault locations of a certain component of an internal combustion engine; comprehensively processing multiple signals from different sensors or data sources to obtain a new vibration signal to improve the overall performance and reliability of the data; using MTF to process the one-dimensional vibration signal data obtained in step 2 to obtain a two-dimensional image, and dividing it into a training set and a test set; using the training set to train the GAN model to obtain a trained discriminator and a trained generator, naming the trained discriminator a GAN discriminator model, using the GAN discriminator model as a trained fault diagnosis model, and using the test set to verify the trained fault diagnosis model; using a confusion matrix to visualize the classification results.
[0074] The entire model training process is introduced using the bearing failure dataset published by Case Western Reserve University:
[0075] S1.1: Input fault vibration signal. The public bearing fault dataset is input. The dataset consists of vibration signals from three different sensors and the faults are divided into 13 categories.
[0076] S1.2: Standardization. The above dataset is standardized to ensure that each variable has the same weight.
[0077] S1.3: Calculate the covariance matrix. Calculate the covariance matrix of the standardized data set.
[0078] S1.4: Eigenvalue decomposition. Perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues and eigenvectors.
[0079] S1.5: Sorting and selecting principal components. Arrange the eigenvalues in descending order and select the first k largest eigenvalues and their corresponding eigenvectors as principal components.
[0080] S1.6: Data transformation. Projecting the original data onto the new coordinate system will yield the fused data after dimensionality reduction.
[0081] S2.1: Initialization parameters. Read the fused data and set the initialization parameters, including setting the data size to 802, the window overlap rate to 0, the number of bins to 5, and the number of categories to 13.
[0082] S2.2: Fill the cell array. Divide the data into small windows and store them in a cell array.
[0083] S2.3: Generate a two-dimensional image. Convert the one-dimensional vibration signal data into a two-dimensional image for output and save it to a specified directory.
[0084] S2.4: Cropping and processing. The above two-dimensional images are cropped to remove the boundaries that are not related to the vibration signal data to reduce the image size for subsequent processing. The processed images contain 13 types of faults, a total of 1963 images. The 13 types of faults are as follows: Figure 4 shown.
[0085] like Figure 8 As shown, the fault diagnosis model training process includes the following steps 501 to 504.
[0086] Step 501: Set the ratio of training set to test set = 8:2, that is, the number of images in the training set is 1573 and the number of images in the test set is 390. And set the relevant parameters of the model such as the number of convolutional layers, pooling layers, flattening layers, fully connected layers, activation functions, and Dropout.
[0087] Step 502: Create a GAN model.
[0088] The trained fault diagnosis model is obtained by training a generative adversarial network; a generative adversarial network is a deep learning model that consists of a generator and a discriminator. The goal of the generator is to generate images that are difficult to distinguish from real pictures, while the goal of the discriminator is to distinguish between real data and generated data. Through adversarial training, the two networks continuously improve their respective capabilities.
[0089] The trained fault diagnosis model is a GAN discriminator model; the GAN discriminator model includes a first convolutional layer, a second convolutional layer, a maximum pooling layer and a fully connected layer connected in sequence. The main function of the GAN discriminator model is to distinguish between fake samples and real samples generated by the generator, and is used for image classification tasks. The specific structure of the GAN discriminator model extracts image features through two convolutional layers and a maximum pooling layer, connects to the fully connected layer through a flattening layer for classification, and finally outputs the category probability. The model uses LeakyReLU activation function and Dropout to prevent overfitting, and implements multi-class classification through the softmax activation function. The first convolutional layer consists of 64 3*3 convolution kernels; the second convolutional layer consists of 128 3*3 convolution kernels; the maximum pooling layer uses a 2×2 pooling window; the fully connected layer contains 512 neurons, and the activation function is ReLU; the activation layer is LeakyReLU; the output layer activation function is softmax, and the output dimension is the number of categories; the Dropout layer randomly discards 30% of the neurons to prevent overfitting; the optimizer uses Adam, uses the classification cross entropy loss function, and uses accuracy to evaluate model performance indicators.
[0090] Step 503: training the model, using the training set to train the GAN model, and using the test set to verify, wherein the number of training steps is set to 100.
[0091] Step 504: Evaluate the model, evaluate the performance of the model using accuracy, etc., and represent it using the elements in the confusion matrix, draw the confusion matrix, and use the confusion matrix to visualize the classification results, such as Figure 5 shown.
[0092] Accuracy is an important indicator to measure the performance of a classification model. It indicates the proportion of samples correctly classified by the model to the total number of samples. The formula for accuracy is as follows:
[0093]
[0094] The present application also provides an application scenario, which applies the above-mentioned internal combustion engine fault diagnosis method. Specifically: the internal combustion engine fault diagnosis method provided in this embodiment can be applied in a content distribution scenario. The internal distribution scenario includes a signal measurement link and a fault diagnosis link; the multi-source vibration signal set enters the fault diagnosis link from the signal measurement link to obtain the corresponding fault type. The internal combustion engine fault diagnosis method provided in this embodiment belongs to the fault diagnosis link. Specifically, in the fault diagnosis link process for the multi-source vibration signal set, the original vibration signals in the multi-source vibration signal set can be fused to obtain a fused vibration signal, and the fused vibration signal can be converted into an image to obtain a target two-dimensional image, and the target two-dimensional image can be input into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine.
[0095] Based on the same inventive concept, the embodiment of the present application also provides an internal combustion engine fault diagnosis device for implementing the internal combustion engine fault diagnosis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the internal combustion engine fault diagnosis device provided below can refer to the limitations of the internal combustion engine fault diagnosis method above, and will not be repeated here.
[0096] In an exemplary embodiment, Fig. 9 As shown, a device for diagnosing faults of an internal combustion engine is provided, comprising:
[0097] The signal acquisition module T1 is used to: acquire a multi-source vibration signal set of a target internal combustion engine; the multi-source vibration signal set includes a plurality of original vibration signals; the original vibration signals in the multi-source vibration signal set are obtained by measuring the target internal combustion engine with different sensors.
[0098] The fusion module T2 is used to fuse the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal.
[0099] The image conversion module T3 is used to perform image conversion on the fused vibration signal to obtain a target two-dimensional image.
[0100] The diagnostic module T4 is used to: input the target two-dimensional image into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine; the trained fault diagnosis model is a model trained with the sample two-dimensional image as input and the fault type corresponding to the sample two-dimensional image as a label.
[0101] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Fig.10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store a multi-source vibration signal set. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for diagnosing internal combustion engine faults is implemented.
[0102] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0103] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.
[0104] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0105] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0106] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0107] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.
[0108] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0109] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A method for diagnosing internal combustion engine faults, It is characterized in that The internal combustion engine fault diagnosis method comprises: Acquire a multi-source vibration signal set of a target internal combustion engine; the multi-source vibration signal set includes a plurality of original vibration signals; the original vibration signals in the multi-source vibration signal set are obtained by measuring the target internal combustion engine with different sensors; fusing the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal; Performing image conversion on the fused vibration signal to obtain a target two-dimensional image; The target two-dimensional image is input into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine; the trained fault diagnosis model is a model trained with the sample two-dimensional image as input and the sample fault type corresponding to the sample two-dimensional image as a label.
2. The internal combustion engine fault diagnosis method according to claim 1, It is characterized in that Fusing the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal specifically includes: The original vibration signals in the multi-source vibration signal set are fused using the principal component analysis method to obtain a fused vibration signal.
3. The internal combustion engine fault diagnosis method according to claim 2, It is characterized in that The original vibration signals in the multi-source vibration signal set are fused by using the principal component analysis method to obtain a fused vibration signal, which specifically includes: Performing standardization processing on each original vibration signal to obtain standard vibration data corresponding to each original vibration signal; The covariance matrix is calculated based on the standard vibration data corresponding to all the original vibration signals; Perform eigenvalue decomposition on the covariance matrix to obtain several eigenvalues and eigenvectors corresponding to each eigenvalue; the eigenvalue is used to represent the variance of the data in the direction of the corresponding eigenvector; the eigenvector is used to represent the direction of the new coordinate axis; Arrange the eigenvalues in descending order, select the first k eigenvalues and the eigenvectors corresponding to each eigenvalue as the principal components, and construct a feature matrix with the selected eigenvectors; Based on the feature matrix, the original vibration signal is converted into a fused vibration signal.
4. The internal combustion engine fault diagnosis method according to claim 1, It is characterized in that Performing image conversion on the fused vibration signal to obtain a target two-dimensional image specifically includes: The Markov transfer field is used to perform image conversion on the fused vibration signal to obtain a target two-dimensional image.
5. The internal combustion engine fault diagnosis method according to claim 4, It is characterized in that Using the Markov transfer field, the fused vibration signal is image-converted to obtain a target two-dimensional image, specifically including: Segmenting the fused vibration signal to obtain a number of window data; Perform image conversion on each window data to obtain a two-dimensional image corresponding to the window data.
6. The internal combustion engine fault diagnosis method according to claim 5, It is characterized in that After obtaining the two-dimensional image corresponding to the window data, it also includes: The two-dimensional image corresponding to the window data is cropped to obtain a cropped image.
7. The internal combustion engine fault diagnosis method according to claim 1, It is characterized in that The trained fault diagnosis model is obtained by training a generative adversarial network; the trained fault diagnosis model is a GAN discriminator model; the GAN discriminator model includes a first convolutional layer, a second convolutional layer, a maximum pooling layer and a fully connected layer connected in sequence.
8. An internal combustion engine fault diagnosis device, It is characterized in that The internal combustion engine fault diagnosis device comprises: A signal acquisition module is used to: acquire a multi-source vibration signal set of a target internal combustion engine; the multi-source vibration signal set includes a plurality of original vibration signals; the original vibration signals in the multi-source vibration signal set are obtained by measuring the target internal combustion engine with different sensors; A fusion module is used to: fuse the original vibration signals in the multi-source vibration signal set to obtain a fused vibration signal; An image conversion module is used to: perform image conversion on the fused vibration signal to obtain a target two-dimensional image; The diagnostic module is used to: input the target two-dimensional image into a trained fault diagnosis model to obtain the fault type of the target internal combustion engine; the trained fault diagnosis model is a model trained with the sample two-dimensional image as input and the fault type corresponding to the sample two-dimensional image as a label.
9. A computer device, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the internal combustion engine fault diagnosis method according to any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the internal combustion engine fault diagnosis method according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Mechanical fault diagnosis method based on multi-sensor multivariate data fusion
CN112906158A
Diesel engine health state assessment method, device and equipment based on digital twinning
CN113537373A
Bearing fault diagnosis method, system and equipment and storage medium
CN116610993A
Centrifugal pump fault diagnosis method, system, medium and equipment
CN117803579A
Cited By
Rolling bearing fault diagnosis method and system based on multi-source vibration signals and medium
CN119643147A
