Vehicle fault early warning method and system based on convolutional neural network
Through the vehicle fault warning method based on convolutional neural network, preprocessing and deep learning analysis is used to use vehicle sensor data for preprocessing and deep learning analysis, the traditional fault detection problems are solved, real-time monitoring and early warning are realized, and the safety of the vehicle is improved and reliability.
Patent Information
- Application Number
- CN202411187492.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-06
AI Technical Summary
Traditional vehicle fault detection methods rely on manual inspection and empirical judgment, are inefficient and prone to missed inspections, and cannot effectively monitor the operating status of the vehicle, resulting in low accuracy of fault detection and poor generalization capabilities.
Using a vehicle fault warning method based on convolutional neural network, multiple sensor data are obtained through on-board sensors, preprocessed and converted into image data, and the image features are analyzed and compared with the deep sample features in the fault library to judge Whether the vehicle has a risk of failure and send out early warning signals in a timely manner.
Real-time monitoring of vehicle operating status is achieved, the accuracy and reliability of fault detection is improved, manual intervention is reduced, the intelligence level of fault detection is enhanced, the false alarm rate is reduced, and the safety of the vehicle is ensured during operation. And promptly remind the driver to take measures to avoid accidents.
Smart Images

Figure CN119091158A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of vehicle fault detection and early warning, and specifically relates to a vehicle fault early warning method and system based on convolutional neural network. Background Art
[0002] With the development of artificial intelligence technology, convolutional neural networks have achieved remarkable results in the field of computer vision. In the field of vehicle fault warning, traditional methods mainly rely on manual feature extraction, which has problems such as low accuracy and poor generalization ability. Convolutional neural networks can automatically learn features and have high accuracy and generalization ability. With the rapid development of the automotive industry, the electronic systems of vehicles are becoming increasingly complex, and the probability of failure is also increasing.
[0003] The existing method includes: performing edge detection on lane lines, matching with lane line models to determine the road area of interest, then collecting road videos, marking vehicle targets therein, creating a vehicle data set, and then designing a convolutional neural network, using the vehicle data set to train the detector, thereby achieving vehicle detection; constructing a convolutional neural network for lane detection, the convolutional neural network includes N structures; using a first training image set to train the constructed convolutional neural network for lane detection to obtain a trained convolutional neural network; inputting an image set taken when weather conditions are good into the trained convolutional neural network to obtain a lane departure warning result.
[0004] However, traditional fault detection methods mostly rely on manual inspection and empirical judgment, which are inefficient and prone to missed detection. Data-driven intelligent fault warning systems have gradually become a research hotspot. As an effective deep learning model, convolutional neural network (CNN) can automatically extract data features and is suitable for image processing and pattern recognition. Therefore, vehicle fault warning systems based on CNN have broad application prospects. Summary of the invention
[0005] Based on this, the present application proposes a vehicle fault warning method and system based on convolutional neural network, aiming to monitor the vehicle's operating status in real time, detect fault risks in advance, and improve the safety and reliability of the vehicle.
[0006] The first aspect of the present application provides a vehicle fault warning method based on a convolutional neural network, the method comprising: Obtain various sensor data during vehicle operation through on-board sensors; Preprocess the collected data from various sensors and convert the preprocessed data into image data; Analyze the preprocessed image data using a convolutional neural network to extract image features, and compare the image features with deep sample features in the fault library to determine whether the vehicle has a fault risk; When it is determined that the vehicle is at risk of failure, a warning signal will be issued in a timely manner to remind the driver to take appropriate measures.
[0007] Compared with the prior art, the present application provides a vehicle fault warning method based on a convolutional neural network. The method monitors the running status of the vehicle in real time, can comprehensively collect the performance data of the vehicle (such as speed, temperature, pressure, etc.), and provides a rich information basis for subsequent analysis, thereby improving the safety and reliability of the vehicle. Through data preprocessing, noise and outliers can be removed to improve the quality and accuracy of the data. Converting data into image form helps to use visual features for more intuitive analysis, which is convenient for subsequent deep learning processing. The convolutional neural network can automatically identify and extract important features, reduce manual intervention, and improve analysis efficiency and accuracy. Through advanced technologies such as deep learning, potential fault modes can be discovered and the intelligent level of fault detection can be enhanced. By comparing with the fault library, the potential fault risks of the vehicle can be quickly and accurately identified, the false alarm rate can be reduced, the reliability of fault diagnosis can be improved, and the safety of the vehicle during operation can be ensured. Timely warning signals can enable the driver to take necessary measures before the fault occurs, avoid accidents, and ensure driving safety. At the same time, the driver's cognition of the vehicle status is enhanced, and the driving experience and confidence are improved. Therefore, the method can monitor the running status of the vehicle in real time, discover fault risks in advance, and improve the safety and reliability of the vehicle.
[0008] As an optional implementation of the first aspect of the present application, the multiple sensor data include vehicle speed, engine speed, oil temperature, and water temperature.
[0009] As an optional implementation manner of the first aspect, the step of preprocessing the collected multiple sensor data and converting the preprocessed data into image data includes: The collected sensor data is preprocessed including cleaning, denoising and normalization, and the preprocessed data is converted into image data, which includes a time-frequency spectrum diagram and a vibration signal diagram.
[0010] As an optional implementation manner of the first aspect, in the step of preprocessing the collected multiple sensor data and converting the preprocessed data into image data, the conversion process includes: Segmenting the one-dimensional sensor data according to preset time intervals to generate a plurality of segmented sensor data; Perform Fourier transform on each segmented sensor data to obtain frequency domain data image data; The frequency domain data is plotted into an image to form image data including a time-frequency spectrum diagram and a vibration signal diagram.
[0011] As an optional implementation manner of the first aspect, the step of analyzing the preprocessed image data using a convolutional neural network to extract image features, and comparing the image features with deep sample features in a fault library to determine whether the vehicle has a fault risk includes: Collecting a large number of vehicle fault data samples from the fault library, the vehicle fault data samples including vehicle fault cases, laboratory test data, and simulation data, and marking all vehicle data in the vehicle fault data samples; Use convolutional neural networks to perform regional segmentation and category analysis training on labeled fault data samples to extract deep sample features from fault data samples; Extract preprocessed image data; It is determined whether the deep sample feature matches the preprocessed image data. If the deep sample feature does not match the preprocessed image data, it is determined that there is a risk of vehicle failure.
[0012] As an optional implementation manner of the first aspect, when it is determined that the vehicle has a risk of failure, the step of timely issuing a warning signal to remind the driver to take corresponding measures includes: The warning signal is divided into three levels of warning standards, and different colors, sounds, frequencies and prompt contents are used to display them in the vehicle instrument panel respectively; The first level of warning is set to be displayed in the vehicle instrument panel using red, above 100 decibels, 2kHz and prompting immediate action; The second level warning is set to be displayed in the vehicle instrument panel using orange, 80-100 decibel range, 20Hz to 2kHz and prompting attention; The third level warning is set to be displayed on the vehicle instrument panel using yellow, 60-80 decibel range, below 20 Hz and general situation prompts.
[0013] A second aspect of the present application provides a vehicle fault warning system based on a convolutional neural network, the system comprising: A data acquisition module is used to obtain various sensor data during vehicle operation through on-board sensors; A data preprocessing module is used to preprocess the collected data of various sensors and convert the preprocessed data into image data; A fault analysis module is used to analyze the preprocessed image data using a convolutional neural network to extract image features, and compare the image features with deep sample features in the fault library to determine whether the vehicle has a fault risk; The early warning module is used to send out early warning signals in time when it is judged that the vehicle has a risk of failure, reminding the driver to take corresponding measures.
[0014] As an optional implementation manner of the second aspect, the fault analysis module includes a marking unit, a training unit, an extraction unit and a vehicle fault risk identification unit, wherein: A marking unit, used to collect a large number of vehicle fault data samples from a fault library, wherein the vehicle fault data samples include vehicle fault cases, laboratory test data, and simulation data, and mark all vehicle data in the vehicle fault data samples; A training unit, used for performing regional segmentation and category analysis training on the marked fault data samples using a convolutional neural network, so as to extract deep sample features in the fault data samples; An extraction unit, used for extracting pre-processed image data; The vehicle failure risk identification unit is used to determine whether the deep sample features match the preprocessed image data. If the deep sample features do not match the preprocessed image data, it is determined that the vehicle has a failure risk.
[0015] The third aspect of the present application provides a computer device, which includes a memory, a processor, and a processing program stored in the memory and executable on the processor, and when the processing program is executed by the processor, the above-mentioned vehicle fault warning method based on convolutional neural network is implemented.
[0016] A fourth aspect of the present application provides a storage medium, on which a processing program is stored. When the processing program is run by a processor, the above-mentioned vehicle fault warning method based on convolutional neural network is executed.
[0017] Additional aspects and advantages of the present application will be given in part in the following description, and in part will become apparent from the following description, or will be understood through the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 A flow chart of a vehicle fault warning method based on a convolutional neural network proposed in the first embodiment of the present application; Figure 2 A schematic diagram of the structure of a vehicle fault warning system based on a convolutional neural network proposed in the second embodiment of the present application.
[0019] The following specific implementation methods will further illustrate the present application in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION
[0020] 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 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.
[0021] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here. In addition, the "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated with each other are in an "or" relationship.
[0022] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.
[0023] See also Figure 1 , is a flow chart of a vehicle fault warning method based on a convolutional neural network proposed in the first embodiment of the present application, and the proposed method includes: S1: Obtain various sensor data during vehicle operation through on-board sensors.
[0024] Specifically, a variety of vehicle-mounted sensors may be used to acquire a variety of sensor data from the vehicle during the operation of the vehicle. Exemplarily, the acquired various sensor data include vehicle speed, engine speed, oil temperature, and water temperature.
[0025] S2: Preprocess the collected data from various sensors and convert the preprocessed data into image data.
[0026] Specifically, the collected sensor data is preprocessed including cleaning, denoising and normalization, and the preprocessed data is converted into image data, wherein the image data includes a time-frequency spectrum diagram and a vibration signal diagram.
[0027] Further, the one-dimensional sensor data is segmented according to preset time intervals to generate a plurality of segmented sensor data; Perform Fourier transform on each segmented sensor data to obtain frequency domain data image data; The frequency domain data is plotted into an image to form image data including a time-frequency spectrum diagram and a vibration signal diagram.
[0028] S3: Analyze the preprocessed image data using a convolutional neural network to extract image features, and compare the image features with deep sample features in the fault library to determine whether the vehicle has a fault risk.
[0029] Specifically, a large number of vehicle fault data samples are collected from the fault library, wherein the vehicle fault data samples include vehicle fault cases, laboratory test data, and simulation data, and all vehicle data in the vehicle fault data samples are marked; Use convolutional neural networks to perform regional segmentation and category analysis training on labeled fault data samples to extract deep sample features from fault data samples; Extract preprocessed image data; It is determined whether the deep sample feature matches the preprocessed image data. If the deep sample feature does not match the preprocessed image data, it is determined that there is a risk of vehicle failure.
[0030] S4: When it is determined that the vehicle is at risk of failure, a warning signal is issued in a timely manner to remind the driver to take appropriate measures.
[0031] Exemplarily, the warning signal is divided into three levels of warning standards, which are displayed in the vehicle instrument panel using different colors, sounds, frequencies and prompt contents respectively; The first level of warning is set to be displayed in the vehicle instrument panel using red, above 100 decibels, 2kHz and prompting immediate action; The second level warning is set to be displayed in the vehicle instrument panel using orange, 80-100 decibel range, 20Hz to 2kHz and prompting attention; The third level warning is set to be displayed on the vehicle instrument panel using yellow, 60-80 decibel range, below 20 Hz and general situation prompts.
[0032] In summary, the implementation method can comprehensively collect vehicle performance data (such as speed, temperature, pressure, etc.) by real-time monitoring of the vehicle's operating status, provide a rich information basis for subsequent analysis, and improve the safety and reliability of the vehicle; through data preprocessing, noise and outliers can be removed to improve the quality and accuracy of the data. Converting data into image form helps to use visual features for more intuitive analysis, which is convenient for subsequent deep learning processing; convolutional neural networks can automatically identify and extract important features, reduce manual intervention, and improve analysis efficiency and accuracy. Through advanced technologies such as deep learning, potential fault modes can be discovered and the intelligent level of fault detection can be enhanced; by comparing with the fault library, the potential fault risks of the vehicle can be quickly and accurately identified, the false alarm rate can be reduced, the reliability of fault diagnosis can be improved, and the safety of the vehicle during operation can be ensured; timely warning signals can enable the driver to take necessary measures before the fault occurs, avoid accidents, and ensure driving safety. At the same time, the driver's cognition of the vehicle status is enhanced, and the driving experience and confidence are improved. Therefore, the method can monitor the vehicle's operating status in real time, discover fault risks in advance, and improve the safety and reliability of the vehicle.
[0033] See also Figure 2 , which is a schematic diagram of the structure of a vehicle fault warning system based on a convolutional neural network proposed in the second embodiment of the present application, and the system includes: The data acquisition module 10 is used to obtain various sensor data during the operation of the vehicle through the vehicle-mounted sensors; The data preprocessing module 20 is used to preprocess the collected data of various sensors and convert the preprocessed data into image data; A fault analysis module 30 is used to analyze the preprocessed image data using a convolutional neural network to extract image features, and compare the image features with deep sample features in the fault library to determine whether the vehicle has a fault risk; The fault analysis module 30 includes a marking unit 301, a training unit 302, an extraction unit 303 and a vehicle fault risk identification unit 304, wherein: The marking unit 301 is used to collect a large number of vehicle fault data samples from the fault library, wherein the vehicle fault data samples include vehicle fault cases, laboratory test data, and simulation data, and mark all vehicle data in the vehicle fault data samples; The training unit 302 is used to perform regional segmentation and category analysis training on the marked fault data samples using a convolutional neural network to extract deep sample features in the fault data samples; An extraction unit 303, used to extract pre-processed image data; The vehicle failure risk identification unit 304 is used to determine whether the deep sample feature matches the pre-processed image data. If the deep sample feature does not match the pre-processed image data, it is determined that the vehicle has a failure risk.
[0034] The warning module 40 is used to send out a warning signal in time when it is determined that there is a risk of failure of the vehicle, so as to remind the driver to take corresponding measures.
[0035] On the other hand, the present application also proposes a computer device, which includes a memory, a processor, and a processing program stored in the memory and executable on the processor. When the processing program is executed by the processor, the above-mentioned vehicle fault warning method based on convolutional neural network is implemented.
[0036] On the other hand, the present application further proposes a storage medium on which a processing program is stored. When the processing program is run by a processor, the above-mentioned vehicle fault warning method based on convolutional neural network is executed.
[0037] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.
[0038] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.
[0039] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.
Claims
1. A vehicle fault warning method based on convolutional neural network, characterized in that: The method comprises: Obtain various sensor data during vehicle operation through on-board sensors; Preprocess the collected data from various sensors and convert the preprocessed data into image data; Analyze the preprocessed image data using a convolutional neural network to extract image features, and compare the image features with deep sample features in the fault library to determine whether the vehicle has a fault risk; When it is determined that the vehicle is at risk of failure, a warning signal will be issued in a timely manner to remind the driver to take appropriate measures.
2. A vehicle fault warning method based on convolutional neural network according to claim 1, characterized in that: The various sensor data include vehicle speed, engine speed, oil temperature, and water temperature.
3. The vehicle fault warning method based on convolutional neural network according to claim 1 is characterized in that: The step of preprocessing the collected multiple sensor data and converting the preprocessed data into image data includes: The collected sensor data is preprocessed including cleaning, denoising and normalization, and the preprocessed data is converted into image data, which includes a time-frequency spectrum diagram and a vibration signal diagram.
4. The vehicle fault warning method based on convolutional neural network according to claim 1 is characterized in that: In the step of preprocessing the collected multiple sensor data and converting the preprocessed data into image data, the conversion process includes: Segmenting the one-dimensional sensor data according to preset time intervals to generate a plurality of segmented sensor data; Perform Fourier transform on each segmented sensor data to obtain frequency domain data image data; The frequency domain data is plotted into an image to form image data including a time-frequency spectrum diagram and a vibration signal diagram.
5. The vehicle fault warning method based on convolutional neural network according to claim 1 is characterized in that: The steps of analyzing the preprocessed image data using a convolutional neural network to extract image features, and comparing the image features with deep sample features in a fault library to determine whether the vehicle has a fault risk include: Collecting a large number of vehicle fault data samples from the fault library, the vehicle fault data samples including vehicle fault cases, laboratory test data, and simulation data, and marking all vehicle data in the vehicle fault data samples; Use convolutional neural networks to perform regional segmentation and category analysis training on labeled fault data samples to extract deep sample features from fault data samples; Extract preprocessed image data; It is determined whether the deep sample feature matches the preprocessed image data. If the deep sample feature does not match the preprocessed image data, it is determined that there is a risk of vehicle failure.
6. The vehicle fault warning method based on convolutional neural network according to claim 1 is characterized in that: When it is determined that the vehicle has a risk of failure, the steps of timely issuing a warning signal to remind the driver to take corresponding measures include: The warning signal is divided into three levels of warning standards, and different colors, sounds, frequencies and prompt contents are used to display them in the vehicle instrument panel respectively; The first level of warning is set to be displayed in the vehicle instrument panel using red, above 100 decibels, 2kHz and prompting immediate action; The second level warning is set to be displayed in the vehicle instrument panel using orange, 80-100 decibel range, 20Hz to 2kHz and prompting attention; The third level warning is set to be displayed on the vehicle instrument panel using yellow, 60-80 decibel range, below 20 Hz and general situation prompts.
7. A vehicle fault warning system based on convolutional neural network, characterized in that: The system comprises: A data acquisition module is used to obtain various sensor data during vehicle operation through on-board sensors; A data preprocessing module is used to preprocess the collected data of various sensors and convert the preprocessed data into image data; A fault analysis module is used to analyze the preprocessed image data using a convolutional neural network to extract image features, and compare the image features with deep sample features in the fault library to determine whether the vehicle has a fault risk; The early warning module is used to send out early warning signals in time when it is judged that the vehicle has a risk of failure, reminding the driver to take corresponding measures.
8. A vehicle fault warning system based on convolutional neural network according to claim 7, characterized in that: The fault analysis module includes a marking unit, a training unit, an extraction unit and a vehicle fault risk identification unit, wherein: A marking unit, used to collect a large number of vehicle fault data samples from a fault library, wherein the vehicle fault data samples include vehicle fault cases, laboratory test data, and simulation data, and mark all vehicle data in the vehicle fault data samples; A training unit, used for performing regional segmentation and category analysis training on the marked fault data samples using a convolutional neural network, so as to extract deep sample features in the fault data samples; An extraction unit, used for extracting pre-processed image data; The vehicle failure risk identification unit is used to determine whether the deep sample feature matches the preprocessed image data. If the deep sample feature does not match the preprocessed image data, it is determined that the vehicle has a failure risk.
9. A computer device, characterized in that: The computer device includes a memory, a processor, and a processing program stored in the memory and executable on the processor. When the processing program is executed by the processor, a vehicle fault warning method based on a convolutional neural network as described in any one of claims 1 to 6 is implemented.
10. A storage medium, characterized in that: The storage medium stores a processing program, and when the processing program is executed by the processor, a vehicle fault warning method based on a convolutional neural network as described in any one of claims 1 to 6 is implemented.
Citation Information
Patent Citations
Fault prediction and health management device for vehicle transmission control system and fault prediction method
CN113391612A
Demagnetization fault diagnosis method, system and device for permanent magnet synchronous motor
CN114264953A
Mechanical fault diagnosis method based on combination of time-frequency analysis and convolutional neural network
CN116735203A
Amphibious vehicle water jet propulsion system state monitoring device and fault diagnosis method
CN117451333A
Power machinery fault diagnosis system
CN118032364A
