Vehicle and fault prediction method, device and storage medium thereof, computer program product
By acquiring vehicle inspection data, analyzing driving conditions and habits, establishing inspection models, and updating and predicting the degree of degradation of vehicle components in real time, the problem of failure to consider the influence of operating conditions and habits in existing technologies is solved, and accurate prediction of fault levels and targeted maintenance are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BYD CO LTD
- Filing Date
- 2024-12-16
- Publication Date
- 2026-06-16
AI Technical Summary
Existing technologies fail to effectively consider the impact of vehicle operating conditions and driver habits on component lifespan, and fail to update prediction models in a timely manner, resulting in poor generalization ability of prediction models and an inability to accurately classify and grade faults.
By acquiring vehicle inspection data, analyzing driving conditions and habits, establishing targeted inspection models, updating and iterating prediction models in real time, diagnosing the degree of degradation of vehicle components, predicting their future health status, and generating fault levels.
It improves the generalization ability and fault level accuracy of the predictive model, provides targeted maintenance suggestions, reduces maintenance costs, and improves vehicle reliability and safety.
Smart Images
Figure CN122221075A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of vehicle technology, and in particular to a vehicle and its fault prediction method, device, storage medium, and computer program product. Background Technology
[0002] Currently, sensors collect relevant data about vehicle components and transmit the data to the cloud. The cloud then calculates the health status and remaining lifespan of all components. When a component's health status is abnormal, the system searches for nearby repair shops and transmits warning messages and the repair shop's location to the vehicle.
[0003] However, the above method does not take into account the impact of other factors that may change in real time, such as vehicle operating conditions and driver habits, on the service life of vehicle components; nor does it take into account the update and iteration of the prediction model system. Summary of the Invention
[0004] The present invention aims to solve at least one of the technical problems existing in the prior art.
[0005] Therefore, one objective of this invention is to propose a vehicle fault prediction method that improves the generalization ability of the detection model and the accuracy of predicting the fault level of vehicle components by generating a targeted prediction model based on continuously updated detection data.
[0006] Therefore, a second objective of the present invention is to provide a vehicle fault prediction device.
[0007] Therefore, a third objective of the present invention is to provide a vehicle.
[0008] Therefore, a fourth object of the present invention is to provide a computer-readable storage medium.
[0009] Therefore, the fifth objective of this invention is to provide a computer program product.
[0010] To achieve the above objectives, an embodiment of the first aspect of the present invention provides a vehicle fault prediction method, the vehicle fault prediction method comprising: acquiring detection data of the vehicle; determining a detection model based on the detection data; and predicting the fault level of the vehicle based on the detection model.
[0011] According to the vehicle fault prediction method of the present invention, by acquiring vehicle detection data, it is possible to determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0012] In some embodiments, determining a detection model based on the detection data includes: acquiring target detection data from the detection data; determining a target dataset based on the target detection data; and performing algorithmic processing on the target dataset to obtain the detection model.
[0013] In some embodiments, acquiring target detection data from the detection data includes: removing first abnormal detection data and normal detection data from the detection data, wherein the first abnormal data is used to indicate the absence of vehicle components; The second abnormal detection data in the detection data is determined as the target detection data, and the second abnormal data is used to indicate abnormal vehicle device operating data.
[0014] In some embodiments, determining a target dataset based on the target detection data includes: extracting object features and / or fault features from the target detection data; classifying the target detection data based on the object features and / or fault features to generate the target dataset.
[0015] In some embodiments, predicting the fault level of the vehicle based on the detection model includes: predicting the failure rate and remaining life of the vehicle based on the detection model; and determining the fault level based on the failure rate and the remaining life.
[0016] In some embodiments, obtaining the vehicle's detection data includes: obtaining historical detection data from the vehicle's historical detection database, and / or obtaining the vehicle's real-time detection data.
[0017] In some embodiments, after acquiring the detection data of the vehicle, the method further includes: saving the detection data to the cloud.
[0018] In some embodiments, after predicting the fault level of the vehicle based on the detection model, the method further includes: controlling the vehicle's in-vehicle infotainment system to issue a preset prompt message based on the fault level.
[0019] To achieve the above objectives, a second aspect of the present invention provides a vehicle fault prediction device, the vehicle fault prediction device comprising: an acquisition module for acquiring detection data of the vehicle; a determination module for determining a detection model based on the detection data; and a prediction module for predicting the fault level of the vehicle based on the detection model.
[0020] According to the vehicle fault prediction device of the present invention, by acquiring vehicle detection data, it can determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0021] To achieve the above objectives, a third aspect of the present invention provides a vehicle that includes the vehicle fault prediction device described in the above embodiments.
[0022] According to embodiments of the present invention, by acquiring vehicle detection data, it is possible to determine the real-time changing driving conditions of the vehicle and the driver's driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0023] To achieve the above objectives, a fourth aspect of the present invention provides a computer-readable storage medium storing a vehicle fault prediction program, which, when executed by a processor, implements the vehicle fault prediction method described in the above embodiments.
[0024] According to the computer-readable storage medium of the present invention, by acquiring vehicle detection data, it is possible to determine the real-time changing driving conditions of the vehicle and the driver's driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0025] To achieve the above objectives, a fifth aspect of the present invention provides a computer program product comprising a computer program that, when executed by a processor, implements the vehicle fault prediction method described in the above embodiments.
[0026] According to the computer program product of the present invention, by acquiring vehicle detection data, it is able to determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0027] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description
[0028] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of a vehicle fault prediction method according to an embodiment of the present invention; Figure 2 This is a network structure diagram of a vehicle fault prediction method according to an embodiment of the present invention; Figure 3 This is a flowchart of a vehicle fault prediction method according to a specific embodiment of the present invention; Figure 4 This is a structural block diagram of a vehicle fault prediction device according to an embodiment of the present invention; Figure 5 This is a structural block diagram of a vehicle according to an embodiment of the present invention.
[0029] Attached image caption: Vehicle fault prediction device 2; Acquisition module 21; Determination module 22; Prediction module 23; Vehicle 3. Detailed Implementation
[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0031] In the description of this invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.
[0032] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.
[0033] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0034] In related technologies, devices for cloud-based real-time predictive maintenance of vehicle components include: sensors coupled to components in the vehicle, the sensors including direct sensors and / or surrogate sensors, the direct sensors being used to directly measure the life determinants of the corresponding components, and the surrogate sensors being used to measure the surrogate life determinants of the corresponding components; a controller communicatively coupled to the direct sensors and / or surrogate sensors, a user interface and a transceiver communicatively coupled to the controller, the controller including a microprocessor and a communication interface, the microprocessor being used to receive measurement results of life determinants from the direct sensors and / or measurement results of surrogate life determinants from the surrogate sensors to obtain a dataset, and the communication interface being used to transmit the dataset to a cloud computing center via the transceiver and an antenna.
[0035] The above methods only predict battery failures and do not consider the failures of other vehicle components. They also do not consider the impact of driving conditions and driver habits on device lifespan, resulting in poor generalization ability. Furthermore, they do not consider the updating and iteration of the prediction model system, do not classify failures, making it impossible to grade and rank the damage caused by failures, and do not consider the causes of vehicle component failures or solutions.
[0036] The following is combined Figures 1-3This invention describes a vehicle fault prediction method according to an embodiment of the present invention.
[0037] like Figure 1 As shown, the vehicle fault prediction method of this embodiment of the invention includes at least steps S1 and S2.
[0038] Step S1: Obtain vehicle detection data.
[0039] In this embodiment, the detection data is data used to check the technical condition or working capacity of the vehicle. For example, the detection data can be collected in real time by the internal detection equipment and / or the detection data pre-recorded by the vehicle repair service company can be directly obtained. The vehicle's detection data is continuously collected through the cloud server to obtain the vehicle's driving condition and the driver's driving habits.
[0040] Step S2: Determine the detection model based on the detection data.
[0041] In this embodiment, after acquiring vehicle detection data, the vehicle's driving conditions and the driver's driving habits are analyzed. Targeted detection models are built for drivers with different driving habits, including but not limited to algorithms such as neural networks, support vector machines, clustering algorithms, and random forests. The prediction model is updated and iterated based on continuously updated detection data to improve the accuracy of the detection model's predictions.
[0042] Step S3: Predict the vehicle's fault level based on the detection model.
[0043] In this embodiment, after the detection model is determined, the detection model can use algorithms to diagnose the degree of degradation of vehicle components in real time, predict the future health status of vehicle components, and generate quantitative indicators, i.e., fault levels, so as to provide different solutions according to the harm caused by the fault, improve the reliability and safety of the vehicle, reduce maintenance costs, and improve the maintenance quality and efficiency of vehicle repair service companies.
[0044] According to the vehicle fault prediction method of the present invention, by acquiring vehicle detection data, it is possible to determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0045] In some embodiments, determining a detection model based on detection data includes: acquiring target detection data from the detection data; determining a target dataset based on the target detection data; and performing algorithmic processing on the target dataset to obtain a detection model.
[0046] In this embodiment, after acquiring the vehicle detection data, since the detection data is raw data, it is first preprocessed to remove detection data that is irrelevant to the prediction results, and the remaining usable data is determined as the target detection data. After determining the target detection data, effective features are extracted from the target detection data from the perspectives of time domain, frequency domain, etc., and the better features are selected. Useful features that are more consistent with the degradation state of the monitored vehicle are screened out. The target dataset is classified according to the features and used for the next step of model training to obtain the detection model, which provides a guarantee for obtaining higher prediction accuracy in the future.
[0047] For example, based on the usage time and operating conditions of vehicle components, vehicle failure modes, vehicle driving conditions, and signals such as temperature, vibration, and pressure collected by sensors, a suitable machine learning or deep learning detection model can be established.
[0048] In some embodiments, acquiring target detection data from the detection data includes: removing first abnormal detection data and normal detection data from the detection data, wherein the first abnormal data is used to indicate the absence of vehicle components; and determining second abnormal detection data from the detection data as target detection data, wherein the second abnormal data is used to indicate abnormal vehicle component operating data.
[0049] In this embodiment, after acquiring the vehicle's detection data, the first abnormal detection data and normal detection data are removed from the detection data. The first abnormal data is used to indicate that the vehicle's components are missing, and the normal detection data is used to indicate that the vehicle is not faulty, so as to remove detection data that is irrelevant to the prediction result and prevent adverse effects on the prediction result. The second abnormal detection data in the detection data is determined as the target detection data. The second abnormal data is used to indicate that the vehicle's component operating data is abnormal. That is, only the data in the detection data that is relevant to the prediction result is retained, thereby improving the accuracy of subsequent predictions.
[0050] In addition, a fixed time window is set for the detection data to facilitate subsequent analysis, and data normalization is used to eliminate differences between different units.
[0051] In some embodiments, determining a target dataset based on target detection data includes: extracting object features and / or fault features from the target detection data; classifying the target detection data based on the object features and / or fault features to generate a target dataset.
[0052] In this embodiment, after determining the target detection data, the object features of the target detection data are extracted to classify the detection objects corresponding to the target detection data, such as electrical systems, braking systems, transmission systems, steering systems, power batteries, drive motors, air conditioning systems, driver assistance systems, and the driver's driving habits, and / or fault features, to classify different fault types of the same detection object, such as current faults; the target detection data is classified according to the object features and / or fault features to generate a target dataset, which facilitates users in making decisions about impending faults and improves maintenance efficiency.
[0053] In some embodiments, predicting the fault level of a vehicle based on a detection model includes: predicting the vehicle's failure rate and remaining lifespan based on the detection model; and determining the fault level based on the failure rate and remaining lifespan.
[0054] In this embodiment, after determining the detection model, the detection model can use algorithms to diagnose the usage status and lifespan of vehicle components in real time, predict the future health status of vehicle components, determine the vehicle's failure rate and remaining lifespan, and comprehensively determine the vehicle's failure level based on the failure rate and remaining lifespan. The severity level of the failure can be divided into three levels: high, medium, and low, which facilitates providing users with targeted maintenance and repair suggestions, such as deciding whether the vehicle needs maintenance / minor repair / major repair / replacement, and recommending maintenance and repair institutions based on the current location of the vehicle. At the same time, maintenance and repair institutions can also anticipate the vehicle information that needs maintenance and repair, improving maintenance efficiency while reducing maintenance costs. This application, by classifying failures and grading and ranking the severity of failures, can not only guide maintenance and repair institutions to adjust their inventory of commonly damaged vehicle parts, reduce maintenance costs, and improve the repair quality and efficiency of vehicle repair service companies, but also guide vehicle R&D companies to improve the problems existing in the components or systems.
[0055] In some embodiments, obtaining vehicle detection data includes: obtaining historical detection data from a historical vehicle detection database, and / or obtaining real-time vehicle detection data.
[0056] In an embodiment, such as Figure 2 As shown, historical inspection data is obtained from the vehicle repair and maintenance historical inspection database established by vehicle repair service companies to obtain relatively comprehensive inspection data covering the entire life cycle of the vehicle, and / or real-time inspection data such as driver speed, accelerator, and brake are obtained through in-vehicle inspection equipment such as OBD and sensors to obtain inspection data that conforms to reality. Figure 2 As shown, the above detection data is sent to the cloud via communication methods such as wireless network and 5G network to achieve data sharing.
[0057] In some embodiments, after predicting the fault level of the vehicle based on the detection model, the method further includes: controlling the vehicle's in-vehicle terminal to issue a preset prompt message based on the fault level.
[0058] In this embodiment, after predicting the vehicle's fault level, a preset prompt message is issued through the vehicle's in-vehicle infotainment system, such as... Figure 2 As shown, the system visualizes the health status of components on the web or mobile platform, issuing predictive warnings or maintenance reminders for severely degraded parts. This allows users to take appropriate preventative maintenance measures, reminds drivers of improper driving habits, reduces sudden vehicle malfunctions, and promptly reminds drivers to purchase relevant parts or contact maintenance personnel. This forms a complete equipment health management system with predictive capabilities.
[0059] The following is for reference. Figure 3 The vehicle fault prediction method of this invention will be illustrated by example.
[0060] like Figure 3 As shown, the vehicle fault prediction method of this embodiment of the invention includes at least steps S11-S16.
[0061] Step S11: Obtain historical inspection data from the vehicle's historical inspection database, and / or obtain real-time inspection data of the vehicle.
[0062] Step S12: Save the detection data to the cloud.
[0063] Step S13: Remove the first abnormal detection data and normal detection data from the detection data. The first abnormal data is used to indicate that the vehicle component is missing. The second abnormal detection data in the detection data is determined as the target detection data. The second abnormal data is used to indicate that the vehicle component's operating data is abnormal.
[0064] Step S14: Extract object features and / or fault features from the target detection data; classify the target detection data according to the object features and / or fault features to generate a target dataset.
[0065] Step S15: Process the target dataset using an algorithm to obtain the detection model.
[0066] Step S16: Predict the vehicle's failure rate and remaining lifespan based on the detection model; determine the failure level based on the failure rate and remaining lifespan.
[0067] According to the vehicle fault prediction method of the present invention, by acquiring vehicle detection data, it is possible to determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0068] The following is combined Figure 4 The present invention describes a vehicle fault prediction device 2 according to an embodiment of the present invention.
[0069] like Figure 4 As shown, the vehicle fault prediction device 2 of this embodiment includes: an acquisition module 21, a determination module 22, and a prediction module, wherein, The acquisition module 21 is used to acquire vehicle detection data; the determination module 22 is used to determine the detection model based on the detection data; and the prediction module 23 is used to predict the vehicle's fault level based on the detection model.
[0070] In this embodiment, the detection data is data for checking the technical condition or working capacity of the vehicle. For example, the acquisition module 21 can collect detection data in real time according to the internally set detection equipment, and / or directly acquire the detection data pre-recorded by the vehicle repair service enterprise. The vehicle's detection data is continuously collected through the cloud server to obtain the vehicle's driving conditions and the driver's driving habits.
[0071] After obtaining the vehicle's detection data, module 22 analyzes the vehicle's driving conditions and the driver's driving habits, and builds targeted detection models for drivers with different driving habits, including but not limited to neural networks, support vector machines, clustering algorithms, and random forests. The prediction model is updated and iterated based on the continuously updated detection data to improve the accuracy of the detection model's predictions.
[0072] After the prediction module 23 determines the detection model, the detection model can diagnose the degree of degradation of vehicle components in real time through algorithms, predict the future health status of vehicle components, and generate quantitative indicators, i.e., fault levels, so as to provide different solutions according to the harm caused by the fault, improve the reliability and safety of the vehicle, reduce maintenance costs, and improve the maintenance quality and efficiency of vehicle repair service companies.
[0073] According to the vehicle fault prediction device 2 of the present invention, by acquiring vehicle detection data, it can determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0074] In some embodiments, when determining the detection model based on the detection data, the determining module 22 is specifically used to: acquire target detection data in the detection data; determine the target dataset based on the target detection data; and perform algorithmic processing on the target dataset to obtain the detection model.
[0075] In this embodiment, after the determination module 22 acquires the vehicle detection data, since the detection data is raw data, it first preprocesses the detection data, removes the detection data that is irrelevant to the prediction result, and determines the remaining usable data as the target detection data. After determining the target detection data, it extracts effective features from the time domain, frequency domain and other perspectives, selects the better features, and filters out the useful features that are more consistent with the deterioration state of the monitored vehicle. Based on the feature classification, the target dataset is used for the next step of model training to obtain the detection model, which provides a guarantee for obtaining higher prediction accuracy in the future.
[0076] For example, based on the usage time and operating conditions of vehicle components, vehicle failure modes, vehicle driving conditions, and signals such as temperature, vibration, and pressure collected by sensors, a suitable machine learning or deep learning detection model can be established.
[0077] In some embodiments, when the determining module 22 acquires the target detection data in the detection data, it is specifically used to: remove the first abnormal detection data and normal detection data in the detection data, the first abnormal data being used to indicate the absence of vehicle components; and determine the second abnormal detection data in the detection data as the target detection data, the second abnormal data being used to indicate abnormal vehicle component operation data.
[0078] In this embodiment, after the determining module 22 acquires the vehicle's detection data, it removes the first abnormal detection data and the normal detection data from the detection data. The first abnormal data is used to indicate that the vehicle's components are missing, and the normal detection data is used to indicate that the vehicle is not faulty, so as to remove detection data that is irrelevant to the prediction result and prevent adverse effects on the prediction result. The second abnormal detection data in the detection data is determined as the target detection data. The second abnormal data is used to indicate that the vehicle's component operating data is abnormal. That is, only the data in the detection data that is relevant to the prediction result is retained, thereby improving the accuracy of subsequent predictions.
[0079] In addition, a fixed time window is set for the detection data to facilitate subsequent analysis, and data normalization is used to eliminate differences between different units.
[0080] In some embodiments, when determining the target dataset based on the target detection data, the determining module 22 is specifically used to: extract object features and / or fault features from the target detection data; classify the target detection data based on the object features and / or fault features to generate the target dataset.
[0081] In this embodiment, after determining the target detection data, the determining module 22 extracts the object features of the target detection data to classify the detection objects corresponding to the target detection data, such as electrical systems, braking systems, transmission systems, steering systems, power batteries, drive motors, air conditioning systems, driver assistance systems, and the driver's driving habits, and / or fault features to classify different fault types of the same detection object, such as current faults; the target detection data is classified according to the object features and / or fault features to generate a target dataset, which facilitates users in making decisions about impending faults and improves maintenance efficiency.
[0082] In some embodiments, when the prediction module 23 predicts the fault level of a vehicle based on the detection model, it is specifically used to: predict the failure rate and remaining life of the vehicle based on the detection model; and determine the fault level based on the failure rate and remaining life.
[0083] In this embodiment, after the prediction module 23 determines the detection model, the detection model can use algorithms to diagnose the usage status and lifespan of vehicle components in real time, predict the future health status of vehicle components, determine the vehicle's failure rate and remaining lifespan, and comprehensively determine the vehicle's failure level based on the failure rate and remaining lifespan. The severity level caused by the failure can be divided into three levels: high, medium, and low, which facilitates providing users with targeted maintenance suggestions, such as deciding whether the vehicle needs maintenance / minor repair / major repair / replacement, and recommending maintenance and repair institutions based on the current location of the vehicle. At the same time, maintenance and repair institutions can also know the information of vehicles that need maintenance in advance, improving maintenance efficiency while reducing maintenance costs. This application, by classifying failures and classifying and ranking the severity caused by failures, can not only guide maintenance and repair institutions to adjust the inventory of commonly damaged vehicle parts, reduce maintenance costs, and improve the repair quality and efficiency of vehicle repair service companies, but also guide vehicle R&D companies to improve the problems existing in the device or system.
[0084] In some embodiments, when the acquisition module 21 acquires vehicle detection data, it is specifically used to: acquire historical detection data from the vehicle's historical detection database, and / or acquire real-time vehicle detection data.
[0085] In this embodiment, the acquisition module 21 acquires historical test data from the vehicle repair and maintenance historical test database established by the vehicle repair service enterprise to obtain relatively comprehensive test data covering the entire life cycle of the vehicle, and / or acquires real-time test data such as driver speed, accelerator, and brake through the test equipment installed inside the vehicle, such as OBD and sensors, to obtain test data that conforms to reality. The above test data is then sent to the cloud through communication methods such as wireless network and 5G network to achieve data sharing.
[0086] In some embodiments, after the prediction module 23 predicts the fault level of the vehicle according to the detection model, it further includes: controlling the vehicle's in-vehicle terminal to issue a preset prompt message according to the fault level.
[0087] In this embodiment, after predicting the vehicle's fault level, the prediction module 23 sends a preset prompt message through the vehicle's infotainment system, such as... Figure 2 As shown, the system visualizes the health status of components on the web or mobile platform, issuing predictive warnings or maintenance reminders for severely degraded parts. This allows users to take appropriate preventative maintenance measures, reminds drivers of improper driving habits, reduces sudden vehicle malfunctions, and promptly reminds drivers to purchase relevant parts or contact maintenance personnel. This forms a complete equipment health management system with predictive capabilities.
[0088] According to the vehicle fault prediction device 2 of the present invention, by acquiring vehicle detection data, it can determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0089] The following is combined Figure 5 Vehicle 3, as described in an embodiment of the present invention.
[0090] like Figure 5 As shown, the vehicle 3 in this embodiment of the invention includes the vehicle fault prediction device 2 of the above embodiment.
[0091] According to the vehicle 3 of the present invention, by acquiring vehicle detection data, the real-time changing driving conditions of the vehicle and the driving habits of the driver can be determined. Based on the detection data, an updated and iterative prediction model is determined. The detection model is used to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0092] The following describes a computer-readable storage medium according to embodiments of the present invention.
[0093] The computer-readable storage medium of this invention stores a vehicle fault prediction program, which, when executed by a processor, implements the vehicle fault prediction method of the above embodiments.
[0094] According to the computer-readable storage medium of the present invention, by acquiring vehicle detection data, it is possible to determine the real-time changing driving conditions of the vehicle and the driver's driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0095] The following describes a computer program product based on an embodiment of the present invention.
[0096] The computer program product of this invention includes a computer program that, when executed by a processor, implements the vehicle fault prediction method described in the above embodiments.
[0097] According to the computer program product of the present invention, by acquiring vehicle detection data, it is able to determine the real-time changing vehicle driving conditions and driver driving habits, determine an updated and iterative prediction model based on the detection data, and use the detection model to diagnose the degree of degradation of vehicle components and predict their future health status to generate a fault level. By generating a targeted prediction model based on continuously updated detection data, the generalization ability of the detection model is improved, and the accuracy of predicting the fault level of vehicle components is improved.
[0098] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example.
[0099] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims
1. A method for predicting vehicle faults, characterized in that, include: Obtain the detection data of the vehicle; Determine the detection model based on the detection data; The detection model is used to predict the fault level of the vehicle.
2. The vehicle fault prediction method according to claim 1, characterized in that, Determining the detection model based on the detection data includes: Obtain the target detection data from the detection data; Determine the target dataset based on the target detection data; The target dataset is processed by an algorithm to obtain the detection model.
3. The vehicle fault prediction method according to claim 2, characterized in that, Obtaining target detection data from the detection data includes: The first abnormal detection data and the normal detection data in the detection data are removed. The first abnormal data is used to indicate the absence of vehicle components. The second abnormal detection data in the detection data is determined as the target detection data, and the second abnormal data is used to indicate abnormal vehicle device operating data.
4. The vehicle fault prediction method according to claim 2, characterized in that, The target dataset is determined based on the target detection data, including: Extract object features and / or fault features from the target detection data; The target detection data is classified according to the object characteristics and / or fault characteristics to generate the target dataset.
5. The vehicle fault prediction method according to claim 1, characterized in that, Predicting the fault level of the vehicle based on the detection model includes: The failure rate and remaining lifespan of the vehicle are predicted based on the detection model. The failure level is determined based on the failure rate and the remaining lifetime.
6. The vehicle fault prediction method according to claim 1, characterized in that, Acquiring the vehicle's detection data includes: Obtain historical detection data from the vehicle's historical detection database, and / or obtain real-time detection data for the vehicle.
7. The vehicle fault prediction method according to claim 1, characterized in that, After obtaining the detection data of the vehicle, the method further includes: The detection data is saved to the cloud.
8. The vehicle fault prediction method according to claim 1, characterized in that, After predicting the fault level of the vehicle based on the detection model, the method further includes: Based on the fault level, the vehicle's infotainment system will issue a preset prompt message.
9. A vehicle fault prediction device, characterized in that, include: An acquisition module is used to acquire the detection data of the vehicle; The determination module is used to determine the detection model based on the detection data; The prediction module is used to predict the fault level of the vehicle based on the detection model.
10. A vehicle, characterized in that, include: The vehicle fault prediction device as described in claim 9.
11. A computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores a vehicle fault prediction program, which, when executed by a processor, implements the vehicle fault prediction method as described in any one of claims 1-8.
12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the vehicle fault prediction method as described in any one of claims 1-8.