A vehicle intelligent monitoring method and system based on machine learning

Abstract

The present invention relates to the technical field of drive control systems. The present invention provides a vehicle intelligent monitoring method and system based on machine learning, which aims to solve the problem that existing vehicle intelligent monitoring methods cannot efficiently and real-time integrate multi-source data and perform accurate analysis. By acquiring basic vehicle data, combining real-time location information and natural environment data, a vehicle driving data set is constructed. A neural network is used to train the electro-hydraulic brake valve operation data and the vehicle driving data set to obtain a vehicle intelligent monitoring model. The model can output the vehicle intelligent monitoring results in real time and substitute them into the vehicle operation fault prediction model for fault prediction. If there is fault information, the vehicle drive parameters corresponding to the fault information are matched, and the vehicle drive system is dynamically adjusted to improve the safety and stability of the vehicle. The present invention also reduces maintenance costs and time by monitoring the electro-hydraulic brake valve to predict its remaining service life and leakage location.

Description

Technical Field

[0001] The present invention relates to the technical field of drive control systems, and in particular to a vehicle intelligent monitoring method and system based on machine learning. Background Art

[0002] With the continuous advancement of science and technology, intelligent control technology has been widely used in the field of vehicle engineering, and has provided new possibilities for improving vehicle performance and driving experience. Among them, the application of vehicle intelligent monitoring methods based on machine learning in drive control systems is particularly important, which can improve vehicle safety and stability by real-time monitoring and analysis of vehicle status.

[0003] In the drive control system, the vehicle intelligent monitoring method based on machine learning is implemented in the following way: various sensors (such as lidar, camera, millimeter wave radar, etc.) are used to collect data about the vehicle's surrounding environment and vehicle status in real time. The collected data is processed and analyzed through advanced algorithm technology, including feature extraction, data cleaning and pattern recognition. Machine learning algorithms play a key role in this process and can automatically discover patterns and trends in the data. Based on the processed data, the intelligent monitoring system can make intelligent decisions, such as adjusting the vehicle speed, steering angle or braking force, to adapt to different driving conditions and road conditions. At the same time, it can also achieve precise control of the vehicle by controlling the vehicle's power system, steering system and braking system.

[0004] In the actual application of the vehicle intelligent monitoring method in the drive control system for autonomous road vehicles, there are the following technical pain points. Autonomous road vehicles need to accurately perceive the surrounding environment, including roads, obstacles, traffic signs, pedestrians, etc. However, the real environment is complex and changeable, such as weather changes (rain, snow, fog, etc.), lighting conditions (day, night, backlight, etc.), road conditions (construction, potholes, slippery, etc.) will affect the accuracy of the perception system. In addition, the recognition of special or rare objects is also a challenge, such as animals suddenly breaking into the road, non-standard traffic signs, etc. Autonomous road vehicles are usually equipped with a variety of sensors (such as cameras, radars, lidars, etc.) to obtain more comprehensive environmental information. However, as the vehicle speed increases, the multi-source data collected during the vehicle driving process is efficiently and real-time integrated, and accurately analyzed and processed, any delay may lead to safety problems. For this reason, the present invention provides a vehicle intelligent monitoring method and system based on machine learning. Summary of the invention

[0005] In view of the shortcomings of the existing technology, the present invention provides a vehicle intelligent monitoring method and system based on machine learning, which solves the problem that the existing vehicle intelligent monitoring method cannot efficiently and real-time integrate multi-source data collected during vehicle driving and perform accurate analysis and processing.

[0006] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0007] In a first aspect, the present invention provides a vehicle intelligent monitoring method based on machine learning, comprising:

[0008] Step S101, acquiring basic vehicle data, the basic vehicle data including environmental perception data, vehicle status data and timestamp data, the environmental perception data including road data, obstacle data, traffic data, signal data, weather data and lighting condition data, the vehicle status data including vehicle dynamics data, vehicle position data and vehicle sensor data;

[0009] Step S102, collecting the real-time position of the vehicle, determining the natural environment of the vehicle according to the real-time position of the vehicle, establishing a correspondence between the real-time position of the vehicle and the corresponding natural environment data and the corresponding vehicle driving and operation data, and obtaining a vehicle driving data set;

[0010] Step S103, matching the data in the vehicle driving data set with the preset vehicle driving parameters, and grouping the matched data to obtain vehicle driving group data, and retrieving the corresponding electro-hydraulic brake valve operation data in the database according to the time period range generated by the vehicle driving group data;

[0011] Step S104, using the electro-hydraulic brake valve operation data and the data in the vehicle driving data set to train the neural network to obtain a vehicle intelligent monitoring model, substituting the real-time vehicle basic data into the vehicle intelligent monitoring model, running the vehicle intelligent monitoring model to output the vehicle intelligent monitoring results;

[0012] Step S105, substitute the vehicle intelligent monitoring result into the vehicle operation fault prediction model to obtain the vehicle operation fault prediction result. If there is fault information in the vehicle operation fault prediction result, the fault information in the vehicle operation fault prediction result is matched in a preset vehicle drive parameter knowledge base to obtain the vehicle drive parameters corresponding to the fault information, transmit the vehicle drive parameters corresponding to the fault information to the vehicle drive system, and obtain feedback information of the vehicle drive system executing the vehicle drive parameters corresponding to the fault information.

[0013] Furthermore, the vehicle intelligent monitoring method based on machine learning described in the present invention, said S103, comprises:

[0014] Performing cluster analysis on feedback information of vehicle driving parameters to obtain cluster analysis results after the vehicle executes the driving parameters, wherein the cluster analysis results after the vehicle executes the driving parameters include parameter data at different time periods and under different natural environments;

[0015] The cluster analysis results after the vehicle executes the driving parameters are matched in a preset vehicle operation standard parameter knowledge base to determine the type of vehicle operation parameters to be compared, and the vehicle operation parameter types to be compared are matched in the preset vehicle operation standard parameter knowledge base to obtain the value of the vehicle operation standard parameter to be compared;

[0016] The feedback information of the vehicle driving parameters is compared with the standard operating parameter values ​​of the vehicle to be compared to obtain a vehicle operating parameter comparison result, which includes parameter data that meets the standard and parameter data that does not meet the standard.

[0017] Furthermore, the vehicle intelligent monitoring method based on machine learning described in the present invention, said S104, comprises:

[0018] Establishing vehicle operation time intervals, which include short-term and long-term operation time intervals. The short-term time interval ranges from 0 minutes to 30 minutes, and the long-term time interval ranges from 30 minutes to 600 minutes.

[0019] Screen the data of parameters that do not meet the standards, filter out the emergency braking information after the vehicle travels at a speed higher than 60 kilometers per hour and in a long period of time, and retrieve the response speed of the electro-hydraulic brake valve within the corresponding time range;

[0020] Retrieve the vehicle's external temperature data for natural environment information within the corresponding time range to obtain the real-time operating temperature of the electro-hydraulic brake valve;

[0021] Substitute the response speed of the electro-hydraulic brake valve within the corresponding time range, the vehicle external temperature data of the natural environment information within the corresponding time range, and the real-time working temperature of the electro-hydraulic brake valve into the vehicle operation fault prediction model to obtain a preset temperature threshold;

[0022] The real-time operating temperature of the electro-hydraulic brake valve is compared with a preset temperature threshold. If the real-time operating temperature exceeds the preset temperature threshold, it is determined that the insulation performance of the electromagnet coil has degraded, and a warning message is issued to the driver.

[0023] Furthermore, the vehicle intelligent monitoring method based on machine learning described in the present invention, said S105, comprises:

[0024] By installing a micro displacement sensor inside the electro-hydraulic brake valve, the clearance between the valve core and the valve body is monitored in real time. Based on the clearance data obtained, a correlation model between the clearance change and the valve action performance is established using a machine learning algorithm to predict the remaining service life of the valve.

[0025] A pressure sensor is installed on the brake line to monitor the line pressure in real time. If the pressure is lower than the preset threshold, it is judged that there is a leak. The pressure change pattern is analyzed through a machine learning algorithm to determine the leak location.

[0026] In a second aspect, the present invention provides a vehicle intelligent monitoring system based on machine learning, comprising:

[0027] A data acquisition unit, used to acquire basic vehicle data, the basic vehicle data including environmental perception data, vehicle status data and timestamp data, the environmental perception data including road data, obstacle data, traffic data, signal data, weather data and lighting condition data, the vehicle status data including vehicle dynamics data, vehicle position data and vehicle sensor data;

[0028] A data reorganization unit is used to collect the real-time position of the vehicle, determine the natural environment of the vehicle according to the real-time position of the vehicle, establish a corresponding relationship between the real-time position of the vehicle and the corresponding natural environment data and the corresponding vehicle driving and operation data, and obtain a vehicle driving data set;

[0029] A data retrieval unit is used to match the data types of the data in the vehicle driving data set with the preset vehicle driving parameters, and group the matched data to obtain vehicle driving group data, and retrieve the corresponding electro-hydraulic brake valve operation data in the database according to the time period range generated by the vehicle driving group data;

[0030] The vehicle operation monitoring unit is used to train the neural network using the electro-hydraulic brake valve operation data and the data in the vehicle driving data set to obtain a vehicle intelligent monitoring model, substitute the vehicle basic data obtained in real time into the vehicle intelligent monitoring model, and run the vehicle intelligent monitoring model to output the vehicle intelligent monitoring results;

[0031] The vehicle drive unit is used to substitute the vehicle intelligent monitoring results into the vehicle operation fault prediction model to obtain the vehicle operation fault prediction results. If there is fault information in the vehicle operation fault prediction results, the fault information in the vehicle operation fault prediction results is matched in a preset vehicle drive parameter knowledge base to obtain the vehicle drive parameters corresponding to the fault information, and the vehicle drive parameters corresponding to the fault information are transmitted to the vehicle drive system to obtain feedback information of the vehicle drive system executing the vehicle drive parameters corresponding to the fault information.

[0032] Furthermore, in the vehicle intelligent monitoring system based on machine learning of the present invention, the data retrieval unit is also used for:

[0033] Performing cluster analysis on feedback information of vehicle driving parameters to obtain cluster analysis results after the vehicle executes the driving parameters, wherein the cluster analysis results after the vehicle executes the driving parameters include parameter data at different time periods and under different natural environments;

[0034] The cluster analysis results after the vehicle executes the driving parameters are matched in a preset vehicle operation standard parameter knowledge base to determine the type of vehicle operation parameters to be compared, and the vehicle operation parameter types to be compared are matched in the preset vehicle operation standard parameter knowledge base to obtain the value of the vehicle operation standard parameter to be compared;

[0035] The feedback information of the vehicle driving parameters is compared with the standard operating parameter values ​​of the vehicle to be compared to obtain a vehicle operating parameter comparison result, which includes parameter data that meets the standard and parameter data that does not meet the standard.

[0036] Furthermore, in the vehicle intelligent monitoring system based on machine learning of the present invention, the vehicle operation monitoring unit is further used for:

[0037] Establishing vehicle operation time intervals, which include short-term and long-term operation time intervals. The short-term time interval ranges from 0 minutes to 30 minutes, and the long-term time interval ranges from 30 minutes to 600 minutes.

[0038] Screen the data of parameters that do not meet the standards, filter out the emergency braking information after the vehicle travels at a speed higher than 60 kilometers per hour and in a long period of time, and retrieve the response speed of the electro-hydraulic brake valve within the corresponding time range;

[0039] Retrieve the vehicle's external temperature data for natural environment information within the corresponding time range to obtain the real-time operating temperature of the electro-hydraulic brake valve;

[0040] Substitute the response speed of the electro-hydraulic brake valve within the corresponding time range, the vehicle external temperature data of the natural environment information within the corresponding time range, and the real-time working temperature of the electro-hydraulic brake valve into the vehicle operation fault prediction model to obtain a preset temperature threshold;

[0041] The real-time operating temperature of the electro-hydraulic brake valve is compared with a preset temperature threshold. If the real-time operating temperature exceeds the preset temperature threshold, it is determined that the insulation performance of the electromagnet coil has degraded, and a warning message is issued to the driver.

[0042] Furthermore, in the vehicle intelligent monitoring system based on machine learning of the present invention, the vehicle drive unit is also used for:

[0043] By installing a micro displacement sensor inside the electro-hydraulic brake valve, the clearance between the valve core and the valve body is monitored in real time. Based on the clearance data obtained, a correlation model between the clearance change and the valve action performance is established using a machine learning algorithm to predict the remaining service life of the valve.

[0044] A pressure sensor is installed on the brake line to monitor the line pressure in real time. If the pressure is lower than the preset threshold, it is judged that there is a leak. The pressure change pattern is analyzed through a machine learning algorithm to determine the leak location.

[0045] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0046] The present invention obtains basic vehicle data (including environmental perception data, vehicle status data and timestamp data) and uses advanced algorithm technology to efficiently fuse and process multi-source data. This method solves the problem that existing vehicle intelligent monitoring methods cannot fuse and process multi-source data in real time, and improves the efficiency and accuracy of data processing.

[0047] By real-time monitoring and analysis of vehicle status, the present invention can intelligently adjust parameters such as vehicle speed, steering angle or braking force to adapt to different driving conditions and road conditions. This helps to improve the safety and stability of the vehicle and reduce the risk of traffic accidents.

[0048] The present invention uses the vehicle intelligent monitoring results and the vehicle operation fault prediction model to accurately predict the potential fault information of the vehicle. By early warning and taking corresponding measures, the occurrence of faults can be effectively avoided or the impact of faults can be reduced, thereby improving the reliability and service life of the vehicle.

[0049] Dynamically adjust vehicle driving parameters:

[0050] When a vehicle operation failure is predicted, the present invention will match the vehicle drive parameter knowledge base, obtain the vehicle drive parameter corresponding to the fault information, and dynamically adjust the vehicle drive system. This dynamic adjustment capability can improve the optimal operation state of the vehicle under different conditions and enhance the driving experience.

[0051] By monitoring the working status of key components such as electro-hydraulic brake valves in real time and predicting their remaining service life and leakage locations, the present invention can arrange maintenance or replacement work in advance, which helps to reduce vehicle downtime and maintenance costs caused by failures and improve vehicle operating efficiency.

[0052] The present invention proposes a vehicle intelligent monitoring method and system based on machine learning, which brings a new solution to the field of vehicle engineering. This will help promote the further application and development of intelligent control technology in vehicle engineering and improve the performance and intelligence level of vehicles.

[0053] In summary, the present invention has demonstrated significant beneficial effects by efficiently and in real time integrating multi-source data, improving vehicle safety and stability, accurately predicting vehicle operating failures, dynamically adjusting vehicle driving parameters, reducing maintenance costs and time, and promoting the application of intelligent control technology in vehicle engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0055] Figure 1 A schematic flow chart of a vehicle intelligent monitoring method based on machine learning provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0057] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0058] In a first aspect, the present invention provides a vehicle intelligent monitoring method based on machine learning, comprising:

[0059] Step S101, acquiring basic vehicle data, the basic vehicle data including environmental perception data, vehicle status data and timestamp data, the environmental perception data including road data, obstacle data, traffic data, signal data, weather data and lighting condition data, the vehicle status data including vehicle dynamics data, vehicle position data and vehicle sensor data;

[0060] In step S101, the present invention provides comprehensive information support for subsequent intelligent monitoring and fault prediction by acquiring basic vehicle data. These data include environmental perception data, vehicle status data, and timestamp data, which are specifically described as follows:

[0061] Road data: including road type (such as highways, urban roads, rural roads, etc.), road conditions (such as construction, potholes, slippery, etc.) and road geometry parameters (such as curvature, slope, etc.).

[0062] Obstacle data: involves information about various obstacles around the vehicle, such as the position, speed, direction, etc. of other vehicles, pedestrians, stationary objects, etc.

[0063] Traffic data: including traffic flow, vehicle density, average speed, etc. This information helps to assess the current traffic conditions.

[0064] Signal data: mainly refers to the status of traffic lights, such as red light, green light, yellow light, etc., which has a direct impact on vehicle driving.

[0065] Weather data: Weather conditions such as temperature, humidity, rainfall, snowfall, fog, etc., which have a significant impact on vehicle driving safety and sensor performance.

[0066] Lighting condition data: including the light intensity and direction of the light source in the current environment, which is crucial to the performance of visual sensors such as cameras.

[0067] Vehicle dynamics data: involves the vehicle's motion state, such as speed, acceleration, steering angle, braking force, etc.

[0068] Vehicle location data: includes the vehicle’s latitude and longitude coordinates, altitude, etc., which helps determine the specific location of the vehicle.

[0069] Vehicle sensor data: Real-time data from various sensors on the vehicle, such as radar, lidar, camera, temperature sensor, pressure sensor, etc., which directly reflect the real-time status of the vehicle and its surrounding environment.

[0070] Timestamp data: Recording the timestamp of each piece of data improves the timing and synchronization of the data, which is crucial for subsequent data analysis and processing.

[0071] By acquiring these detailed basic vehicle data, the present invention can fully understand the real-time status of the vehicle and its surrounding environment, providing a solid foundation for subsequent intelligent monitoring and fault prediction. At the same time, the multi-source and diversity of these data also require the system to have efficient data fusion and processing capabilities, which is one of the key issues to be solved by the present invention.

[0072] Step S102, collecting the real-time position of the vehicle, determining the natural environment of the vehicle according to the real-time position of the vehicle, establishing a correspondence between the real-time position of the vehicle and the corresponding natural environment data and the corresponding vehicle driving and operation data, and obtaining a vehicle driving data set;

[0073] In step S102, the present invention collects the real-time location information of the vehicle and determines the natural environment in which the vehicle is currently located based on the location information. Subsequently, the system establishes a corresponding relationship between the real-time location of the vehicle and the corresponding natural environment data and the operation data during the vehicle's driving process, thereby forming a comprehensive vehicle driving data set. The specific implementation process of this step is as follows:

[0074] The global positioning system (GPS) or other positioning technologies are used to obtain the vehicle's longitude, latitude, altitude and other location information in real time.

[0075] Based on the real-time location information of the vehicle, combined with the pre-established geographic environment database or real-time weather and environmental data services, the natural environment in which the vehicle is currently located is determined. This includes but is not limited to weather conditions (such as sunny, rainy, snowy, etc.), lighting conditions (such as daytime, nighttime, backlight, etc.), road conditions (such as dry, slippery, potholes, etc.), and special environmental factors (such as construction areas, animal-infested areas, etc.).

[0076] The real-time position of the vehicle is associated with the natural environment data determined above, and combined with the real-time operation data of the vehicle during driving (such as speed, acceleration, steering angle, braking force, etc.), to form a comprehensive vehicle driving data set. This set contains comprehensive information about the vehicle's current position, surrounding environment, and vehicle driving status.

[0077] Through the above steps, the system can generate a vehicle driving data set containing rich information. This set not only records the location and driving status of the vehicle, but also associates the relevant natural environment data, providing comprehensive data support for subsequent data analysis, intelligent monitoring and fault prediction.

[0078] Through the implementation of step S102, the present invention can realize comprehensive perception and real-time monitoring of the vehicle driving environment, laying a solid foundation for subsequent data processing and analysis. At the same time, this also reflects the advantages of the present invention in data fusion and real-time processing, which helps to improve the accuracy and real-time performance of vehicle intelligent monitoring.

[0079] Step S103, matching the data in the vehicle driving data set with the preset vehicle driving parameters, and grouping the matched data to obtain vehicle driving group data, and retrieving the corresponding electro-hydraulic brake valve operation data in the database according to the time period range generated by the vehicle driving group data;

[0080] In step S103, the present invention performs a series of processing on the data in the vehicle driving data set, aiming to provide more accurate and targeted data support for subsequent intelligent monitoring and fault prediction through steps such as data type matching, data grouping, and database retrieval. The specific steps are as follows:

[0081] The system will first perform type matching on the data in the vehicle driving data set. The purpose of this step is to compare the data in the set with the preset vehicle driving parameters to improve the accuracy and consistency of the data. The preset vehicle driving parameters include speed range, acceleration threshold, steering angle limit, braking force standard, etc. These data are usually set based on the vehicle's technical specifications and driving safety requirements.

[0082] After completing the data type matching, the system will group the data according to the matching results. The basis for grouping includes various factors such as driving speed, driving time, and vehicle status. For example, the system will separate high-speed driving data from low-speed driving data, or group data from different time periods (such as morning and evening peak hours and non-peak hours). Through data grouping, the system can better understand and analyze the performance of the vehicle under different driving conditions.

[0083] After the above steps, the system will generate one or more groups of vehicle driving group data. These data not only contain the original driving information, but also have classification and labeling according to preset parameters, making subsequent data processing and analysis more efficient and accurate.

[0084] According to the time period range generated by the vehicle driving group data, the system will further retrieve the corresponding electro-hydraulic brake valve operation data in the database. The electro-hydraulic brake valve is an important part of the vehicle braking system, and its operating status directly affects the vehicle's driving safety and braking performance. By retrieving the electro-hydraulic brake valve operation data that matches the current driving conditions, the system can have a more comprehensive understanding of the actual performance of the vehicle's braking system, providing strong support for subsequent intelligent monitoring and fault prediction.

[0085] Through the implementation of step S103, the present invention realizes the refined processing and classification of vehicle driving data and the accurate retrieval of electro-hydraulic brake valve operation data, which not only improves the efficiency and accuracy of data processing, but also provides more abundant and targeted data support for subsequent intelligent monitoring and fault prediction.

[0086] Step S104, using the electro-hydraulic brake valve operation data and the data in the vehicle driving data set to train the neural network to obtain a vehicle intelligent monitoring model, substituting the real-time vehicle basic data into the vehicle intelligent monitoring model, running the vehicle intelligent monitoring model to output the vehicle intelligent monitoring results;

[0087] In step S104, the present invention trains a neural network by using the electro-hydraulic brake valve operation data and the data in the vehicle driving data set, thereby constructing a vehicle intelligent monitoring model. The model can analyze the vehicle's operating status in real time and output intelligent monitoring results. The specific steps are as follows:

[0088] The electro-hydraulic brake valve operation data retrieved in step S103 is integrated with the data in the vehicle driving data set. These data cover the braking performance, driving speed, acceleration, steering angle, and environmental perception of the vehicle under different driving conditions, providing comprehensive data support for the training of the neural network.

[0089] The integrated data is used to train a neural network. A neural network is a powerful machine learning model that can model and predict complex systems by learning the inherent laws and patterns of data. In the present invention, a neural network is used to learn the correlation between the vehicle driving state and the performance of the electro-hydraulic brake valve, and how these relationships change with changes in environmental conditions and vehicle status.

[0090] After training, the neural network can build a vehicle intelligent monitoring model. This model can receive real-time basic vehicle data as input, including current environmental perception data, vehicle status data, and timestamp data, and analyze and process this data in real time based on the learned rules and patterns.

[0091] Substitute the basic vehicle data acquired in real time into the vehicle intelligent monitoring model, which will quickly process the data and output the vehicle intelligent monitoring results. These results include the vehicle’s current driving status assessment, potential fault warning, braking performance, etc.

[0092] Finally, the vehicle intelligent monitoring model will output the intelligent monitoring results in an easy-to-understand form to the driver or related management system. These results are presented in the form of graphics, text or alarm signals, etc., to help the driver understand the vehicle's operating status in a timely manner and take corresponding measures to improve driving safety.

[0093] Through the implementation of step S104, the present invention successfully constructs a vehicle intelligent monitoring model based on machine learning, which can analyze the vehicle's operating status in real time and output accurate intelligent monitoring results, providing a strong guarantee for the safe driving of the vehicle. At the same time, this step also reflects the technical innovation and advantages of the present invention in data fusion, machine learning algorithm application and real-time intelligent monitoring.

[0094] Step S105, substitute the vehicle intelligent monitoring result into the vehicle operation fault prediction model to obtain the vehicle operation fault prediction result. If there is fault information in the vehicle operation fault prediction result, the fault information in the vehicle operation fault prediction result is matched in a preset vehicle drive parameter knowledge base to obtain the vehicle drive parameters corresponding to the fault information, transmit the vehicle drive parameters corresponding to the fault information to the vehicle drive system, and obtain feedback information of the vehicle drive system executing the vehicle drive parameters corresponding to the fault information.

[0095] In step S105, the present invention realizes accurate prediction and timely processing of potential vehicle faults by combining the vehicle intelligent monitoring results with the vehicle operation fault prediction model. The specific steps are as follows:

[0096] The vehicle intelligent monitoring results obtained in step S104 are substituted into the vehicle operation fault prediction model. The model is based on a large amount of historical data and machine learning algorithms, and can analyze abnormal indicators and trends in the current operation state of the vehicle and predict the type and time of future faults.

[0097] If there is fault information in the vehicle operation fault prediction results, the system will immediately start the fault handling process. First, the fault information will be matched in the preset vehicle drive parameter knowledge base. This knowledge base contains the correspondence between various fault types and vehicle drive parameters, and the system can quickly locate the vehicle drive parameters that match the current fault information.

[0098] When the vehicle driving parameters corresponding to the fault information are matched, the system will transmit these parameters to the vehicle driving system. After receiving the parameters, the vehicle driving system will adjust the vehicle's operating status according to the instructions, such as adjusting the speed, steering angle, braking force, etc., to avoid the occurrence of the fault or reduce the impact of the fault.

[0099] After the vehicle drive system executes the vehicle drive parameters corresponding to the fault information, it will feedback the execution results to the system. This feedback information includes the adjusted vehicle status data, the actual application effect of the drive parameters, etc. By collecting and analyzing this feedback information, the system can further verify the accuracy of fault prediction and optimize the fault handling process.

[0100] Throughout the process, the system will continue to monitor the vehicle's operating status and make necessary adjustments based on actual conditions. If the fault information persists or becomes more serious, the system will repeat the above steps until the fault is completely resolved or the vehicle reaches a safe operating state.

[0101] Through the implementation of step S105, the present invention not only achieves accurate prediction and timely processing of potential vehicle faults, but also continuously optimizes the fault handling process and improves prediction accuracy through the collection and analysis of feedback information. This step fully demonstrates the technical innovation and practical application value of the present invention in vehicle intelligent monitoring and fault prediction.

[0102] Specifically, the vehicle intelligent monitoring method based on machine learning of the present invention, said S103, comprises:

[0103] Performing cluster analysis on feedback information of vehicle driving parameters to obtain cluster analysis results after the vehicle executes the driving parameters, wherein the cluster analysis results after the vehicle executes the driving parameters include parameter data at different time periods and under different natural environments;

[0104] The cluster analysis results after the vehicle executes the driving parameters are matched in a preset vehicle operation standard parameter knowledge base to determine the type of vehicle operation parameters to be compared, and the vehicle operation parameter types to be compared are matched in the preset vehicle operation standard parameter knowledge base to obtain the value of the vehicle operation standard parameter to be compared;

[0105] The feedback information of the vehicle driving parameters is compared with the standard operating parameter values ​​of the vehicle to be compared to obtain a vehicle operating parameter comparison result, which includes parameter data that meets the standard and parameter data that does not meet the standard.

[0106] In step S103, the present invention further processes and analyzes the vehicle driving data, especially the feedback information of the vehicle driving parameters, and step S103 improves the accuracy and safety of vehicle operation. The specific steps are as follows:

[0107] First, cluster analysis is performed on the feedback information of vehicle driving parameters. Cluster analysis is an unsupervised learning algorithm that can classify similar data points into one category. In this step, cluster analysis is used to process the driving parameter data of the vehicle at different time periods and in different natural environments, so as to obtain the cluster analysis results after the vehicle executes the driving parameters. These results help to reveal the operating characteristics and laws of the vehicle under different conditions.

[0108] Next, the cluster analysis results are matched with the preset vehicle operation standard parameter knowledge base. This knowledge base contains the standard parameter values ​​of the vehicle under various operating conditions and is an important basis for evaluating the vehicle's operating status. Through matching, the system can determine the type of vehicle operating parameters to be compared and find the corresponding standard parameter values ​​in the knowledge base.

[0109] Then, the feedback information of the vehicle driving parameters is compared with the vehicle operation standard parameter values ​​to be compared. This step is to evaluate whether the current operation status of the vehicle meets the standards. The comparison results will be classified into parameter data that meets the standards and parameter data that does not meet the standards. Data that meets the standards indicates that the current operation status of the vehicle is normal, while data that does not meet the standards means that there are potential problems or failures in the vehicle.

[0110] For data that does not meet the standard parameters, the system will conduct further analysis and processing. This includes issuing alarms, adjusting the vehicle's operating status, or suggesting that the driver take appropriate measures. At the same time, these data that do not meet the standards will also be recorded for subsequent fault prediction and model optimization.

[0111] Through the implementation of step S103, the present invention can more accurately evaluate the vehicle operating status, timely discover potential problems or faults, and take corresponding measures to deal with them.

[0112] Specifically, the vehicle intelligent monitoring method based on machine learning of the present invention, said S104, comprises:

[0113] Establishing vehicle operation time intervals, which include short-term and long-term operation time intervals. The short-term time interval ranges from 0 minutes to 30 minutes, and the long-term time interval ranges from 30 minutes to 600 minutes.

[0114] Screen the data of parameters that do not meet the standards, filter out the emergency braking information after the vehicle travels at a speed higher than 60 kilometers per hour and in a long period of time, and retrieve the response speed of the electro-hydraulic brake valve within the corresponding time range;

[0115] Retrieve the vehicle's external temperature data for natural environment information within the corresponding time range to obtain the real-time operating temperature of the electro-hydraulic brake valve;

[0116] Substitute the response speed of the electro-hydraulic brake valve within the corresponding time range, the vehicle external temperature data of the natural environment information within the corresponding time range, and the real-time working temperature of the electro-hydraulic brake valve into the vehicle operation fault prediction model to obtain a preset temperature threshold;

[0117] The real-time operating temperature of the electro-hydraulic brake valve is compared with a preset temperature threshold. If the real-time operating temperature exceeds the preset temperature threshold, it is determined that the insulation performance of the electromagnet coil has degraded, and a warning message is issued to the driver.

[0118] In step S104, the present invention further refines the training process of the vehicle intelligent monitoring model and introduces a method for in-depth analysis of the vehicle operating status under specific conditions. Specifically, this step includes the following key sub-steps:

[0119] In order to more accurately analyze the behavior characteristics of the vehicle at different operating time periods, the present invention first establishes the vehicle operating time interval, which is divided into two categories: "short time" and "long time". The short time interval is defined as 0 minutes to 30 minutes, while the long time interval is defined as 30 minutes to 600 minutes. This division helps to distinguish the performance differences of vehicles at different operating time periods.

[0120] Next, the system screens the data that does not meet the standard parameters, paying special attention to those cases where the vehicle speed is higher than 60 kilometers per hour for a long period of time and then an emergency brake occurs. For these screened data, the system further retrieves the response speed information of the electro-hydraulic brake valve within the corresponding time range. This step is intended to identify the performance of the brake system in emergency braking after long-term high-speed driving.

[0121] In addition to the performance data of the braking system, the system also retrieves the natural environment information within the corresponding time range, especially the vehicle's external temperature data. At the same time, the real-time operating temperature of the electro-hydraulic brake valve is obtained. These environmental data are crucial for understanding the impact of external conditions on the performance of the braking system.

[0122] The selected emergency braking information, the response speed of the electro-hydraulic brake valve, the vehicle's external temperature data, and the real-time operating temperature of the electro-hydraulic brake valve are substituted into the vehicle operation fault prediction model. The model analyzes the relationship between these parameters through a complex machine learning algorithm and sets a preset temperature threshold. This threshold represents the maximum temperature that the electro-hydraulic brake valve can withstand under different conditions.

[0123] Finally, the system compares the real-time operating temperature of the electro-hydraulic brake valve with the preset temperature threshold. If the real-time operating temperature exceeds this threshold, the system determines that the insulation performance of the electromagnet coil has degraded, which will affect the reliability and safety of the brake system. In this case, the system will immediately send a warning message to the driver, reminding him to pay attention to potential problems in the brake system and take corresponding maintenance or replacement measures.

[0124] Through the implementation of step S104, the present invention not only improves the accuracy and reliability of the vehicle intelligent monitoring model, but also can perform a more in-depth analysis of the vehicle operating status under specific conditions, thereby timely discovering and handling potential safety hazards.

[0125] Specifically, the vehicle intelligent monitoring method based on machine learning of the present invention, said S105, comprises:

[0126] By installing a micro displacement sensor inside the electro-hydraulic brake valve, the clearance between the valve core and the valve body is monitored in real time. Based on the clearance data obtained, a correlation model between the clearance change and the valve action performance is established using a machine learning algorithm to predict the remaining service life of the valve.

[0127] A pressure sensor is installed on the brake line to monitor the line pressure in real time. If the pressure is lower than the preset threshold, it is judged that there is a leak. The pressure change pattern is analyzed through a machine learning algorithm to determine the leak location.

[0128] In step S105, the present invention conducts in-depth monitoring and prediction on the electro-hydraulic brake valve, a key component of the vehicle braking system, in order to improve the safety and reliability of vehicle driving. Specifically, this step includes two main aspects:

[0129] In order to accurately predict the remaining service life of the electro-hydraulic brake valve, the present invention installs a micro displacement sensor inside the electro-hydraulic brake valve. The sensor can monitor the matching clearance between the valve core and the valve body in real time, which is a key indicator for evaluating the wear degree and performance status of the brake valve.

[0130] The acquired clearance data is then fed into a machine learning algorithm, which analyzes the data and builds a correlation model between clearance changes and valve performance. Through this model, the system can predict the remaining service life of the electro-hydraulic brake valve, thereby arranging maintenance or replacement in advance and avoiding safety hazards caused by brake system failure.

[0131] In addition, in order to promptly detect leakage problems in the brake pipeline, the present invention installs pressure sensors on the brake pipeline. These sensors can monitor the pressure changes in the pipeline in real time and improve the normal operation of the brake system.

[0132] When the system detects that the line pressure is lower than the preset threshold, it will immediately determine that there is a leak. At this time, the machine learning algorithm will further analyze the pressure change pattern and determine the specific location of the leak through complex algorithm calculations. This function is crucial for quickly locating and repairing leaks, which can significantly reduce vehicle downtime and repair costs caused by brake system failures.

[0133] Through the implementation of step S105, the present invention not only improves the reliability and safety of the vehicle braking system, but also reduces the maintenance cost and time through real-time monitoring and prediction technology. This intelligent monitoring method based on machine learning brings a new solution to the field of vehicle engineering, which helps to promote the further application and development of intelligent control technology in vehicle engineering.

[0134] In a second aspect, the present invention provides a vehicle intelligent monitoring system based on machine learning, comprising:

[0135] A data acquisition unit, used to acquire basic vehicle data, the basic vehicle data including environmental perception data, vehicle status data and timestamp data, the environmental perception data including road data, obstacle data, traffic data, signal data, weather data and lighting condition data, the vehicle status data including vehicle dynamics data, vehicle position data and vehicle sensor data;

[0136] A data reorganization unit is used to collect the real-time position of the vehicle, determine the natural environment of the vehicle according to the real-time position of the vehicle, establish a corresponding relationship between the real-time position of the vehicle and the corresponding natural environment data and the corresponding vehicle driving and operation data, and obtain a vehicle driving data set;

[0137] A data retrieval unit is used to match the data types of the data in the vehicle driving data set with the preset vehicle driving parameters, and group the matched data to obtain vehicle driving group data, and retrieve the corresponding electro-hydraulic brake valve operation data in the database according to the time period range generated by the vehicle driving group data;

[0138] The vehicle operation monitoring unit is used to train the neural network using the electro-hydraulic brake valve operation data and the data in the vehicle driving data set to obtain a vehicle intelligent monitoring model, substitute the vehicle basic data obtained in real time into the vehicle intelligent monitoring model, and run the vehicle intelligent monitoring model to output the vehicle intelligent monitoring results;

[0139] The vehicle drive unit is used to substitute the vehicle intelligent monitoring results into the vehicle operation fault prediction model to obtain the vehicle operation fault prediction results. If there is fault information in the vehicle operation fault prediction results, the fault information in the vehicle operation fault prediction results is matched in a preset vehicle drive parameter knowledge base to obtain the vehicle drive parameters corresponding to the fault information, and the vehicle drive parameters corresponding to the fault information are transmitted to the vehicle drive system to obtain feedback information of the vehicle drive system executing the vehicle drive parameters corresponding to the fault information.

[0140] Specifically, in the vehicle intelligent monitoring system based on machine learning of the present invention, the data retrieval unit is further used for:

[0141] Performing cluster analysis on feedback information of vehicle driving parameters to obtain cluster analysis results after the vehicle executes the driving parameters, wherein the cluster analysis results after the vehicle executes the driving parameters include parameter data at different time periods and under different natural environments;

[0142] The cluster analysis results after the vehicle executes the driving parameters are matched in a preset vehicle operation standard parameter knowledge base to determine the type of vehicle operation parameters to be compared, and the vehicle operation parameter types to be compared are matched in the preset vehicle operation standard parameter knowledge base to obtain the value of the vehicle operation standard parameter to be compared;

[0143] The feedback information of the vehicle driving parameters is compared with the standard operating parameter values ​​of the vehicle to be compared to obtain a vehicle operating parameter comparison result, which includes parameter data that meets the standard and parameter data that does not meet the standard.

[0144] Specifically, in the vehicle intelligent monitoring system based on machine learning of the present invention, the vehicle operation monitoring unit is further used for:

[0145] Establishing vehicle operation time intervals, which include short time and long time. The short time interval ranges from 0 minutes to 30 minutes, and the long time interval ranges from 30 minutes to 600 minutes.

[0146] Screen the data of parameters that do not meet the standards, screen out the emergency braking information after the vehicle travels at a speed higher than 60 kilometers per hour and in a long period of time, and retrieve the response speed of the electro-hydraulic brake valve within the corresponding time range;

[0147] Retrieve the vehicle's external temperature data for natural environment information within the corresponding time range to obtain the real-time operating temperature of the electro-hydraulic brake valve;

[0148] Substitute the response speed of the electro-hydraulic brake valve within the corresponding time range, the vehicle external temperature data of the natural environment information within the corresponding time range, and the real-time working temperature of the electro-hydraulic brake valve into the vehicle operation fault prediction model to obtain a preset temperature threshold;

[0149] The real-time operating temperature of the electro-hydraulic brake valve is compared with a preset temperature threshold. If the real-time operating temperature exceeds the preset temperature threshold, it is determined that the insulation performance of the electromagnet coil has degraded, and a warning message is sent to the driver.

[0150] Specifically, in the vehicle intelligent monitoring system based on machine learning of the present invention, the vehicle drive unit is further used for:

[0151] By installing a micro displacement sensor inside the electro-hydraulic brake valve, the clearance between the valve core and the valve body is monitored in real time. Based on the clearance data obtained, a correlation model between the clearance change and the valve action performance is established using a machine learning algorithm to predict the remaining service life of the valve.

[0152] A pressure sensor is installed on the brake line to monitor the line pressure in real time. If the pressure is lower than the preset threshold, it is judged that there is a leak. The pressure change pattern is analyzed through a machine learning algorithm to determine the leak location.

[0153] The technical solution provided by the present invention effectively solves the problem that the existing vehicle intelligent monitoring methods cannot efficiently and real-time integrate the multi-source data collected during vehicle driving and perform accurate analysis and processing through a vehicle intelligent monitoring method and system based on machine learning. The specific technical solution is as follows:

[0154] Obtain basic vehicle data, including environmental perception data (such as road data, obstacle data, traffic data, signal data, weather data, and lighting condition data) and vehicle status data (such as vehicle dynamics data, vehicle location data, and vehicle sensor data), as well as timestamp data.

[0155] Collect the real-time location of the vehicle, determine the natural environment of the vehicle based on the real-time location of the vehicle, establish a corresponding relationship between the real-time location of the vehicle and the corresponding natural environment data and vehicle driving data, and obtain a vehicle driving data set. This step realizes the initial fusion of multi-source data.

[0156] The data in the vehicle driving data set is matched with the preset vehicle driving parameters, and the matched data is grouped to obtain vehicle driving group data. According to the time period range generated by the vehicle driving group data, the corresponding electro-hydraulic brake valve operation data in the database is retrieved. This step improves the efficiency and accuracy of data processing through data matching and grouping.

[0157] The neural network is trained using the data in the electro-hydraulic brake valve operation data and the vehicle driving data set to obtain the vehicle intelligent monitoring model. The basic vehicle data obtained in real time is substituted into the vehicle intelligent monitoring model, and the model is run to output the vehicle intelligent monitoring results. This step uses machine learning algorithms, especially neural networks, to deeply analyze and process multi-source data, realizing efficient and real-time intelligent monitoring.

[0158] Substitute the vehicle intelligent monitoring results into the vehicle operation fault prediction model to obtain the vehicle operation fault prediction results. If there is fault information, match it in the preset vehicle drive parameter knowledge base to obtain the vehicle drive parameters corresponding to the fault information, and transmit these parameters to the vehicle drive system. After the vehicle drive system executes the corresponding drive parameters, it feeds back the execution information. This step improves the safety and stability of the vehicle through fault prediction and dynamic adjustment of drive parameters.

[0159] In steps S103, S104 and S105, the present invention also includes a series of optimization measures, such as cluster analysis, establishing vehicle operation time intervals, screening out parameter data that does not meet the standards, predicting the remaining service life of the electro-hydraulic brake valve, and determining the location of brake line leakage, etc., which further improve the accuracy of data processing and the predictive ability of the model.

[0160] To summarize, the present invention effectively solves the problem that existing vehicle intelligent monitoring methods are unable to efficiently and real-time fuse and process multi-source data through a series of steps such as acquiring multi-source data, performing data fusion and reorganization, matching and retrieval, model training and intelligent monitoring, fault prediction and drive parameter adjustment, thereby improving the accuracy and real-time performance of vehicle intelligent monitoring.

[0161] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations, and the above-described embodiments of the present invention do not constitute a limitation on the scope of protection of the present invention.

Claims

1. A vehicle intelligent monitoring method based on machine learning, characterized in that: include: Step S101, acquiring basic vehicle data, the basic vehicle data including environmental perception data, vehicle status data and timestamp data, the environmental perception data including road data, obstacle data, traffic data, signal data, weather data and lighting condition data, the vehicle status data including vehicle dynamics data, vehicle position data and vehicle sensor data; Step S102, collecting the real-time position of the vehicle, determining the natural environment of the vehicle according to the real-time position of the vehicle, establishing a correspondence between the real-time position of the vehicle and the corresponding natural environment data and the corresponding vehicle driving and operation data, and obtaining a vehicle driving data set; Step S103, matching the data in the vehicle driving data set with the preset vehicle driving parameters, and grouping the matched data to obtain vehicle driving group data, and retrieving the corresponding electro-hydraulic brake valve operation data in the database according to the time period range generated by the vehicle driving group data; Step S104, using the electro-hydraulic brake valve operation data and the data in the vehicle driving data set to train the neural network to obtain a vehicle intelligent monitoring model, substituting the real-time vehicle basic data into the vehicle intelligent monitoring model, running the vehicle intelligent monitoring model to output the vehicle intelligent monitoring results; Step S105, substitute the vehicle intelligent monitoring result into the vehicle operation fault prediction model to obtain the vehicle operation fault prediction result. If there is fault information in the vehicle operation fault prediction result, the fault information in the vehicle operation fault prediction result is matched in a preset vehicle drive parameter knowledge base to obtain the vehicle drive parameters corresponding to the fault information, transmit the vehicle drive parameters corresponding to the fault information to the vehicle drive system, and obtain feedback information of the vehicle drive system executing the vehicle drive parameters corresponding to the fault information.

2. The vehicle intelligent monitoring method based on machine learning as claimed in claim 1, characterized in that: The S103 includes: Performing cluster analysis on feedback information of vehicle driving parameters to obtain cluster analysis results after the vehicle executes the driving parameters, wherein the cluster analysis results after the vehicle executes the driving parameters include parameter data at different time periods and under different natural environments; The cluster analysis results after the vehicle executes the driving parameters are matched in a preset vehicle operation standard parameter knowledge base to determine the type of vehicle operation parameters to be compared, and the vehicle operation parameter types to be compared are matched in the preset vehicle operation standard parameter knowledge base to obtain the value of the vehicle operation standard parameter to be compared; The feedback information of the vehicle driving parameters is compared with the standard operating parameter values ​​of the vehicle to be compared to obtain a vehicle operating parameter comparison result, which includes parameter data that meets the standard and parameter data that does not meet the standard.

3. The vehicle intelligent monitoring method based on machine learning as claimed in claim 1, characterized in that: The S104 includes: Establishing vehicle operation time intervals, which include short-term and long-term operation time intervals. The short-term time interval ranges from 0 minutes to 30 minutes, and the long-term time interval ranges from 30 minutes to 600 minutes. Screen the data of parameters that do not meet the standards, filter out the emergency braking information after the vehicle travels at a speed higher than 60 kilometers per hour and in a long period of time, and retrieve the response speed of the electro-hydraulic brake valve within the corresponding time range; Retrieve the vehicle's external temperature data for natural environment information within the corresponding time range to obtain the real-time operating temperature of the electro-hydraulic brake valve; Substitute the response speed of the electro-hydraulic brake valve within the corresponding time range, the vehicle external temperature data of the natural environment information within the corresponding time range, and the real-time working temperature of the electro-hydraulic brake valve into the vehicle operation fault prediction model to obtain a preset temperature threshold; The real-time operating temperature of the electro-hydraulic brake valve is compared with a preset temperature threshold. If the real-time operating temperature exceeds the preset temperature threshold, it is determined that the insulation performance of the electromagnet coil has degraded, and a warning message is issued to the driver.

4. The vehicle intelligent monitoring method based on machine learning as claimed in claim 1, characterized in that: The S105 includes: By installing a micro displacement sensor inside the electro-hydraulic brake valve, the clearance between the valve core and the valve body is monitored in real time. Based on the clearance data obtained, a correlation model between the clearance change and the valve action performance is established using a machine learning algorithm to predict the remaining service life of the valve. A pressure sensor is installed on the brake line to monitor the line pressure in real time. If the pressure is lower than the preset threshold, it is judged that there is a leak. The pressure change pattern is analyzed through a machine learning algorithm to determine the leak location.

5. A vehicle intelligent monitoring system based on machine learning, characterized in that: include: A data acquisition unit, used to acquire basic vehicle data, the basic vehicle data including environmental perception data, vehicle status data and timestamp data, the environmental perception data including road data, obstacle data, traffic data, signal data, weather data and lighting condition data, the vehicle status data including vehicle dynamics data, vehicle position data and vehicle sensor data; A data reorganization unit is used to collect the real-time position of the vehicle, determine the natural environment of the vehicle according to the real-time position of the vehicle, establish a corresponding relationship between the real-time position of the vehicle and the corresponding natural environment data and the corresponding vehicle driving and operation data, and obtain a vehicle driving data set; A data retrieval unit is used to match the data types of the data in the vehicle driving data set with the preset vehicle driving parameters, and group the matched data to obtain vehicle driving group data, and retrieve the corresponding electro-hydraulic brake valve operation data in the database according to the time period range generated by the vehicle driving group data; The vehicle operation monitoring unit is used to train the neural network using the electro-hydraulic brake valve operation data and the data in the vehicle driving data set to obtain a vehicle intelligent monitoring model, substitute the vehicle basic data obtained in real time into the vehicle intelligent monitoring model, and run the vehicle intelligent monitoring model to output the vehicle intelligent monitoring results; The vehicle drive unit is used to substitute the vehicle intelligent monitoring results into the vehicle operation fault prediction model to obtain the vehicle operation fault prediction results. If there is fault information in the vehicle operation fault prediction results, the fault information in the vehicle operation fault prediction results is matched in a preset vehicle drive parameter knowledge base to obtain the vehicle drive parameters corresponding to the fault information, and the vehicle drive parameters corresponding to the fault information are transmitted to the vehicle drive system to obtain feedback information of the vehicle drive system executing the vehicle drive parameters corresponding to the fault information.

6. The vehicle intelligent monitoring system based on machine learning as claimed in claim 5, characterized in that: The data retrieval unit is further used for: Performing cluster analysis on feedback information of vehicle driving parameters to obtain cluster analysis results after the vehicle executes the driving parameters, wherein the cluster analysis results after the vehicle executes the driving parameters include parameter data at different time periods and under different natural environments; The cluster analysis results after the vehicle executes the driving parameters are matched in a preset vehicle operation standard parameter knowledge base to determine the type of vehicle operation parameters to be compared, and the vehicle operation parameter types to be compared are matched in the preset vehicle operation standard parameter knowledge base to obtain the value of the vehicle operation standard parameter to be compared; The feedback information of the vehicle driving parameters is compared with the standard operating parameter values ​​of the vehicle to be compared to obtain a vehicle operating parameter comparison result, which includes parameter data that meets the standard and parameter data that does not meet the standard.

7. The vehicle intelligent monitoring system based on machine learning as claimed in claim 5, characterized in that: The vehicle operation monitoring unit is further used for: Establishing vehicle operation time intervals, which include short-term and long-term operation time intervals. The short-term time interval ranges from 0 minutes to 30 minutes, and the long-term time interval ranges from 30 minutes to 600 minutes. Screen the data of parameters that do not meet the standards, filter out the emergency braking information after the vehicle travels at a speed higher than 60 kilometers per hour and in a long period of time, and retrieve the response speed of the electro-hydraulic brake valve within the corresponding time range; Retrieve the vehicle's external temperature data for natural environment information within the corresponding time range to obtain the real-time operating temperature of the electro-hydraulic brake valve; Substitute the response speed of the electro-hydraulic brake valve within the corresponding time range, the vehicle external temperature data of the natural environment information within the corresponding time range, and the real-time working temperature of the electro-hydraulic brake valve into the vehicle operation fault prediction model to obtain a preset temperature threshold; The real-time operating temperature of the electro-hydraulic brake valve is compared with a preset temperature threshold. If the real-time operating temperature exceeds the preset temperature threshold, it is determined that the insulation performance of the electromagnet coil has degraded, and a warning message is issued to the driver.

8. The vehicle intelligent monitoring system based on machine learning as claimed in claim 5, characterized in that: The vehicle drive unit is further used for: By installing a micro displacement sensor inside the electro-hydraulic brake valve, the clearance between the valve core and the valve body is monitored in real time. Based on the clearance data obtained, a correlation model between the clearance change and the valve action performance is established using a machine learning algorithm to predict the remaining service life of the valve. A pressure sensor is installed on the brake line to monitor the line pressure in real time. If the pressure is lower than the preset threshold, it is judged that there is a leak. The pressure change pattern is analyzed through a machine learning algorithm to determine the leak location.

Images