A vehicle posture control optimization system based on electro-hydraulic brake valve

Abstract

The present invention relates to the technical field of vehicle control systems. The present invention provides a vehicle posture control optimization system based on an electro-hydraulic brake valve, including units such as data acquisition, driving mode recognition, electro-hydraulic brake valve adjustment simulation, vehicle driving simulation, data comparison and control optimization. The system recognizes the driving mode in real time through a deep learning model, simulates the electro-hydraulic brake valve adjustment strategy according to the driving mode and vehicle driving state data, and verifies the adjustment effect using a vehicle driving simulation model. If the simulation result is inconsistent with the real-time data, the model parameters are optimized through a genetic algorithm. The system can adapt to different driving modes and road conditions, dynamically adjust the control strategy, improve the electro-hydraulic brake valve adjustment smoothness and fuel efficiency, and provide a stable power drive for the vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle control systems, and in particular to a vehicle posture control optimization system based on an electro-hydraulic brake valve. Background Art

[0002] An electro-hydraulic brake valve is a device that converts electrical signals into hydraulic energy. It combines the advantages of electronic control and hydraulic transmission. In the braking system, the electro-hydraulic brake valve receives signals from the electronic control unit (ECU) and uses the magnetic field generated by the electromagnet to control the position of the valve core, thereby adjusting the flow direction and flow rate of the hydraulic oil, thereby achieving precise control of the brake pressure and applying it to the vehicle braking system.

[0003] During the adjustment process of the vehicle's electro-hydraulic brake valve, the electro-hydraulic brake valve achieves smooth control by precisely regulating the pressure and flow of the hydraulic system. When receiving an electro-hydraulic brake valve adjustment command from the vehicle's transmission control unit (TCU) or electronic control unit (ECU), the electro-hydraulic brake valve will respond quickly, drive the structure inside the valve to change through electromagnetic force, adjust the flow direction and pressure of the brake fluid, and then control the engagement and disengagement of the clutch or brake, so as to achieve electro-hydraulic brake valve adjustment and improve the vehicle's fuel efficiency.

[0004] The vehicle posture control system of the electro-hydraulic brake valve has the following technical pain points during actual use: there are multiple driving modes during vehicle driving, including economic mode, normal mode and sports mode. Different driving modes have different requirements for the control of the electro-hydraulic brake valve. The switching of driving modes will directly affect the control logic and parameter settings of the electro-hydraulic brake valve. When driving on rough roads, road conditions and vehicle loads also affect the control of the electro-hydraulic brake valve. The electro-hydraulic brake valve needs to adjust the control strategy more frequently. The existing vehicle posture control system of the electro-hydraulic brake valve cannot adapt to the ever-changing vehicle dynamic adjustments, resulting in the inability to provide stable power drive for the vehicle. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a vehicle posture control optimization system based on an electro-hydraulic brake valve to solve the problem that the existing vehicle posture control system of the electro-hydraulic brake valve cannot adapt to the ever-changing vehicle dynamic adjustment, resulting in the inability to provide stable power drive for the vehicle.

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

[0007] The present invention provides a vehicle posture control optimization system based on an electro-hydraulic brake valve, comprising:

[0008] A data acquisition unit, used to acquire vehicle driving status historical data, driving mode historical data, electro-hydraulic brake valve adjustment historical data, electro-hydraulic brake valve status historical data, road condition historical data, and vehicle load historical data;

[0009] A driving mode recognition unit is used to preprocess the vehicle driving state historical data and the driving mode historical data, use the preprocessed driving mode historical data to train the deep learning model to obtain the driving mode recognition model, obtain the real-time vehicle driving state data, substitute the vehicle driving state data into the driving mode recognition model, and obtain the real-time driving mode recognition result;

[0010] The electro-hydraulic brake valve adjustment simulation unit is used to match the real-time driving mode recognition result in a preset electro-hydraulic brake valve adjustment model database to obtain a real-time electro-hydraulic brake valve adjustment simulation model, substitute the real-time vehicle driving state data into the real-time electro-hydraulic brake valve adjustment simulation model to obtain real-time electro-hydraulic brake valve adjustment simulation data, generate real-time electro-hydraulic brake valve adjustment simulation data of different time periods according to a preset time period, and establish corresponding timestamps for the real-time electro-hydraulic brake valve adjustment simulation data of different time periods;

[0011] A vehicle driving simulation unit is used to obtain real-time vehicle driving data, perform feature extraction on the real-time vehicle driving data, and obtain real-time vehicle driving feature data, the real-time vehicle driving feature data includes real-time vehicle driving state data, driving mode data, electro-hydraulic brake valve adjustment data, electro-hydraulic brake valve state data, road condition data, and vehicle load data, perform cluster analysis on the real-time vehicle driving feature data, and obtain cluster analysis results of the real-time vehicle driving feature data, match the cluster analysis results of the real-time vehicle driving feature data in a preset algorithm knowledge base, and obtain an algorithm model corresponding to the real-time vehicle driving feature data, and use the real-time vehicle driving feature data to train the algorithm model to obtain a vehicle driving simulation model;

[0012] The data comparison unit is used to obtain vehicle external image data during vehicle driving, pre-process the vehicle external image data during vehicle driving to obtain pre-processed vehicle external image data during vehicle driving, identify the pre-processed vehicle external image data during vehicle driving through a preset natural environment image recognition model to obtain vehicle external image recognition results during vehicle driving, establish corresponding timestamps for the vehicle external image recognition results during vehicle driving, randomly generate data retrieval time information, and adjust the real-time electro-hydraulic brake valve simulation data during different time periods and the vehicle driving process according to the data retrieval time information. Based on the timestamp in the vehicle external image recognition result, the data corresponding to the data retrieval time information is retrieved to obtain the first data to be compared, and according to the time information of the first data to be compared, the vehicle driving history data corresponding to the time information of the first data to be compared is retrieved, and the vehicle driving history data corresponding to the time information of the first data to be compared is substituted into the vehicle driving simulation model to obtain the vehicle driving simulation data corresponding to the time information of the first data to be compared, and the vehicle driving simulation data corresponding to the time information of the first data to be compared is used as the second data to be compared, and the first data to be compared is compared with the second data to be compared to obtain the comparison result of the first data to be compared with the second data to be compared;

[0013] A control optimization unit is used to analyze the real-time vehicle driving data using a vehicle driving simulation model if the comparison results of the first data to be compared are consistent with the second data to be compared, and substitute the output results into a preset electro-hydraulic brake valve drive control model to output real-time electro-hydraulic brake valve drive parameters, and use the real-time electro-hydraulic brake valve drive parameters as control parameters of the electro-hydraulic brake valve during the adjustment of the vehicle electro-hydraulic brake valve; if the comparison results of the first data to be compared are inconsistent with the second data to be compared, use the comparison results of the first data to be compared with the second data to be compared, vehicle driving status historical data, driving mode historical data, electro-hydraulic brake valve adjustment historical data, and electro-hydraulic brake valve status historical data The vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model and the driving mode recognition model are optimized by genetic algorithm based on the historical data of road conditions and the historical data of vehicle load to obtain the optimized vehicle driving simulation model, the optimized real-time electro-hydraulic brake valve adjustment simulation model and the optimized driving mode recognition model. The optimized vehicle driving simulation model, the optimized real-time electro-hydraulic brake valve adjustment simulation model and the optimized driving mode recognition model are used to process the data to obtain the optimized real-time electro-hydraulic brake valve driving parameters, and the optimized real-time electro-hydraulic brake valve driving parameters are used as the control parameters of the electro-hydraulic brake valve in the optimized vehicle electro-hydraulic brake valve adjustment process.

[0014] Furthermore, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the driving mode recognition unit is also used for:

[0015] Obtain historical data on vehicle driving status and driving mode, clean the historical data on vehicle driving status and driving mode, remove outliers and missing values, normalize the historical data on vehicle driving status and driving mode, use the preprocessed driving mode historical data as input and the corresponding driving mode as output label, train the selected deep learning model to obtain a trained driving mode recognition model, and deploy the trained driving mode recognition model to the system for real-time recognition of driving modes.

[0016] Furthermore, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the electro-hydraulic brake valve adjustment simulation unit is also used for:

[0017] Determine a preset time period, the preset time period includes a fixed time interval, and match a corresponding electro-hydraulic brake valve adjustment simulation model in a preset electro-hydraulic brake valve adjustment model database according to the real-time driving mode recognition result, the preset electro-hydraulic brake valve adjustment model database includes electro-hydraulic brake valve adjustment models corresponding to multiple driving modes;

[0018] Substituting the real-time acquired vehicle driving state data into the matched electro-hydraulic brake valve adjustment simulation model, and obtaining real-time electro-hydraulic brake valve adjustment simulation data through model calculation, the real-time electro-hydraulic brake valve adjustment simulation data including electro-hydraulic brake valve adjustment time, electro-hydraulic brake valve adjustment speed and electro-hydraulic brake valve adjustment smoothness data;

[0019] According to a preset time period, a corresponding time stamp is established for the generated real-time electro-hydraulic brake valve adjustment simulation data, and the time stamp is used to reflect the time point when the data is generated.

[0020] Furthermore, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the vehicle driving simulation unit is also used for:

[0021] The cluster analysis results are matched with the preset algorithm knowledge base to obtain an algorithm model that matches the real-time vehicle driving feature data. The algorithm knowledge base includes algorithm models under various driving modes. The matching process uses similarity calculation to match the most appropriate algorithm model.

[0022] The matched algorithm model is trained using real-time vehicle driving characteristic data, and the algorithm model is trained to simulate the dynamic characteristics and response behaviors of the vehicle during driving.

[0023] Furthermore, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the data comparison unit is also used for:

[0024] The vehicle's external image data during driving includes weather conditions, lighting conditions, and road types;

[0025] Using a preset annotation tool to annotate the vehicle's external image data during the vehicle's driving process, to obtain annotation information of the vehicle's external image data during the vehicle's driving process, wherein the annotation information of the vehicle's external image data during the vehicle's driving process includes road signs, pedestrians, vehicles, and obstacles;

[0026] Using a deep learning model as a training model, the labeled vehicle external image data during driving is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used to evaluate model performance and adjust hyperparameters during training, and the test set is used to finally evaluate model performance.

[0027] Receive and set model training hyperparameters, including learning rate, batch size, and number of training rounds;

[0028] The training set is input into the model for training. During the training process, the model parameters are adjusted through the back propagation algorithm to minimize the loss function value. The model performance is monitored using the validation set. If the validation passes, the trained natural environment image recognition model is exported into a deployable format and deployed.

[0029] Furthermore, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the control optimization unit is also used for:

[0030] The real-time vehicle driving data, electro-hydraulic brake valve status data, driving mode recognition results and sensor data are used as input parameters.

[0031] Real-time vehicle driving data includes vehicle speed, acceleration and engine speed, and sensor data includes brake pressure and throttle opening;

[0032] Using real-time electro-hydraulic brake valve driving parameters as output parameters, the real-time electro-hydraulic brake valve driving parameters including brake pressure setting value and brake response time;

[0033] Based on the physical principles and the working principle of the electro-hydraulic brake system, a mathematical model for the drive control of the electro-hydraulic brake valve is preliminarily established. The physical principles and the working principle of the electro-hydraulic brake system include the relationship between the brake pressure and the current, and the relationship between the brake response time and the oil pressure change rate.

[0034] The model parameters are optimized using the historical data of vehicle driving status and the historical data of electro-hydraulic brake valve adjustment. In the simulation environment, the model is verified using historical data to check whether the model output meets expectations, and the electro-hydraulic brake valve drive control model is obtained.

[0035] Furthermore, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the control optimization unit is also used for:

[0036] Initialize the population and randomly generate a set of initial solutions as the initial population of the genetic algorithm. The solutions are the model parameters.

[0037] Receive the optimization goal, and define a fitness function to evaluate the quality of each solution based on the optimization goal. The fitness function is constructed based on the error between the model prediction result and the actual data, the smoothness index of the electro-hydraulic brake valve adjustment, and the driving mode recognition accuracy;

[0038] Perform selection operations to retain individuals with higher fitness;

[0039] Perform crossover operations to combine the good genes of different individuals;

[0040] Perform mutation operations, introduce new genes, and increase the population size;

[0041] Repeat the selection, crossover and mutation operations until the preset number of iterations or fitness threshold is reached;

[0042] The genetic algorithm optimization framework is used to optimize the parameters of the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model, and the driving mode recognition model to obtain the optimized model parameters;

[0043] Use the optimized model parameters to update the model, verify the optimized model in a simulation environment or using actual vehicle data, and output the optimized vehicle driving simulation model, real-time electro-hydraulic brake valve adjustment simulation model, and driving mode recognition model.

[0044] Beneficial effects of the present invention:

[0045] The present invention can dynamically adjust the control strategy to adapt to different driving modes and road conditions by acquiring and analyzing various information such as vehicle driving status historical data, driving mode historical data, and electro-hydraulic brake valve adjustment historical data in real time, and provide stable power drive for the vehicle. The present invention can train the vehicle driving status historical data and driving mode historical data using a deep learning model, and can accurately identify the current driving mode in real time, ensure that the control logic and parameter settings match the current driving mode, and improve driving comfort and safety.

[0046] According to the real-time driving mode recognition results, the corresponding electro-hydraulic brake valve adjustment simulation model is matched in the preset electro-hydraulic brake valve adjustment model database, and the electro-hydraulic brake valve adjustment simulation data is calculated in real time, including the electro-hydraulic brake valve adjustment time, electro-hydraulic brake valve adjustment speed and electro-hydraulic brake valve adjustment smoothness data, which improves the accuracy and smoothness of the electro-hydraulic brake valve adjustment. By performing feature extraction and cluster analysis on real-time vehicle driving data, matching the most appropriate algorithm model for training, a vehicle driving simulation model is obtained, which can simulate the dynamic characteristics and response behavior of the vehicle under different driving conditions, and provide reliable data support for control optimization.

[0047] The vehicle's external image data is obtained during the vehicle's driving process, and is identified through the natural environment image recognition model. It is then compared with the real-time electro-hydraulic brake valve adjustment simulation data and vehicle driving simulation data to verify the accuracy of the model's prediction results. If the model's prediction results are inconsistent with the actual data, the model is optimized through a genetic algorithm to improve the accuracy and stability of the control. Through precise electro-hydraulic brake valve control, the electro-hydraulic brake valve adjustment process is optimized, unnecessary power loss is reduced, and fuel economy is improved. At the same time, the smooth electro-hydraulic brake valve adjustment process also improves driving comfort.

[0048] The present invention adopts a modular design concept, and each unit is relatively independent and closely cooperates with each other, which is convenient for system maintenance and upgrading. At the same time, the model parameters are optimized through genetic algorithms, which improves the robustness and scalability of the system, making it able to adapt to more complex driving environments and needs.

[0049] In summary, the present invention realizes the optimized control of the vehicle electro-hydraulic brake valve adjustment process by comprehensively applying technologies such as deep learning and genetic algorithms, which significantly improves the adaptability of control, the accuracy of driving mode recognition, the precision of electro-hydraulic brake valve adjustment simulation and the reliability of vehicle driving simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0051] Figure 1 A schematic diagram of a unit module in a vehicle posture control optimization system based on an electro-hydraulic brake valve provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] 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.

[0053] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0054] The present invention provides a vehicle posture control optimization system based on an electro-hydraulic brake valve, comprising:

[0055] A data acquisition unit, used to acquire vehicle driving status historical data, driving mode historical data, electro-hydraulic brake valve adjustment historical data, electro-hydraulic brake valve status historical data, road condition historical data, and vehicle load historical data;

[0056] The data acquisition unit in the present invention is responsible for collecting and organizing various types of historical data for subsequent data processing, model training and control optimization. Specifically, the data acquisition unit mainly acquires the following types of data:

[0057] The vehicle driving status historical data includes but is not limited to real-time dynamic data such as vehicle speed, acceleration, engine speed, and vehicle position, which reflects the vehicle's driving status at different time points and is used to analyze the vehicle's operating rules and predict future driving status.

[0058] Driving mode history data records the driving modes used by the vehicle in different time periods, such as economy mode, normal mode, sports mode, etc., which helps the system understand the driver's driving preferences and driving habits in different situations, thereby optimizing the control strategy.

[0059] The electro-hydraulic brake valve adjustment history data includes the electro-hydraulic brake valve adjustment operation records of the vehicle at different time points, such as the electro-hydraulic brake valve adjustment time, the electro-hydraulic brake valve adjustment speed, the electro-hydraulic brake valve adjustment smoothness, etc.

[0060] The electro-hydraulic brake valve status history data reflects the status information of the electro-hydraulic brake valve at different time points, such as valve core position, hydraulic oil pressure, solenoid current, etc., which helps to understand the working status of the electro-hydraulic brake valve, enable its normal operation and optimize control parameters.

[0061] Road condition historical data collects information on road conditions encountered by vehicles during driving, such as road type (urban roads, highways, mountain roads, etc.), road flatness, traffic congestion, etc.

[0062] Vehicle load history data: records the vehicle load at different time points, such as the number of passengers, cargo weight, etc. Load changes will directly affect the vehicle's power demand and control effect, so it is also crucial for optimizing control strategies.

[0063] By acquiring these comprehensive historical data, the data acquisition unit provides a solid foundation for subsequent driving mode recognition, electro-hydraulic brake valve adjustment simulation, vehicle driving simulation and control optimization. It not only helps to understand the behavior of vehicles and drivers, but also can continuously improve control strategies through machine learning and optimization algorithms, thereby improving vehicle driving smoothness and fuel efficiency.

[0064] A driving mode recognition unit is used to preprocess the vehicle driving state historical data and the driving mode historical data, use the preprocessed driving mode historical data to train the deep learning model to obtain the driving mode recognition model, obtain the real-time vehicle driving state data, substitute the vehicle driving state data into the driving mode recognition model, and obtain the real-time driving mode recognition result;

[0065] The driving mode recognition unit in the present invention is responsible for real-time recognition of the vehicle's driving mode so that the control logic and parameter settings match the current driving mode. The following is a specific working process of the driving mode recognition unit:

[0066] The driving mode recognition unit first preprocesses the vehicle driving status history data and driving mode history data, including data cleaning (removing outliers and missing values) and data normalization (converting the data to the same scale for subsequent processing and analysis).

[0067] The selected deep learning model is trained using the preprocessed driving mode history data. During this process, the driving mode history data is used as input and the corresponding driving mode (such as economic mode, normal mode, and sports mode) is used as the output label. Through multiple iterations of training, the model gradually learns the mapping relationship from vehicle driving status to driving mode.

[0068] The trained driving mode recognition model is deployed to the system to prepare for real-time driving mode recognition. The system obtains the vehicle's driving status data in real time, such as vehicle speed, acceleration, engine speed, etc.

[0069] The real-time vehicle driving status data is substituted into the trained driving mode recognition model. The model calculates the current optimal driving mode based on the input data, that is, the real-time driving mode recognition result.

[0070] The driving mode recognition unit can accurately and real-time identify the vehicle's current driving mode, providing an important basis for subsequent control strategies. This driving mode recognition method based on a deep learning model has higher accuracy and adaptability than traditional methods, and can better respond to control requirements under different driving modes and road conditions.

[0071] The electro-hydraulic brake valve adjustment simulation unit is used to match the real-time driving mode recognition result in a preset electro-hydraulic brake valve adjustment model database to obtain a real-time electro-hydraulic brake valve adjustment simulation model, substitute the real-time vehicle driving state data into the real-time electro-hydraulic brake valve adjustment simulation model to obtain real-time electro-hydraulic brake valve adjustment simulation data, generate real-time electro-hydraulic brake valve adjustment simulation data of different time periods according to a preset time period, and establish corresponding timestamps for the real-time electro-hydraulic brake valve adjustment simulation data of different time periods;

[0072] In the vehicle posture control optimization system based on the electro-hydraulic brake valve, the electro-hydraulic brake valve adjustment simulation unit simulates the best electro-hydraulic brake valve adjustment strategy according to the real-time driving mode recognition results and vehicle driving status data. The following is the specific work flow of the electro-hydraulic brake valve adjustment simulation unit:

[0073] The electro-hydraulic brake valve adjustment simulation unit first receives the real-time driving mode recognition result from the driving mode recognition unit. Then, it matches the preset electro-hydraulic brake valve adjustment model database to find the electro-hydraulic brake valve adjustment simulation model corresponding to the current driving mode. The database contains electro-hydraulic brake valve adjustment models under multiple driving modes, and each model is optimized for the shifting requirements under a specific driving mode.

[0074] Real-time vehicle driving status data, such as vehicle speed, acceleration, engine speed, etc., are obtained, and these data are substituted into the matched real-time electro-hydraulic brake valve adjustment simulation model. The model calculates the optimal electro-hydraulic brake valve adjustment time, electro-hydraulic brake valve adjustment speed, and electro-hydraulic brake valve adjustment smoothness data under the current driving state based on the input data, thereby generating real-time electro-hydraulic brake valve adjustment simulation data.

[0075] In order to better simulate the vehicle's gear shifting needs in different time periods, the electro-hydraulic brake valve adjustment simulation unit will generate real-time electro-hydraulic brake valve adjustment simulation data for different time periods according to a preset time period (such as every few seconds or minutes). This can improve the system's ability to adapt to dynamic changes during vehicle driving and provide more precise control strategies.

[0076] For each period of real-time electro-hydraulic brake valve adjustment simulation data, the electro-hydraulic brake valve adjustment simulation unit will create a corresponding timestamp for it. The timestamp records the time point when the data was generated, which is convenient for subsequent data comparison and analysis. Through the timestamp, the system can track and verify the effectiveness and stability of the electro-hydraulic brake valve adjustment strategy in different time periods.

[0077] Through the operation of the electro-hydraulic brake valve adjustment simulation unit, the system can simulate the optimal electro-hydraulic brake valve adjustment strategy in real time, and make dynamic adjustments based on changes in the vehicle's driving state and different driving modes. This not only improves the smoothness of gear shifting, but also helps to improve the vehicle's fuel efficiency.

[0078] A vehicle driving simulation unit is used to obtain real-time vehicle driving data, perform feature extraction on the real-time vehicle driving data, and obtain real-time vehicle driving feature data, the real-time vehicle driving feature data includes real-time vehicle driving state data, driving mode data, electro-hydraulic brake valve adjustment data, electro-hydraulic brake valve state data, road condition data, and vehicle load data, perform cluster analysis on the real-time vehicle driving feature data, and obtain cluster analysis results of the real-time vehicle driving feature data, match the cluster analysis results of the real-time vehicle driving feature data in a preset algorithm knowledge base, and obtain an algorithm model corresponding to the real-time vehicle driving feature data, and use the real-time vehicle driving feature data to train the algorithm model to obtain a vehicle driving simulation model;

[0079] The vehicle driving simulation unit in the present invention constructs a model that can simulate the actual driving process of the vehicle by extracting features, clustering analysis, model matching and training of real-time vehicle driving data. The following is the specific workflow of the vehicle driving simulation unit:

[0080] Acquire real-time vehicle driving data. The vehicle driving simulation unit first acquires real-time vehicle driving data including vehicle driving status data (such as vehicle speed, acceleration, engine speed, etc.), driving mode data, electro-hydraulic brake valve adjustment data, electro-hydraulic brake valve status data, road condition data, and vehicle load data, etc.

[0081] Feature extraction is performed on the acquired real-time vehicle driving data to extract features that can reflect the vehicle driving characteristics, forming real-time vehicle driving feature data. The feature data provides a basis for subsequent analysis and modeling.

[0082] Cluster analysis is performed on real-time vehicle driving characteristic data, and data points with similar characteristics are classified into one category to obtain cluster analysis results of real-time vehicle driving characteristic data. Cluster analysis helps to identify different modes and scenarios during vehicle driving.

[0083] The cluster analysis results are matched with the preset algorithm knowledge base to find the algorithm model that matches the real-time vehicle driving feature data. The algorithm knowledge base contains algorithm models under various driving modes. The models have been pre-trained and optimized to simulate the vehicle driving process in different scenarios.

[0084] The matched algorithm model is further trained using real-time vehicle driving characteristic data. Through training, the model can better adapt to the current vehicle driving status and environmental conditions, improving the accuracy and reliability of the simulation.

[0085] After the above steps, the vehicle driving simulation unit finally obtains a model that can simulate the actual driving process of the vehicle. The model can predict the vehicle's driving status and gear shifting requirements based on the input real-time vehicle driving data, providing an important basis for subsequent control optimization.

[0086] Through the operation of the vehicle driving simulation unit, the system can more accurately simulate the behavioral characteristics of the vehicle under different driving conditions, provide strong support for the optimization of the control strategy, and help improve the vehicle's fuel economy, driving comfort and driving safety.

[0087] The data comparison unit is used to obtain vehicle external image data during vehicle driving, pre-process the vehicle external image data during vehicle driving to obtain pre-processed vehicle external image data during vehicle driving, identify the pre-processed vehicle external image data during vehicle driving through a preset natural environment image recognition model to obtain vehicle external image recognition results during vehicle driving, establish corresponding timestamps for the vehicle external image recognition results during vehicle driving, randomly generate data retrieval time information, and adjust the real-time electro-hydraulic brake valve simulation data during different time periods and the vehicle driving process according to the data retrieval time information. Based on the timestamp in the vehicle external image recognition result, the data corresponding to the data retrieval time information is retrieved to obtain the first data to be compared, and according to the time information of the first data to be compared, the vehicle driving history data corresponding to the time information of the first data to be compared is retrieved, and the vehicle driving history data corresponding to the time information of the first data to be compared is substituted into the vehicle driving simulation model to obtain the vehicle driving simulation data corresponding to the time information of the first data to be compared, and the vehicle driving simulation data corresponding to the time information of the first data to be compared is used as the second data to be compared, and the first data to be compared is compared with the second data to be compared to obtain the comparison result of the first data to be compared with the second data to be compared;

[0088] The data comparison unit improves the accuracy of the control strategy by comparing and verifying the real-time data with the simulation data in the vehicle posture control optimization system based on the electro-hydraulic brake valve. The following is the specific workflow of the data comparison unit:

[0089] The data comparison unit first obtains the image data of the external environment of the vehicle during driving, which includes weather conditions, lighting conditions, road type, road signs, pedestrians, vehicles and obstacles.

[0090] The acquired external image data is preprocessed, including denoising, contrast enhancement, color correction, etc., to improve the image clarity and recognition accuracy.

[0091] The preprocessed image data is recognized through a preset natural environment image recognition model, which can identify information such as weather, lighting, road type, and road elements in the image to obtain external image recognition results during vehicle driving.

[0092] Create corresponding timestamps for image recognition results to record the time point of each recognition result and improve the accuracy of subsequent data comparison.

[0093] Data retrieval time information is randomly generated, and according to the time information, data corresponding to the data retrieval time information is retrieved based on the timestamp from the real-time electro-hydraulic brake valve adjustment simulation data at different time periods and the external image recognition results during vehicle driving, to obtain the first data to be compared.

[0094] According to the time information of the first data to be compared, the corresponding vehicle driving history data is retrieved, and then the history data is substituted into the vehicle driving simulation model to simulate the vehicle driving state corresponding to the time information of the first data to be compared, and obtain the second data to be compared.

[0095] The first data to be compared (real-time data) is compared with the second data to be compared (simulated data) to analyze the differences and consistency between the two. The comparison results will be used to evaluate the effectiveness and accuracy of the current control strategy.

[0096] According to the comparison results, if there is a large difference between the two, it means that the current control strategy cannot fully adapt to the current driving environment and vehicle status. At this time, the data comparison unit will feedback this information to the control optimization unit, triggering the optimization process of the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model, and the driving mode recognition model to improve the accuracy and adaptability of the control.

[0097] Through the work of the data comparison unit, the system can evaluate the effectiveness of the control strategy in real time and make optimization adjustments when necessary, thereby improving the vehicle's optimal control performance in different driving environments and conditions.

[0098] A control optimization unit is used to analyze the real-time vehicle driving data using a vehicle driving simulation model if the comparison results of the first data to be compared are consistent with the second data to be compared, and substitute the output results into a preset electro-hydraulic brake valve drive control model to output real-time electro-hydraulic brake valve drive parameters, and use the real-time electro-hydraulic brake valve drive parameters as control parameters of the electro-hydraulic brake valve during the adjustment of the vehicle electro-hydraulic brake valve; if the comparison results of the first data to be compared are inconsistent with the second data to be compared, use the comparison results of the first data to be compared with the second data to be compared, vehicle driving status historical data, driving mode historical data, electro-hydraulic brake valve adjustment historical data, and electro-hydraulic brake valve status historical data The vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model and the driving mode recognition model are optimized by genetic algorithm based on the historical data of road conditions and the historical data of vehicle load to obtain the optimized vehicle driving simulation model, the optimized real-time electro-hydraulic brake valve adjustment simulation model and the optimized driving mode recognition model. The optimized vehicle driving simulation model, the optimized real-time electro-hydraulic brake valve adjustment simulation model and the optimized driving mode recognition model are used to process the data to obtain the optimized real-time electro-hydraulic brake valve driving parameters, and the optimized real-time electro-hydraulic brake valve driving parameters are used as the control parameters of the electro-hydraulic brake valve in the optimized vehicle electro-hydraulic brake valve adjustment process.

[0099] The control optimization unit first receives the output result of the data comparison unit, that is, the comparison result of the first data to be compared (real-time data) and the second data to be compared (simulation data).

[0100] If the comparison results show that the two are consistent or the difference is within an acceptable range, it means that the current vehicle driving simulation model, real-time electro-hydraulic brake valve adjustment simulation model and driving mode recognition model can accurately reflect the actual driving state of the vehicle and the control strategy is effective.

[0101] If the comparison results show that there are large differences between the two, it means that the current model cannot fully adapt to the current driving environment and vehicle status and needs to be optimized and adjusted.

[0102] When the comparison results are consistent, the control optimization unit will use the vehicle driving simulation model to analyze the real-time vehicle driving data and substitute the output results into the preset electro-hydraulic brake valve drive control model.

[0103] The electro-hydraulic brake valve drive control model is established based on physical principles and the working principles of the electro-hydraulic brake system. It can calculate the real-time electro-hydraulic brake valve drive parameters, including brake pressure setting value and brake response time, according to the input vehicle driving data.

[0104] The real-time electro-hydraulic brake valve driving parameters will be directly used as the control parameters of the electro-hydraulic brake valve during the vehicle's electro-hydraulic brake valve adjustment process, improving the smoothness of the electro-hydraulic brake valve adjustment process and the stability of power output.

[0105] In the case of inconsistent comparison results, the control optimization unit will start the model optimization process. It uses the comparison results of the first data to be compared with the second data to be compared, as well as the vehicle driving state history data, driving mode history data, electro-hydraulic brake valve adjustment history data, electro-hydraulic brake valve state history data, road condition history data and vehicle load history data as inputs, and optimizes the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model and the driving mode recognition model through the genetic algorithm.

[0106] Genetic algorithm is an optimization search algorithm based on the principles of natural selection and genetics. It searches for the optimal solution by simulating the natural evolution process. In the model optimization process, the control optimization unit first initializes the population and randomly generates a set of initial solutions as the initial population of the genetic algorithm. Each solution represents a set of parameters of the model.

[0107] Then, it defines a fitness function to evaluate the quality of each solution. The fitness function is constructed based on factors such as the error between the model prediction result and the actual data, the smoothness index of the electro-hydraulic brake valve adjustment, and the accuracy of driving mode recognition.

[0108] Next, the optimization unit is controlled to perform a selection operation, retaining individuals with higher fitness as candidate solutions for the next generation.

[0109] Perform a crossover operation to generate new solutions by combining the good genes of different individuals.

[0110] Perform mutation operations to randomly introduce new genes to increase the diversity of the population.

[0111] The selection, crossover, and mutation operations are repeated until a preset number of iterations or fitness threshold is reached.

[0112] Finally, the control optimization unit outputs the optimized vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model and the driving mode recognition model.

[0113] After obtaining the optimized models, the control optimization unit uses these models to analyze and process the real-time vehicle driving data. The output results are substituted into the electro-hydraulic brake valve drive control model to calculate the optimized real-time electro-hydraulic brake valve drive parameters.

[0114] The optimized driving parameters will be used as the control parameters of the electro-hydraulic brake valve during the adjustment process of the vehicle electro-hydraulic brake valve to further improve the accuracy and adaptability of the control.

[0115] Through the work of the control optimization unit, the system can evaluate and adjust the control strategy in real time to improve the vehicle's optimal control performance under different driving environments and conditions.

[0116] Specifically, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the driving mode recognition unit is further used for:

[0117] Obtain historical data on vehicle driving status and driving mode, clean the historical data on vehicle driving status and driving mode, remove outliers and missing values, normalize the historical data on vehicle driving status and driving mode, use the preprocessed driving mode historical data as input and the corresponding driving mode as output label, train the selected deep learning model to obtain a trained driving mode recognition model, and deploy the trained driving mode recognition model to the system for real-time recognition of driving modes.

[0118] The driving mode recognition unit obtains the vehicle driving status historical data and driving mode historical data from the system's data storage or other data sources, including vehicle driving status parameters such as vehicle speed, acceleration, engine speed, and driving mode information such as economic mode, normal mode, and sports mode.

[0119] In order to improve the accuracy and reliability of the data, the driving mode recognition unit will clean the acquired data. The main purpose of this step is to remove outliers and missing values. Outliers are caused by sensor failure or data recording errors, while missing values ​​are caused by the failure of some sensors to work properly at certain times or data transmission interruption.

[0120] Since the dimensions and magnitudes of different parameters vary greatly, in order to improve the training efficiency and recognition accuracy of the model, the driving mode recognition unit will normalize the cleaned data. Normalization is to scale all data to the same scale so that each parameter has the same weight when training the model.

[0121] The driving mode recognition unit selects a model suitable for deep learning, such as convolutional neural network (CNN), recurrent neural network (RNN) or its variants (such as LSTM, GRU), etc., for driving mode recognition.

[0122] The selected deep learning model is trained using the preprocessed driving mode history data as input and the corresponding driving mode as output label. During the training process, the model will continuously learn the mapping relationship between input data and output labels, and minimize the prediction error by adjusting the model parameters.

[0123] When the model training reaches the preset convergence conditions or performance requirements, the driving mode recognition unit will deploy the trained model to the system, so that the system can use the model to process the real-time vehicle driving status data during real-time operation to identify the current driving mode. Real-time driving mode recognition, when the system is running, the driving mode recognition unit will continuously obtain real-time vehicle driving status data and input it into the deployed driving mode recognition model. The model will quickly calculate the current driving mode recognition result based on the input data, and output the result to other modules of the system (such as the electro-hydraulic brake valve adjustment simulation unit) for subsequent processing.

[0124] Through the above process, the driving mode recognition unit can accurately identify the current driving mode of the vehicle and provide important reference information for the system's control strategy, thereby improving the vehicle's optimal control performance in different driving modes.

[0125] Specifically, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the electro-hydraulic brake valve adjustment simulation unit is also used for:

[0126] Determine a preset time period, the preset time period includes a fixed time interval, and match a corresponding electro-hydraulic brake valve adjustment simulation model in a preset electro-hydraulic brake valve adjustment model database according to the real-time driving mode recognition result, the preset electro-hydraulic brake valve adjustment model database includes electro-hydraulic brake valve adjustment models corresponding to multiple driving modes;

[0127] Substituting the real-time acquired vehicle driving state data into the matched electro-hydraulic brake valve adjustment simulation model, and obtaining real-time electro-hydraulic brake valve adjustment simulation data through model calculation, the real-time electro-hydraulic brake valve adjustment simulation data including electro-hydraulic brake valve adjustment time, electro-hydraulic brake valve adjustment speed and electro-hydraulic brake valve adjustment smoothness data;

[0128] According to a preset time period, a corresponding time stamp is established for the generated real-time electro-hydraulic brake valve adjustment simulation data, and the time stamp is used to reflect the time point when the data is generated.

[0129] The electro-hydraulic brake valve adjustment simulation unit will first determine a preset time period, which includes a fixed time interval. This time period is set to simulate and predict the vehicle's shifting needs in different time periods to improve the continuity and stability of control.

[0130] According to the real-time driving mode recognition results, the electro-hydraulic brake valve adjustment simulation unit will match the corresponding electro-hydraulic brake valve adjustment simulation model in the preset electro-hydraulic brake valve adjustment model database. The preset electro-hydraulic brake valve adjustment model database is a model set containing electro-hydraulic brake valve adjustment strategies corresponding to various driving modes (such as economic mode, normal mode, sports mode, etc.). The matching process improves the system's ability to adopt the most appropriate electro-hydraulic brake valve adjustment strategy according to the characteristics of different driving modes to meet the vehicle's driving needs in different modes.

[0131] The electro-hydraulic brake valve adjustment simulation unit will substitute the real-time vehicle driving status data (such as vehicle speed, acceleration, engine speed, etc.) into the matched electro-hydraulic brake valve adjustment simulation model. Through the calculation of the model, the unit can generate real-time electro-hydraulic brake valve adjustment simulation data, including important indicators such as electro-hydraulic brake valve adjustment time, electro-hydraulic brake valve adjustment speed, and electro-hydraulic brake valve adjustment smoothness, providing a direct reference for subsequent control.

[0132] According to the preset time period, the electro-hydraulic brake valve adjustment simulation unit will establish a corresponding timestamp for the generated real-time electro-hydraulic brake valve adjustment simulation data. The timestamp is an identifier that reflects the time point when the data is generated, which improves the time sequence and traceability of the data. Through the timestamp, the system can clearly know when each set of electro-hydraulic brake valve adjustment simulation data was generated, which is of great significance for subsequent data analysis and model optimization.

[0133] In summary, the electro-hydraulic brake valve adjustment simulation unit provides prediction and simulation support for vehicle control by matching the appropriate electro-hydraulic brake valve adjustment simulation model, calculating real-time electro-hydraulic brake valve adjustment simulation data and establishing timestamps, which not only improves the accuracy and efficiency of control, but also enhances the adaptability and stability of the vehicle under different driving modes.

[0134] Specifically, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the vehicle driving simulation unit is also used for:

[0135] The cluster analysis results are matched with the preset algorithm knowledge base to obtain an algorithm model that matches the real-time vehicle driving feature data. The matching process uses similarity calculation to match the most appropriate algorithm model;

[0136] The matched algorithm model is trained using real-time vehicle driving characteristic data, and the algorithm model is trained to simulate the dynamic characteristics and response behaviors of the vehicle during driving.

[0137] The vehicle driving simulation unit first performs cluster analysis on the vehicle driving characteristic data obtained in real time. Cluster analysis is an unsupervised learning method that can classify similar data points into one category, thereby revealing the inherent structure and pattern of the data.

[0138] The cluster analysis results reflect the characteristic distribution and pattern of the vehicle under the current driving state. Next, the unit will match the cluster analysis results with the preset algorithm knowledge base. The algorithm knowledge base is a collection of algorithm models under various driving modes, which can simulate the dynamic characteristics and response behaviors of the vehicle under different driving conditions.

[0139] The matching process uses similarity calculation to find the algorithm model that matches the real-time vehicle driving feature data by comparing the similarity of the cluster analysis results with the algorithm models in the algorithm knowledge base. There are many methods for similarity calculation, such as Euclidean distance, cosine similarity, etc. The specific choice depends on the characteristics of the data and the matching requirements.

[0140] When a matching algorithm model is found, the vehicle driving simulation unit will use real-time vehicle driving characteristic data to train the model. The training process is a supervised learning or reinforcement learning process, which aims to enable the algorithm model to better simulate the dynamic characteristics and response behavior of the vehicle under the current driving state.

[0141] Through training, the algorithm model can learn the dynamic characteristics of the vehicle such as acceleration, deceleration, steering response under different driving conditions, as well as the vehicle's response behavior to driver input (such as throttle, brake, and steering wheel operation). Knowledge and capabilities enable the algorithm model to play an important role in subsequent vehicle driving simulations.

[0142] The trained algorithm model can be used for vehicle driving simulation. Vehicle driving simulation is a process based on model prediction, which uses the algorithm model to predict and simulate the driving state of the vehicle in the future. Through vehicle driving simulation, the system can evaluate the impact of different control strategies on vehicle driving performance, optimize control parameters, and improve the vehicle's fuel economy and driving comfort. At the same time, vehicle driving simulation can also be used in scenarios such as driver training and vehicle performance testing, providing strong support for vehicle research and development and improvement.

[0143] In summary, the vehicle driving simulation unit provides important data support and decision-making basis for the vehicle posture control optimization system based on the electro-hydraulic brake valve through cluster analysis and algorithm model matching, algorithm model training and vehicle driving simulation, so that the system can better adapt to the vehicle dynamic characteristics and response behavior under different driving conditions, and improve the accuracy and efficiency of control.

[0144] Specifically, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the data comparison unit is also used for:

[0145] The vehicle's external image data during driving includes weather conditions, lighting conditions, and road types;

[0146] Using a preset annotation tool to annotate the vehicle's external image data during the vehicle's driving process, to obtain annotation information of the vehicle's external image data during the vehicle's driving process, wherein the annotation information of the vehicle's external image data during the vehicle's driving process includes road signs, pedestrians, vehicles, and obstacles;

[0147] Using a deep learning model as a training model, the labeled vehicle external image data during driving is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used to evaluate model performance and adjust hyperparameters during training, and the test set is used to finally evaluate model performance.

[0148] Receive and set model training hyperparameters, including learning rate, batch size, and number of training rounds;

[0149] The training set is input into the model for training. During the training process, the model parameters are adjusted through the back propagation algorithm to minimize the loss function value. The model performance is monitored using the validation set. If the validation passes, the trained natural environment image recognition model is exported into a deployable format and deployed.

[0150] Specifically, the data comparison unit in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention has the following detailed functions and steps in processing and utilizing the vehicle external image data:

[0151] The data comparison unit first collects external image data of the vehicle during driving. The data contains rich environmental information, such as weather conditions (sunny, rainy, snowy, etc.), lighting conditions (day, night, dawn, dusk, etc.) and road types (highways, urban roads, country roads, etc.).

[0152] Subsequently, the collected vehicle external image data is annotated using the preset annotation tools. The annotation information includes but is not limited to road signs (such as traffic lights, road signs, lane lines, etc.), pedestrians, other vehicles, and obstacles, etc. The annotation information provides precise targets and labels for subsequent model training.

[0153] After the annotation is completed, the vehicle external image data is divided into training set, validation set and test set. The training set is used to train the deep learning model so that it can learn and recognize different environmental features and objects; the validation set is used to evaluate the performance of the model during the training process and adjust the model's hyperparameters (such as learning rate, batch size, number of training rounds, etc.) based on the evaluation results; the test set is used to finally evaluate the model's generalization ability and actual application effect.

[0154] Receiving and setting the hyperparameters for model training is a key step to improve the efficiency and effectiveness of model training. The selection and adjustment of hyperparameters requires repeated trials and optimization based on specific data sets and task requirements.

[0155] The training set is input into the deep learning model for training. During the training process, the model parameters are continuously adjusted through the back propagation algorithm to minimize the loss function value, thereby improving the recognition accuracy and generalization ability of the model.

[0156] Use the validation set to monitor the performance of the model during the training process. If the validation results show that the model performance meets the expected standards, the trained natural environment image recognition model can be exported into a deployable format and deployed to actual vehicles or related systems for use.

[0157] The deployed natural environment image recognition model can process and analyze external image data during vehicle driving in real time, providing important environmental perception and decision-making support for the vehicle shift optimization system.

[0158] In summary, the data comparison unit provides a powerful environmental perception capability for the vehicle posture control optimization system based on the electro-hydraulic brake valve by collecting, preprocessing, labeling, dividing, training, verifying and deploying vehicle external image data, which helps the system to more accurately understand the changes in the external environment during vehicle driving, thereby making more reasonable gear shifting decisions and improving the vehicle's driving performance and safety.

[0159] Specifically, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the control optimization unit is further used for:

[0160] The real-time vehicle driving data, electro-hydraulic brake valve status data, driving mode recognition results and sensor data are used as input parameters.

[0161] Real-time vehicle driving data includes vehicle speed, acceleration and engine speed, and sensor data includes brake pressure and throttle opening;

[0162] Using real-time electro-hydraulic brake valve driving parameters as output parameters, the real-time electro-hydraulic brake valve driving parameters including brake pressure setting value and brake response time;

[0163] Based on the physical principles and the working principle of the electro-hydraulic brake system, a mathematical model for the drive control of the electro-hydraulic brake valve is preliminarily established. The physical principles and the working principle of the electro-hydraulic brake system include the relationship between the brake pressure and the current, and the relationship between the brake response time and the oil pressure change rate.

[0164] The model parameters are optimized using the historical data of vehicle driving status and the historical data of electro-hydraulic brake valve adjustment. In the simulation environment, the model is verified using historical data to check whether the model output meets expectations, and the electro-hydraulic brake valve drive control model is obtained.

[0165] Input parameter collection,The control optimization unit first collects real-time vehicle driving data,,which includes key parameters such as vehicle speed, acceleration,,engine speed, etc. The data reflects the current driving status of the,vehicle.

[0166] At the same time, the unit also obtains electro-hydraulic brake valve status data, driving mode recognition results, and sensor data, including brake pressure and throttle opening, which are crucial for understanding the driver's intentions and the vehicle's power requirements.

[0167] Output parameter definition,The output parameters of the control optimization unit are real-time electro-hydraulic brake valve drive parameters, including brake pressure setting value and brake response time. The parameters directly determine the control effect of the electro-hydraulic brake valve during the electro-hydraulic brake valve adjustment process, thereby affecting the vehicle's driving smoothness and fuel efficiency.

[0168] Based on the physical principles and the working principles of the electro-hydraulic brake system, the control optimization unit preliminarily established a mathematical model for the drive control of the electro-hydraulic brake valve. This model takes into account key factors such as the relationship between brake pressure and current, and the relationship between brake response time and oil pressure change rate.

[0169] Subsequently, the model parameters are optimized using the historical data of vehicle driving status and the historical data of electro-hydraulic brake valve adjustment. This step continuously adjusts the model parameters to make the model output more in line with the electro-hydraulic brake valve control requirements during actual vehicle driving.

[0170] In the simulation environment, the optimized electro-hydraulic brake valve drive control model is verified using historical data. The accuracy and reliability of the model are checked by comparing the model output with the actual data.

[0171] If the model is verified, that is, the output of the model meets the expected requirements, then the model will be applied to the actual vehicle shift optimization system. In actual application, the control optimization unit will calculate the optimal electro-hydraulic brake valve drive parameters based on the real-time input parameters through the model, and send them as control instructions to the electro-hydraulic brake valve actuator.

[0172] In summary, the control optimization unit achieves precise control of the electro-hydraulic brake valve by collecting real-time data, establishing and optimizing mathematical models, and verifying and applying models. This not only improves the smoothness of vehicle shifting and fuel efficiency, but also enhances the vehicle's adaptability in different driving modes and road conditions.

[0173] Specifically, in the vehicle posture control optimization system based on the electro-hydraulic brake valve of the present invention, the control optimization unit is further used for:

[0174] Initialize the population and randomly generate a set of initial solutions as the initial population of the genetic algorithm. The solutions are the model parameters.

[0175] Receive the optimization goal, and define a fitness function to evaluate the quality of each solution based on the optimization goal. The fitness function is constructed based on the error between the model prediction result and the actual data, the smoothness index of the electro-hydraulic brake valve adjustment, and the driving mode recognition accuracy;

[0176] Perform selection operations to retain individuals with higher fitness;

[0177] Perform crossover operations to combine the good genes of different individuals;

[0178] Perform mutation operations, introduce new genes, and increase the population size;

[0179] Repeat the selection, crossover and mutation operations until the preset number of iterations or fitness threshold is reached;

[0180] The genetic algorithm optimization framework is used to optimize the parameters of the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model, and the driving mode recognition model to obtain the optimized model parameters;

[0181] Use the optimized model parameters to update the model, verify the optimized model in a simulation environment or using actual vehicle data, and output the optimized vehicle driving simulation model, real-time electro-hydraulic brake valve adjustment simulation model, and driving mode recognition model.

[0182] Initialize the population. The control optimization unit will first initialize a population and randomly generate a set of initial solutions as individuals in the population. The solutions represent the parameters of the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model, and the driving mode recognition model.

[0183] Define the fitness function. The control optimization unit receives the optimization objectives and defines the fitness function based on these objectives. The fitness function is used to evaluate the quality of each solution, that is, the quality of the model parameters. The function is usually constructed based on multiple factors such as the error between the model prediction result and the actual data, the smoothness index of the electro-hydraulic brake valve adjustment, and the accuracy of driving mode recognition.

[0184] Selection operation,In each generation, the control optimization unit performs a selection,operation to retain individuals with higher fitness, which,have better model parameters and can more accurately reflect the,actual conditions during vehicle driving and electro-hydraulic brake valve,adjustment.

[0185] Crossover operation,Then, the control optimization unit will perform a crossover operation to generate new individuals by combining the good genes of different individuals.,This step helps to explore new areas in the parameter space and find better solutions.

[0186] Mutation operation,In addition to the crossover operation, the control optimization unit will also perform a mutation operation, randomly introducing new genes into individuals to increase the diversity of the population, which helps to avoid the algorithm from falling into the local optimal solution and improve the global search capability.

[0187] Iterative optimization, the above selection, crossover and mutation operations are repeated until the preset number of iterations or fitness threshold is reached. In the process, the population will gradually converge to a better solution, and the model parameters will be continuously optimized.

[0188] Finally, the control optimization unit uses the optimized model parameters to update the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model, and the driving mode recognition model. Then, the optimized model is verified in a simulation environment or using actual vehicle data to improve its accuracy and reliability. After verification, the optimized model is output for system use. Through this series of steps, the control optimization unit can effectively use genetic algorithms to optimize model parameters and improve the performance and adaptability of the vehicle shift optimization system.

[0189] The technical solution of the present invention solves the problem that the vehicle posture control system of the existing electro-hydraulic brake valve cannot adapt to the constantly changing vehicle dynamic adjustment, resulting in the inability to provide stable power drive for the vehicle by the following methods:

[0190] Acquire comprehensive historical data including vehicle driving status historical data, driving mode historical data, electro-hydraulic brake valve adjustment historical data, electro-hydraulic brake valve status historical data, road condition historical data and vehicle load historical data, providing a data basis for subsequent analysis and model training.

[0191] Preprocess the historical data of vehicle driving status and driving mode, train the deep learning model to obtain the driving mode recognition model, identify the current driving mode in real time, and improve the control logic and parameter settings to match the current driving mode.

[0192] According to the real-time driving mode recognition results, the corresponding electro-hydraulic brake valve adjustment simulation model is matched in the preset electro-hydraulic brake valve adjustment model database, and the electro-hydraulic brake valve adjustment simulation data, including the electro-hydraulic brake valve adjustment time, the electro-hydraulic brake valve adjustment speed and the electro-hydraulic brake valve adjustment smoothness data, are calculated in real time to adapt to the ever-changing driving modes.

[0193] Perform feature extraction and cluster analysis on real-time vehicle driving data, match the most appropriate algorithm model for training, and obtain a vehicle driving simulation model to simulate the dynamic characteristics and response behavior of the vehicle under different driving conditions.

[0194] The external image data of the vehicle is obtained during the driving process, and the data is identified through the natural environment image recognition model. The data is compared with the real-time electro-hydraulic brake valve adjustment simulation data and the vehicle driving simulation data to verify the accuracy of the model prediction results.

[0195] If the model prediction results are consistent with the actual data, the vehicle driving simulation model is used to analyze the real-time vehicle driving data and output the real-time electro-hydraulic brake valve drive parameters; if they are inconsistent, the model is optimized through a genetic algorithm to obtain the optimized control parameters to improve the accuracy and stability of the control.

[0196] Genetic algorithms are used to optimize the parameters of the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model, and the driving mode recognition model. Through continuous iteration and evolution, the optimal combination of model parameters is found to improve the adaptability and accuracy of the model.

[0197] Through these measures, the technical solution of the present invention can adapt to the dynamic changes of the vehicle during driving in real time, provide stable power drive for the vehicle, and solve the problems existing in the existing electro-hydraulic brake valve vehicle control system.

Claims

1. A vehicle posture control optimization system based on an electro-hydraulic brake valve, characterized in that: include: A data acquisition unit, used to acquire vehicle driving status historical data, driving mode historical data, electro-hydraulic brake valve adjustment historical data, electro-hydraulic brake valve status historical data, road condition historical data, and vehicle load historical data; A driving mode recognition unit is used to preprocess the vehicle driving state historical data and the driving mode historical data, use the preprocessed driving mode historical data to train the deep learning model to obtain the driving mode recognition model, obtain the real-time vehicle driving state data, substitute the vehicle driving state data into the driving mode recognition model, and obtain the real-time driving mode recognition result; The electro-hydraulic brake valve adjustment simulation unit is used to match the real-time driving mode recognition result in a preset electro-hydraulic brake valve adjustment model database to obtain a real-time electro-hydraulic brake valve adjustment simulation model, substitute the real-time vehicle driving state data into the real-time electro-hydraulic brake valve adjustment simulation model to obtain real-time electro-hydraulic brake valve adjustment simulation data, generate real-time electro-hydraulic brake valve adjustment simulation data of different time periods according to a preset time period, and establish corresponding timestamps for the real-time electro-hydraulic brake valve adjustment simulation data of different time periods; A vehicle driving simulation unit is used to obtain real-time vehicle driving data, perform feature extraction on the real-time vehicle driving data, and obtain real-time vehicle driving feature data, the real-time vehicle driving feature data includes real-time vehicle driving state data, driving mode data, electro-hydraulic brake valve adjustment data, electro-hydraulic brake valve state data, road condition data, and vehicle load data, perform cluster analysis on the real-time vehicle driving feature data, and obtain cluster analysis results of the real-time vehicle driving feature data, match the cluster analysis results of the real-time vehicle driving feature data in a preset algorithm knowledge base, and obtain an algorithm model corresponding to the real-time vehicle driving feature data, and use the real-time vehicle driving feature data to train the algorithm model to obtain a vehicle driving simulation model; The data comparison unit is used to obtain vehicle external image data during vehicle driving, pre-process the vehicle external image data during vehicle driving to obtain pre-processed vehicle external image data during vehicle driving, identify the pre-processed vehicle external image data during vehicle driving through a preset natural environment image recognition model to obtain vehicle external image recognition results during vehicle driving, establish corresponding timestamps for the vehicle external image recognition results during vehicle driving, randomly generate data retrieval time information, and adjust the real-time electro-hydraulic brake valve simulation data during different time periods and the vehicle driving process according to the data retrieval time information. Based on the timestamp in the vehicle external image recognition result, the data corresponding to the data retrieval time information is retrieved to obtain the first data to be compared, and according to the time information of the first data to be compared, the vehicle driving history data corresponding to the time information of the first data to be compared is retrieved, and the vehicle driving history data corresponding to the time information of the first data to be compared is substituted into the vehicle driving simulation model to obtain the vehicle driving simulation data corresponding to the time information of the first data to be compared, and the vehicle driving simulation data corresponding to the time information of the first data to be compared is used as the second data to be compared, and the first data to be compared is compared with the second data to be compared to obtain the comparison result of the first data to be compared with the second data to be compared; A control optimization unit is used to analyze the real-time vehicle driving data using a vehicle driving simulation model if the comparison results of the first data to be compared are consistent with the second data to be compared, and substitute the output results into a preset electro-hydraulic brake valve drive control model to output real-time electro-hydraulic brake valve drive parameters, and use the real-time electro-hydraulic brake valve drive parameters as control parameters of the electro-hydraulic brake valve during the adjustment of the vehicle electro-hydraulic brake valve; if the comparison results of the first data to be compared are inconsistent with the second data to be compared, use the comparison results of the first data to be compared with the second data to be compared, vehicle driving status historical data, driving mode historical data, electro-hydraulic brake valve adjustment historical data, and electro-hydraulic brake valve status historical data The vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model and the driving mode recognition model are optimized by genetic algorithm based on the historical data of road conditions and the historical data of vehicle load to obtain the optimized vehicle driving simulation model, the optimized real-time electro-hydraulic brake valve adjustment simulation model and the optimized driving mode recognition model. The optimized vehicle driving simulation model, the optimized real-time electro-hydraulic brake valve adjustment simulation model and the optimized driving mode recognition model are used to process the data to obtain the optimized real-time electro-hydraulic brake valve driving parameters, and the optimized real-time electro-hydraulic brake valve driving parameters are used as the control parameters of the electro-hydraulic brake valve in the optimized vehicle electro-hydraulic brake valve adjustment process.

2. The vehicle posture control optimization system based on the electro-hydraulic brake valve according to claim 1, characterized in that: The driving mode recognition unit is further used for: Obtain historical data on vehicle driving status and driving mode, clean the historical data on vehicle driving status and driving mode, remove outliers and missing values, normalize the historical data on vehicle driving status and driving mode, use the preprocessed driving mode historical data as input and the corresponding driving mode as output label, train the selected deep learning model to obtain a trained driving mode recognition model, and deploy the trained driving mode recognition model to the system for real-time recognition of driving modes.

3. The vehicle posture control optimization system based on the electro-hydraulic brake valve according to claim 1, characterized in that: The electro-hydraulic brake valve adjustment simulation unit is also used for: Determine a preset time period, the preset time period includes a fixed time interval, and match a corresponding electro-hydraulic brake valve adjustment simulation model in a preset electro-hydraulic brake valve adjustment model database according to the real-time driving mode recognition result, the preset electro-hydraulic brake valve adjustment model database includes electro-hydraulic brake valve adjustment models corresponding to multiple driving modes; Substituting the real-time acquired vehicle driving state data into the matched electro-hydraulic brake valve adjustment simulation model, and obtaining real-time electro-hydraulic brake valve adjustment simulation data through model calculation, the real-time electro-hydraulic brake valve adjustment simulation data including electro-hydraulic brake valve adjustment time, electro-hydraulic brake valve adjustment speed and electro-hydraulic brake valve adjustment smoothness data; According to a preset time period, a corresponding time stamp is established for the generated real-time electro-hydraulic brake valve adjustment simulation data, and the time stamp is used to reflect the time point when the data is generated.

4. The vehicle posture control optimization system based on the electro-hydraulic brake valve according to claim 1, characterized in that: The vehicle driving simulation unit is also used for: The cluster analysis results are matched with a preset algorithm knowledge base to obtain an algorithm model that matches the real-time vehicle driving feature data. The algorithm knowledge base includes algorithm models under various driving modes. The matching process uses similarity calculation to match the algorithm model. The matched algorithm model is trained using real-time vehicle driving characteristic data, and the algorithm model is trained to simulate the dynamic characteristics and response behaviors of the vehicle during driving.

5. The vehicle posture control optimization system based on the electro-hydraulic brake valve according to claim 1, characterized in that: The data comparison unit is further used for: The vehicle's external image data during driving includes weather conditions, lighting conditions, and road types; Using a preset annotation tool to annotate the vehicle's external image data during the vehicle's driving process, to obtain annotation information of the vehicle's external image data during the vehicle's driving process, wherein the annotation information of the vehicle's external image data during the vehicle's driving process includes road signs, pedestrians, vehicles, and obstacles; Using a deep learning model as a training model, the labeled vehicle external image data during driving is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used to evaluate model performance and adjust hyperparameters during training, and the test set is used to finally evaluate model performance. Receive and set model training hyperparameters, including learning rate, batch size, and number of training rounds; The training set is input into the model for training. During the training process, the model parameters are adjusted through the back propagation algorithm to minimize the loss function value. The model performance is monitored using the validation set. If the validation passes, the trained natural environment image recognition model is exported into a deployable format and deployed.

6. The vehicle posture control optimization system based on the electro-hydraulic brake valve according to claim 1, characterized in that: The control optimization unit is further used for: The real-time vehicle driving data, electro-hydraulic brake valve status data, driving mode recognition results and sensor data are used as input parameters. Real-time vehicle driving data includes vehicle speed, acceleration and engine speed, and sensor data includes brake pressure and throttle opening; Using real-time electro-hydraulic brake valve driving parameters as output parameters, the real-time electro-hydraulic brake valve driving parameters including brake pressure setting value and brake response time; Based on the physical principles and the working principle of the electro-hydraulic brake system, a mathematical model for the drive control of the electro-hydraulic brake valve is preliminarily established. The physical principles and the working principle of the electro-hydraulic brake system include the relationship between the brake pressure and the current, and the relationship between the brake response time and the oil pressure change rate. The model parameters are optimized using the historical data of vehicle driving status and the historical data of electro-hydraulic brake valve adjustment. In the simulation environment, the model is verified using historical data to check whether the model output meets expectations, and the electro-hydraulic brake valve drive control model is obtained.

7. The vehicle posture control optimization system based on the electro-hydraulic brake valve according to claim 6, characterized in that: The control optimization unit is further used for: Initialize the population and randomly generate a set of initial solutions as the initial population of the genetic algorithm. The solutions are the model parameters. Receive the optimization goal, and define a fitness function to evaluate the quality of each solution based on the optimization goal. The fitness function is constructed based on the error between the model prediction result and the actual data, the smoothness index of the electro-hydraulic brake valve adjustment, and the driving mode recognition accuracy; Perform selection operations to retain individuals with high fitness; Perform crossover operations to combine the good genes of different individuals; Perform mutation operations, introduce new genes, and increase the population size; Repeat the selection, crossover and mutation operations until the preset number of iterations or fitness threshold is reached; The genetic algorithm optimization framework is used to optimize the parameters of the vehicle driving simulation model, the real-time electro-hydraulic brake valve adjustment simulation model, and the driving mode recognition model to obtain the optimized model parameters; Use the optimized model parameters to update the model, verify the optimized model in a simulation environment or using actual vehicle data, and output the optimized vehicle driving simulation model, real-time electro-hydraulic brake valve adjustment simulation model, and driving mode recognition model.

Images