Electro-hydraulic brake-by-wire system based on CPS architecture and energy optimization method

Abstract

The invention discloses an electro-hydraulic brake-by-wire system based on a CPS architecture and an energy optimization method. The method comprises the following steps: acquiring driver input data of a vehicle and working condition data of vehicle driving; performing data processing on the driver input data and the vehicle driving working condition data; working condition prediction is carried out according to the processed data, and a rolling time domain driving working condition prediction result is obtained; according to the obtained working condition prediction result, an electro-hydraulic composite braking overall objective function is solved, and an optimal braking strategy is obtained; and vehicle braking is conducted according to the optimal braking strategy, and braking system energy optimization is completed. Braking is conducted according to the current vehicle state in combination with working condition prediction, the energy optimization strategy is perfected, reasonable braking torque distribution is achieved, energy consumption of the electro-hydraulic composite braking system is further reduced, and energy conservation and emission reduction of the whole life cycle of the vehicle are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle brake-by-wire control, and specifically relates to an electro-hydraulic brake-by-wire control system based on a CPS architecture and an energy optimization method. Background Art

[0002] As the automotive industry develops towards intelligence and electrification, electro-hydraulic composite brake systems are widely used in modern vehicles. Electro-hydraulic composite brake systems eliminate mechanical connections and use electronic controllers, integrated sensors, and actuators. The vehicle brake pedal signal is transmitted to the electronic controller, and the control signal is transmitted to the actuator in the form of an electrical signal.

[0003] The traditional braking system is limited by its inherent mechanical hydraulic transmission mode, and has obvious shortcomings in dynamic response and energy recovery efficiency. Compared with the traditional braking system, the electro-hydraulic composite braking system improves the response speed of the braking system and also realizes the effective recovery of braking energy. During low-intensity braking, the electronic control unit controls the inverter to recover braking energy; during high-intensity braking, the hydraulic braking system intervenes to ensure braking response and safety requirements.

[0004] Due to various physical factors in the vehicle braking process, the braking decision made by the controller will be affected to a certain extent, resulting in low efficiency of braking energy recovery and poor vehicle ride quality. The wire control brake system combines the CPS (cyber-physical system) architecture, collects physical information such as driver input data, vehicle driving data, and road condition data through sensors, and uploads the collected data to the control system. The control system combines the driver input with the vehicle driving conditions to make reasonable braking decisions, improve the energy recovery efficiency of the braking system, and improve the ride quality of the vehicle.

[0005] The Chinese invention patent application number is CN202410154776.2, and the name is "A braking energy recovery system and working method of a pure electric vehicle". It proposes that the motor module is connected to the transmission and the high-frequency switch circuit. During braking, the braking torque is transmitted to the motor module through the transmission, and the motor module is connected to the high-frequency switch circuit to generate an electromotive force to achieve energy recovery. However, due to the long energy transmission path and the existence of friction, the energy recovery efficiency is low, and the motor braking effect is incomplete. The Chinese invention patent application number is CN202210721810.0, and the name is "An electro-hydraulic composite braking force distribution method for regenerative braking energy recovery of electric vehicles". It proposes to collect vehicle braking intensity and battery SOC data, and distribute the braking force of the electro-hydraulic composite braking system according to the vehicle braking intensity and battery SOC state. It only collects part of the braking system data, does not combine the CPS architecture, and collects key physical information such as vehicle driving information, road condition information and driver input. The existing braking strategy has low braking energy recovery efficiency and poor vehicle smoothness.

[0006] At present, there is no relevant public technical information on energy optimization through information-physical systems combined with driving condition prediction in the field of electro-hydraulic braking. Summary of the invention

[0007] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide an electro-hydraulic wire-controlled brake system and energy optimization method based on the CPS architecture, which realizes a reasonable braking torque distribution strategy to solve the problems of excessive braking energy consumption and inability to integrate road traffic information physics for energy prediction and optimization in the prior art. The present invention can comprehensively consider driving intention, battery status, and vehicle status to complete the energy optimization of the braking system.

[0008] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0009] An electro-hydraulic brake-by-wire system based on a CPS architecture of the present invention comprises: a brake-by-wire unit and a brake optimization management unit, which are communicatively connected;

[0010] The wire control brake unit is used to collect vehicle information and realize the braking state of the vehicle according to the braking control command sent by the braking optimization management unit, and includes: a driver input module, a braking execution module, and an environment perception module;

[0011] A driver input module is used to collect braking, driving and steering data input by the driver;

[0012] Environmental perception module, used to collect vehicle driving condition data;

[0013] A braking execution module is used to receive a braking control instruction sent by the braking optimization management unit to realize the braking of the vehicle;

[0014] The braking optimization management unit is used to predict the driving condition of the driver input data and the vehicle driving condition data sent by the wire control braking unit, generate a braking control instruction, and send the braking control instruction to the wire control braking unit.

[0015] Furthermore, the driver input module includes: a brake pedal sensor, an accelerator pedal sensor, a steering wheel angle sensor and a torque sensor.

[0016] Furthermore, the environmental perception module includes: a vehicle speed sensor, a vehicle acceleration sensor, a yaw angular velocity sensor, a road condition detection device, and a battery status sensor; the vehicle driving condition data includes: vehicle speed, vehicle lateral acceleration, vehicle longitudinal acceleration, vehicle yaw angular velocity, battery SOC status, and road curvature.

[0017] Furthermore, the braking execution module is used to implement the braking of the vehicle, and receives the braking control command sent by the braking optimization management unit through the CAN bus.

[0018] Furthermore, the brake execution module includes: a hydraulic wheel cylinder, a solenoid valve, an electric hydraulic pump, a collaborative controller, an electronic booster, a hub motor, an inverter, a power battery and a brake disc; a solenoid valve is provided at the hydraulic oil inlet of the hydraulic wheel cylinder to control the hydraulic pressure; a mechanical connection is adopted between the hydraulic wheel cylinder and the brake disc; the collaborative controller is connected to the electric hydraulic pump, the solenoid valve and the hub motor respectively through a control bus; the electronic booster is connected to the brake pedal through a wiring harness; the hub motor is connected to the wheel through a coupling to drive the wheel to rotate; when the vehicle brakes, the kinetic energy of the hub motor is converted into electrical energy through the inverter and stored in the power battery to realize regenerative braking; at the same time, the electric hydraulic pump moves according to the control instruction, and the solenoid valve realizes the pressure building of the hydraulic wheel cylinder by changing the opening, so that the brake pads and the brake disc are pressed tightly to perform friction braking.

[0019] Furthermore, the braking optimization management unit includes: a data processing module and a decision optimization module;

[0020] A data processing module, used for processing the driver input data and vehicle driving condition data sent by the wire control brake unit;

[0021] The decision optimization module is used to receive the data processed by the data processing module, calculate the working condition prediction result, and obtain the optimal braking strategy based on the working condition prediction result.

[0022] Furthermore, the data processing module processes the data by means of smoothing filtering and correlation analysis.

[0023] Furthermore, the optimal braking strategy includes: optimal braking energy recovery strategy, optimal vehicle ride comfort strategy, optimal tracking braking torque strategy, and optimal regenerative braking power strategy;

[0024] The optimal braking energy recovery strategy is to maximize the energy recovery value achieved by the hub motor through the inverter;

[0025] The optimal vehicle ride comfort strategy is to minimize the change in braking torque and the minimum vehicle bump;

[0026] The optimal tracking braking torque strategy is to have the best vehicle controllability and the shortest time to reach the target torque;

[0027] The optimal regenerative braking power strategy is that the hub motor braking torque does not exceed the maximum torque that the motor can generate and the power battery is in the best state.

[0028] Furthermore, the operating condition prediction uses an encoder decoupled prediction network (ED-TFN) based on a temporal Transformer model, which is implemented based on a self-attention mechanism. According to the driver input data and the vehicle operating condition data sent by the data processing module, the vehicle driving condition at the future moment is calculated to realize the vehicle driving condition prediction.

[0029] The present invention also provides an energy optimization method for an electro-hydraulic brake-by-wire system based on a CPS architecture. Based on the above system, the method steps are as follows:

[0030] 1) Obtaining the driver input data and the vehicle driving condition data;

[0031] 2) Processing the driver input data and vehicle driving condition data;

[0032] 3) Perform operating condition prediction based on the processed data to obtain the rolling time domain driving condition prediction result;

[0033] 4) Solve the overall objective function of the electro-hydraulic composite brake based on the obtained working condition prediction results to obtain the optimal braking strategy;

[0034] 5) Brake the vehicle according to the optimal braking strategy to optimize the braking system energy.

[0035] Furthermore, the driver input data in step 1) includes: the current steering wheel angle θ and torque T, brake pedal travel, accelerator pedal travel; the vehicle driving condition data includes vehicle speed v, vehicle lateral acceleration a x , vehicle longitudinal acceleration a y 、Vehicle yaw angular velocity ω r , battery SOC status, road curvature ρ.

[0036] Furthermore, the data processing method in step 2) is smoothing filtering and correlation analysis;

[0037] The smoothing filter process removes the noise of the data set, and the expression is as follows:

[0038]

[0039] Among them, y t-n is the driver input data and vehicle driving condition data before smoothing, y t is the driver input data set and vehicle driving condition data set obtained after smoothing, and n is the number of filter items;

[0040] The correlation analysis is a Pearson correlation coefficient analysis, which selects key parameters with high correlation with the braking torque, simplifies the calculation amount of driving condition prediction, and performs correlation analysis on the smoothed driver input data set and the vehicle driving condition data set. The expression is as follows:

[0041]

[0042] Among them, X is the driver input data set and vehicle driving condition data set obtained after smoothing, Y is the braking torque, ρ X,Y is the Pearson correlation coefficient, cov(X,Y) is the covariance of X and Y, σ X and σ Y are the standard deviations of X and Y, μ X and μ Y are the average values ​​of X and Y respectively, and E is the mathematical expectation. After correlation analysis, it is found that the correlation between vehicle speed v and braking torque is the largest, and the data set of vehicle speed v is used as the input data for working condition prediction.

[0043] Furthermore, the operating condition prediction in step 3) uses an encoder decoupled prediction network based on a temporal Transformer model, which embeds each variable into a variable token and inputs the time step as a feature of a single variable; the prediction network includes an embedding layer, a linear layer, a normalization layer, and a feedforward neural network.

[0044] Furthermore, the specific steps of performing operating condition prediction in step 3) are:

[0045] 31) The embedding layer maps the processed vehicle speed data set to obtain a mapping vector corresponding to the vehicle speed v data set;

[0046] 32) The linear layer captures the correlation between the mapping vector of the vehicle speed v data set mapped by the embedding layer and the driving condition at the future moment, and obtains the driving condition prediction feature vector;

[0047] 33) The working condition prediction feature vector obtained by the linear layer is normalized through the normalization layer;

[0048] 34) The normalized operating condition prediction feature vector is decoded through a feedforward neural network to obtain the vehicle driving condition prediction result.

[0049] Furthermore, the step 31) is implemented based on a multi-layer perceptron MLP, and feature mapping is performed on the entire historical observation sequence of each variable, and the expression is as follows:

[0050] H 0 =MLP(V 0 )

[0051] Among them, H 0 is the mapping vector, V 0 It is the vehicle speed data set obtained after smoothing.

[0052] Furthermore, the step 32) is implemented based on a multi-head attention mechanism, which consists of multiple self-attention modules. Each self-attention module converts the input vector into a query vector, a key vector and a value vector, and uses the Softmax function to generate a probability distribution according to the pairwise similarity scores between the vectors. The expression is:

[0053]

[0054] Among them, Attention (Q, K, V) is the comprehensive information extracted from the driver input data set and the vehicle driving condition data set, Q is the query vector, K is the key vector, and V is the value vector; the query vector represents the direction or focus of finding information, the key vector determines the information that the query vector should focus on, and the value vector contains the actual information to be output. The final output is obtained based on the weighted sum.

[0055] Furthermore, in step 33), the mean and standard deviation of all features are calculated to improve network convergence and training stability, and the process expression is:

[0056]

[0057]

[0058]

[0059] Output=γ×H norm +β

[0060] Among them, μ l is the mean, is the i-th neuron node of the input l-th group encoding module, n l is the number of neuron nodes in the lth group of coding modules, l∈{1,2,...,L} represents the lth group of coding modules, is the variance, H norm is the normalized value of the vehicle speed v mapping vector, γ and β are learnable scaling and offset parameters, respectively, H l is the input variable of the lth group of encoding modules, Output is the output of the normalization layer, and Output includes the vehicle speed v mapping vector.

[0061] Furthermore, the structure of the feedforward neural network in step 34) is a two-layer fully connected network with an activation function in the middle, and the activation function is a Gaussian error linear unit GELU, and the expression is:

[0062]

[0063] Where GELU is the Gaussian error linear unit activation function, and x is the function input variable.

[0064] Furthermore, the feedforward neural network encodes the observed time series through the stacking of Transformer modules and decodes the representation of the future sequence using dense nonlinear connections to obtain the output, which is expressed as:

[0065] H l =LayerNorm(H l-1 +Feed-Forward(H l-1 ))

[0066] Among them, H l is the output of the current layer, H l-1 is the output of the previous layer, LayerNorm is the layer normalization function, and Forward is the feedforward layer function. After receiving the input, it generates the output through a series of transformations.

[0067] Furthermore, the future sequence is decoded by inputting the output of the feedforward neural network into a multi-layer perceptron for mapping, and the expression is:

[0068]

[0069] in, is the working condition prediction result, H L is the output of the feedforward neural network.

[0070] Furthermore, the optimal braking strategy in step 4) includes: optimal braking energy recovery strategy, optimal vehicle ride comfort strategy, optimal tracking braking torque strategy, and optimal regenerative braking power strategy;

[0071] The goal of the optimal braking energy recovery strategy is to maximize the energy recovered from the vehicle deceleration process. The braking energy recovery rate objective function expression is:

[0072]

[0073] Among them, J bra_rec is the total recovered energy, T mot_ij is the motor torque, ω ij is the motor angular velocity, v i0 is the vehicle speed at the start of braking, v ik is the vehicle speed at the end of braking, η d is the braking system efficiency, η m_reg is the motor regenerative braking efficiency, f is the rolling resistance coefficient, m is the vehicle mass, CD is the air resistance coefficient, A is the frontal area of ​​the vehicle, t ik is the braking end time, t i0 is the braking start time;

[0074] The goal of the optimal vehicle ride comfort strategy is to achieve the best ride comfort during the braking process. The ride comfort objective function expression is:

[0075]

[0076] Among them, J bra_f is the target value of vehicle smoothness, u(k+i|k) represents the vehicle speed at the i-th step in the future with the vehicle speed at the current time step k as the initial value, and u(k+i-1|k) represents the vehicle speed at the i-1-th step in the future with the vehicle speed at the current time step k as the initial value;

[0077] The goal of the optimal tracking braking torque strategy is to ensure that the braking torque actually applied can accurately track and meet the expected braking torque setting value. The tracking braking torque objective function expression is:

[0078]

[0079] Among them, J bra_T is the tracking value, T if (k+i|k) represents the front axle braking torque at the current time step k as the initial value, and the front axle braking torque at the future i-th step, T jr (k+i|k) means that the rear axle braking torque at the current time step k is taken as the initial value, and the rear axle braking torque at the future i-th step, T req (k+i|k) represents the expected braking torque at the future i-th step, taking the expected braking torque at the current time step k as the initial value;

[0080] The optimal regenerative braking power strategy needs to consider the regenerative braking torque of the wheel hub motor and the battery SOC value;

[0081] The regenerative braking torque of the wheel hub motor is less than / equal to the maximum braking torque that the motor can generate, and the expression is:

[0082]

[0083] Where n is the motor speed, η m is the motor efficiency, P chg_max is the maximum charging power of the motor, T m_max is the maximum torque of the motor, T m is the regenerative braking torque;

[0084] The battery SOC value is less than 90%, and the expression is:

[0085]

[0086] Among them, SOC initial is the initial charge state of the battery, C bat is the rated capacity of the battery, and I0 is the current value.

[0087] Furthermore, in step 4), the overall objective function of the electro-hydraulic hybrid brake is obtained by multiplying the braking energy recovery rate objective function, the smoothness objective function, and the tracking braking torque objective function by their weight coefficients respectively, and then summing them up, and the expression is:

[0088] J=min(w rec ·J bra_rec +w f ·J bra_f +w T ·J bra_T )

[0089] Among them, J is the optimal braking strategy target value, w rec is the energy recovery weight coefficient, w f is the smoothness weight coefficient, w T is the braking force tracking target weight coefficient.

[0090] Furthermore, in the step 4), when solving the overall objective function of the electro-hydraulic composite brake, it is necessary to use a deep Q learning algorithm to recalculate the energy recovery rate weight coefficient, smoothness weight coefficient, and tracking braking torque weight coefficient under different braking conditions to ensure that the benefits of each braking are maximized.

[0091] Beneficial effects of the present invention:

[0092] The present invention integrates the wheel hub motor with the hydraulic brake system, cancels the original mechanical connection, simplifies the structure of the brake system, and applies braking torque to different wheels through four wheel hub motors, so as to meet the independent control of the wheels and improve active safety and operational stability; during regenerative braking, energy can be directly recovered through the inverter, which increases the energy recovery efficiency of the brake system; and the wheel hub motor can cooperate with the hydraulic control unit to brake by applying reverse torque, quickly compensate for the required braking torque, and improve braking reliability;

[0093] The present invention combines the CPS architecture with the electro-hydraulic brake-by-wire system, inputs the driving conditions collected by the brake-by-wire unit into the brake optimization management unit, performs control after decision optimization, and realizes a closed loop of "perception-analysis-decision-execution".

[0094] The Transformer network architecture in the present invention improves the traditional deep learning neural network and can adapt to the working conditions where the sensor collects signals of different frequencies. It is suitable for application in electro-hydraulic composite braking systems. Using this network for working condition prediction can improve the stability of the algorithm and ensure the correctness of the braking system control strategy.

[0095] The present invention performs braking according to the current vehicle state in combination with operating condition prediction, improves the energy optimization strategy, realizes reasonable braking torque distribution, further reduces the energy consumption of the electro-hydraulic composite braking system, and realizes energy conservation and emission reduction throughout the life cycle of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Figure 1 The figure is a functional block diagram of the system of the present invention.

[0097] Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0098] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with embodiments and drawings. The contents mentioned in the implementation modes are not intended to limit the present invention.

[0099] Reference Figure 1 As shown, an electro-hydraulic brake-by-wire system based on a CPS architecture of the present invention comprises: a brake-by-wire unit and a brake optimization management unit, which are communicatively connected;

[0100] The wire control brake unit is used to collect vehicle information and realize the braking state of the vehicle according to the braking control command sent by the braking optimization management unit, and includes: a driver input module, a braking execution module, and an environment perception module;

[0101] The driver input module is used to collect the braking, driving and steering data input by the driver, which includes: brake pedal sensor, accelerator pedal sensor, steering wheel angle sensor and torque sensor.

[0102] The environmental perception module is used to collect the vehicle's operating condition data, which includes: vehicle speed sensor, vehicle acceleration sensor, yaw angular velocity sensor, road condition detection equipment, and battery status sensor; the vehicle's operating condition data includes: vehicle speed, vehicle lateral acceleration, vehicle longitudinal acceleration, vehicle yaw angular velocity, battery SOC status, and road condition information includes road curvature.

[0103] The braking execution module is used to receive the braking control command sent by the braking optimization management unit to realize the braking of the vehicle; it receives the braking control command sent by the braking optimization management unit through the CAN bus.

[0104] The brake execution module includes: a hydraulic wheel cylinder, a solenoid valve, an electric hydraulic pump, a cooperative controller, an electronic booster, a hub motor, an inverter, a power battery and a brake disc; a solenoid valve is provided at the hydraulic oil inlet of the hydraulic wheel cylinder to control the hydraulic pressure; a mechanical connection is adopted between the hydraulic wheel cylinder and the brake disc; the cooperative controller is connected to the electric hydraulic pump, the solenoid valve and the hub motor respectively through a control bus; the electronic booster is connected to the brake pedal through a wiring harness; the hub motor is connected to the wheel through a coupling to drive the wheel to rotate; when the vehicle brakes, the kinetic energy of the hub motor is converted into electrical energy through the inverter and stored in the power battery to realize regenerative braking; at the same time, the electric hydraulic pump moves according to the control instruction, and the solenoid valve realizes the pressure building of the hydraulic wheel cylinder by changing the opening, so that the brake pad and the brake disc are pressed tightly to perform friction braking.

[0105] The braking optimization management unit is used to predict the driving condition of the driver input data and the vehicle driving condition data sent by the wire control braking unit, generate a braking control instruction, and send the braking control instruction to the wire control braking unit; it includes: a data processing module and a decision optimization module;

[0106] A data processing module, used for processing the driver input data and vehicle driving condition data sent by the wire control brake unit;

[0107] The decision optimization module is used to receive the data processed by the data processing module, calculate the working condition prediction result, and obtain the optimal braking strategy based on the working condition prediction result.

[0108] The data processing module processes the data by means of smoothing filtering and correlation analysis.

[0109] The optimal braking strategy includes: optimal braking energy recovery strategy, optimal vehicle ride comfort strategy, optimal tracking braking torque strategy, and optimal regenerative braking power strategy;

[0110] The optimal braking energy recovery strategy is to maximize the energy recovery value achieved by the hub motor through the inverter;

[0111] The optimal vehicle ride comfort strategy is to minimize the change in braking torque and the minimum vehicle bump;

[0112] The optimal tracking braking torque strategy is to have the best vehicle controllability and the shortest time to reach the target torque;

[0113] The optimal regenerative braking power strategy is that the hub motor braking torque does not exceed the maximum torque that the motor can generate and the power battery is in the best state.

[0114] Among them, the operating condition prediction uses an encoder decoupled prediction network (ED-TFN) based on a temporal Transformer model, which is implemented based on a self-attention mechanism. According to the driver input data and the vehicle operating condition data sent by the data processing module, the vehicle driving condition at the future moment is calculated to realize the vehicle driving condition prediction.

[0115] Reference Figure 2 As shown, the present invention also provides an energy optimization method for an electro-hydraulic brake-by-wire system based on a CPS architecture. Based on the above system, the method steps are as follows:

[0116] 1) Obtaining the driver input data of the vehicle and the operating condition data of the vehicle;

[0117] The driver input data in step 1) includes: the current steering wheel angle θ and torque T, brake pedal travel, accelerator pedal travel; the vehicle driving condition data includes vehicle speed v, vehicle lateral acceleration a x , vehicle longitudinal acceleration a y 、Vehicle yaw angular velocity ω r , battery SOC status, road curvature ρ.

[0118] 2) Processing the driver input data and vehicle driving condition data;

[0119] The data processing method is smoothing filtering and correlation analysis;

[0120] The smoothing filter process removes the noise of the data set, and the expression is as follows:

[0121]

[0122] Among them, y t-n is the driver input data and vehicle driving condition data before smoothing, y t is the driver input data set and vehicle driving condition data set obtained after smoothing, and n is the number of filter items;

[0123] The correlation analysis is a Pearson correlation coefficient analysis, which selects key parameters with high correlation with the braking torque, simplifies the calculation amount of driving condition prediction, and performs correlation analysis on the smoothed driver input data set and the vehicle driving condition data set. The expression is as follows:

[0124]

[0125] Among them, X is the driver input data set and vehicle driving condition data set obtained after smoothing, Y is the braking torque, ρ X,Y is the Pearson correlation coefficient, cov(X,Y) is the covariance of X and Y, σX and σ Y are the standard deviations of X and Y, μ X and μ Y are the average values ​​of X and Y respectively, and E is the mathematical expectation. After correlation analysis, it is found that the correlation between vehicle speed v and braking torque is the largest, and the data set of vehicle speed v is used as the input data for working condition prediction.

[0126] 3) Perform operating condition prediction based on the processed data to obtain the rolling time domain driving condition prediction result;

[0127] The working condition prediction uses an encoder decoupled prediction network (ED-TFN) based on the temporal Transformer model, which embeds each variable into a variable token and inputs the time step as the feature of a single variable; the prediction network includes an embedding layer, a linear layer, a normalization layer, and a feedforward neural network.

[0128] The specific steps for operating condition prediction are:

[0129] 31) The embedding layer maps the processed vehicle speed data set to obtain a mapping vector corresponding to the vehicle speed v data set;

[0130] 32) The linear layer captures the correlation between the mapping vector of the vehicle speed v data set mapped by the embedding layer and the driving condition at the future moment, and obtains the driving condition prediction feature vector;

[0131] 33) The working condition prediction feature vector obtained by the linear layer is normalized through the normalization layer;

[0132] 34) The normalized operating condition prediction feature vector is decoded through a feedforward neural network to obtain the vehicle driving condition prediction result.

[0133] Wherein, the step 31) is implemented based on a multi-layer perceptron MLP, and feature mapping is performed on the entire historical observation sequence of each variable, and the expression is as follows:

[0134] H 0 =MLP(V 0 )

[0135] Among them, H 0 is the mapping vector, V 0 is the vehicle speed data set obtained after slip processing.

[0136] Wherein, the step 32) is implemented based on a multi-head attention mechanism, which is composed of multiple self-attention modules. Each self-attention module converts the input vector into a query vector, a key vector and a value vector, and uses the Softmax function to generate a probability distribution according to the pairwise similarity scores between the vectors. The expression is:

[0137]

[0138] Among them, Attention (Q, K, V) is the comprehensive information extracted from the driver input data set and the vehicle driving condition data set, Q is the query vector, K is the key vector, and V is the value vector; the query vector represents the direction or focus of finding information, the key vector determines the information that the query vector should focus on, and the value vector contains the actual information to be output. The final output is obtained based on the weighted sum.

[0139] Wherein, in step 33), the mean and standard deviation of all features are calculated to improve the network convergence and training stability, and the process expression is:

[0140]

[0141]

[0142]

[0143] Output=γ×H norm +β

[0144] Among them, μ l is the mean, is the i-th neuron node of the input l-th group encoding module, n l is the number of neuron nodes in the lth group of coding modules, l∈{1,2,...,L} represents the lth group of coding modules, is the variance, H norm is the normalized value of the vehicle speed v mapping vector, γ and β are learnable scaling and offset parameters, respectively, H l is the input variable of the lth group of encoding modules, Output is the output of the normalization layer, and Output includes the vehicle speed v mapping vector.

[0145] The structure of the feedforward neural network in step 34) is a two-layer fully connected network with an activation function in the middle. The activation function is a Gaussian error linear unit GELU, and the expression is:

[0146]

[0147] Where GELU is the Gaussian error linear unit activation function, and x is the function input variable.

[0148] The feedforward neural network encodes the observed time series through the stacking of Transformer modules and decodes the representation of future sequences using dense nonlinear connections to obtain the output, which is expressed as:

[0149] Hl =LayerNorm(H l-1 +Feed-Forward(H l-1 ))

[0150] Among them, H l is the output of the current layer, H l-1 is the output of the previous layer, LayerNorm is the layer normalization function, and Forward is the feedforward layer function. After receiving the input, it generates the output through a series of transformations.

[0151] The future sequence is decoded by inputting the output of the feedforward neural network into a multi-layer perceptron for mapping. The expression is:

[0152]

[0153] in, is the working condition prediction result, H L is the output of the feedforward neural network.

[0154] 4) Solve the overall objective function of the electro-hydraulic composite brake based on the obtained working condition prediction results to obtain the optimal braking strategy;

[0155] The optimal braking strategy includes: optimal braking energy recovery strategy, optimal vehicle smoothness strategy, optimal tracking braking torque strategy, and optimal regenerative braking power strategy;

[0156] The goal of the optimal braking energy recovery strategy is to maximize the energy recovered from the vehicle deceleration process. The braking energy recovery rate objective function expression is:

[0157]

[0158] Among them, J bra_rec is the total recovered energy, T mot_ij is the motor torque, ω ij is the motor angular velocity, v i0 is the vehicle speed at the start of braking, v ik is the vehicle speed at the end of braking, η d is the braking system efficiency, η m_reg is the motor regenerative braking efficiency, f is the rolling resistance coefficient, m is the vehicle mass, C D is the air resistance coefficient, A is the frontal area of ​​the vehicle, t ik is the braking end time, t i0 is the braking start time;

[0159] The goal of the optimal vehicle ride comfort strategy is to achieve the best ride comfort during the braking process. The ride comfort objective function expression is:

[0160]

[0161] Among them, J bra_f is the target value of vehicle smoothness, u(k+i|k) represents the vehicle speed at the i-th step in the future with the vehicle speed at the current time step k as the initial value, and u(k+i-1|k) represents the vehicle speed at the i-1-th step in the future with the vehicle speed at the current time step k as the initial value;

[0162] The goal of the optimal tracking braking torque strategy is to ensure that the braking torque actually applied can accurately track and meet the expected braking torque setting value. The tracking braking torque objective function expression is:

[0163]

[0164] Among them, J bra_T is the tracking value, T if (k+i|k) represents the front axle braking torque at the current time step k as the initial value, and the front axle braking torque at the future i-th step, T jr (k+i|k) means that the rear axle braking torque at the current time step k is taken as the initial value, and the rear axle braking torque at the future i-th step, T req (k+i|k) represents the expected braking torque at the future i-th step, taking the expected braking torque at the current time step k as the initial value;

[0165] The optimal regenerative braking power strategy needs to consider the regenerative braking torque of the wheel hub motor and the battery SOC value;

[0166] The regenerative braking torque of the wheel hub motor is less than / equal to the maximum braking torque that the motor can generate, and the expression is:

[0167]

[0168] Where n is the motor speed, η m is the motor efficiency, P chg_max is the maximum charging power of the motor, T m_max is the maximum torque of the motor, T m is the regenerative braking torque;

[0169] The battery SOC value is less than 90%, and the expression is:

[0170]

[0171] Among them, SOC initial is the initial charge state of the battery, C bat is the rated capacity of the battery, and I0 is the current value.

[0172] The overall objective function of electro-hydraulic hybrid braking is obtained by multiplying the braking energy recovery rate objective function, the smoothness objective function, and the tracking braking torque objective function with their weight coefficients respectively, and then summing them up. The expression is:

[0173] J=min(w rec ·J bra_rec +w f ·J bra_f +w T ·J bra_T )

[0174] Among them, J is the optimal braking strategy target value, w rec is the energy recovery weight coefficient, w f is the smoothness weight coefficient, w T is the braking force tracking target weight coefficient.

[0175] Specifically, when solving the overall objective function of the electro-hydraulic composite brake in step 4), a deep Q-learning algorithm is required to recalculate the energy recovery rate weight coefficient, smoothness weight coefficient, and tracking braking torque weight coefficient under different braking conditions to ensure that the benefits of each braking are maximized.

[0176] 5) Brake the vehicle according to the optimal braking strategy to optimize the braking system energy.

[0177] The present invention has many specific application paths. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements can be made without departing from the principle of the present invention. These improvements should also be regarded as the protection scope of the present invention.

Claims

1. An electro-hydraulic brake-by-wire system based on a CPS architecture, characterized in that: include: The wire control brake unit and the brake optimization management unit are communicatively connected; The wire control brake unit is used to collect vehicle information and realize the braking state of the vehicle according to the braking control command sent by the braking optimization management unit, and includes: a driver input module, a braking execution module, and an environment perception module; A driver input module is used to collect braking, driving and steering data input by the driver; Environmental perception module, used to collect vehicle driving condition data; A braking execution module is used to receive a braking control instruction sent by the braking optimization management unit to realize the braking of the vehicle; The braking optimization management unit is used to predict the driving condition of the driver input data and the vehicle driving condition data sent by the wire control braking unit, generate a braking control instruction, and send the braking control instruction to the wire control braking unit.

2. The electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 1, characterized in that: The brake execution module includes: a hydraulic wheel cylinder, a solenoid valve, an electric hydraulic pump, a cooperative controller, an electronic booster, a hub motor, an inverter, a power battery and a brake disc; a solenoid valve is provided at the hydraulic oil inlet of the hydraulic wheel cylinder to control the hydraulic pressure; a mechanical connection is adopted between the hydraulic wheel cylinder and the brake disc; the cooperative controller is connected to the electric hydraulic pump, the solenoid valve and the hub motor respectively through a control bus; the electronic booster is connected to the brake pedal through a wiring harness; the hub motor is connected to the wheel through a coupling to drive the wheel to rotate; when the vehicle brakes, the kinetic energy of the hub motor is converted into electrical energy through the inverter and stored in the power battery to realize regenerative braking; at the same time, the electric hydraulic pump moves according to the control instruction, and the solenoid valve realizes the pressure building of the hydraulic wheel cylinder by changing the opening, so that the brake pad and the brake disc are pressed tightly to perform friction braking.

3. The electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 1, characterized in that: The braking optimization management unit includes: a data processing module and a decision optimization module; A data processing module, used for processing the driver input data and vehicle driving condition data sent by the wire control brake unit; The decision optimization module is used to receive the data processed by the data processing module, calculate the working condition prediction result, and obtain the optimal braking strategy based on the working condition prediction result.

4. An energy optimization method for an electro-hydraulic brake-by-wire system based on a CPS architecture, based on the system described in any one of claims 1 to 3, characterized in that: The steps are as follows: 1) Obtaining the driver input data of the vehicle and the operating condition data of the vehicle; 2) Processing the driver input data and vehicle driving condition data; 3) Perform operating condition prediction based on the processed data to obtain the rolling time domain driving condition prediction result; 4) Solve the overall objective function of the electro-hydraulic composite brake based on the obtained working condition prediction results to obtain the optimal braking strategy; 5) Brake the vehicle according to the optimal braking strategy to optimize the braking system energy.

5. The energy optimization method of the electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 4 is characterized in that: The driver input data in step 1) includes: current steering wheel angle and torque, brake pedal travel, accelerator pedal travel; vehicle driving condition data includes vehicle speed v, vehicle lateral acceleration a x , vehicle longitudinal acceleration a y 、Vehicle yaw angular velocity ω r , battery SOC status, road curvature ρ.

6. The energy optimization method of the electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 4 is characterized in that: The data processing method in step 2) is smoothing filtering and correlation analysis; The smoothing filter process removes the noise of the data set, and the expression is as follows: Among them, y t-n is the driver input data and vehicle driving condition data before smoothing, y t is the driver input data set and vehicle driving condition data set obtained after smoothing, and n is the number of filter items; The correlation analysis is a Pearson correlation coefficient analysis, which selects key parameters with high correlation with the braking torque, simplifies the calculation amount of driving condition prediction, and performs correlation analysis on the smoothed driver input data set and the vehicle driving condition data set. The expression is as follows: Among them, X is the driver input data set and vehicle driving condition data set obtained after smoothing, Y is the braking torque, ρ X,Y is the Pearson correlation coefficient, cov(X,Y) is the covariance of X and Y, σ X and σ Y are the standard deviations of X and Y, μ X and μ Y are the average values ​​of X and Y respectively, and E is the mathematical expectation. After correlation analysis, it is found that the correlation between vehicle speed v and braking torque is the largest, and the data set of vehicle speed v is used as the input data for working condition prediction.

7. The energy optimization method of the electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 4, characterized in that: The working condition prediction in step 3) uses an encoder decoupled prediction network based on a temporal Transformer model, which embeds each variable into a variable token and inputs the time step as a feature of a single variable; the prediction network includes an embedding layer, a linear layer, a normalization layer, and a feedforward neural network.

8. The energy optimization method of the electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 7 is characterized in that: The specific steps of performing operating condition prediction in step 3) are: 31) The embedding layer maps the processed vehicle speed data set to obtain a mapping vector corresponding to the vehicle speed v data set; 32) The linear layer captures the correlation between the mapping vector of the vehicle speed v data set mapped by the embedding layer and the driving condition at the future moment, and obtains the driving condition prediction feature vector; 33) The working condition prediction feature vector obtained by the linear layer is normalized through the normalization layer; 34) The normalized operating condition prediction feature vector is decoded through a feedforward neural network to obtain the vehicle driving condition prediction result.

9. The energy optimization method of the electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 4, characterized in that: The optimal braking strategy in step 4) includes: optimal braking energy recovery strategy, optimal vehicle ride comfort strategy, optimal tracking braking torque strategy, and optimal regenerative braking power strategy; The goal of the optimal braking energy recovery strategy is to maximize the energy recovered from the vehicle deceleration process. The braking energy recovery rate objective function expression is: Among them, J bra_rec is the total recovered energy, T mot_ij is the motor torque, ω ij is the motor angular velocity, v i0 is the vehicle speed at the start of braking, v ik is the vehicle speed at the end of braking, η d is the braking system efficiency, η m_reg is the motor regenerative braking efficiency, f is the rolling resistance coefficient, m is the vehicle mass, C D is the air resistance coefficient, A is the frontal area of ​​the vehicle, t ik is the braking end time, t i0 is the braking start time; The goal of the optimal vehicle ride comfort strategy is to achieve the best ride comfort during the braking process. The ride comfort objective function expression is: Among them, J bra_f is the target value of vehicle smoothness, u(k+i|k) represents the vehicle speed at the i-th step in the future with the vehicle speed at the current time step k as the initial value, and u(k+i-1|k) represents the vehicle speed at the i-1-th step in the future with the vehicle speed at the current time step k as the initial value; The goal of the optimal tracking braking torque strategy is to ensure that the braking torque actually applied can accurately track and meet the expected braking torque setting value. The tracking braking torque objective function expression is: Among them, J bra_T is the tracking value, T if (k+i|k) represents the front axle braking torque at the current time step k as the initial value, and the front axle braking torque at the future i-th step, T jr (k+i|k) means that the rear axle braking torque at the current time step k is taken as the initial value, and the rear axle braking torque at the future i-th step, T req (k+i|k) represents the expected braking torque at the future i-th step, taking the expected braking torque at the current time step k as the initial value; The optimal regenerative braking power strategy needs to consider the regenerative braking torque of the wheel hub motor and the battery SOC value; The regenerative braking torque of the wheel hub motor is less than / equal to the maximum braking torque that the motor can generate, and the expression is: Where n is the motor speed, η m is the motor efficiency, P chg_max is the maximum charging power of the motor, T m_max is the maximum torque of the motor, T m is the regenerative braking torque; The battery SOC value is less than 90%, and the expression is: Among them, SOC initial is the initial charge state of the battery, C bat is the rated capacity of the battery, and I0 is the current value.

10. The energy optimization method of the electro-hydraulic brake-by-wire system based on the CPS architecture according to claim 4, characterized in that: In step 4), the overall objective function of the electro-hydraulic hybrid brake is obtained by multiplying the braking energy recovery rate objective function, the smoothness objective function, and the tracking braking torque objective function by their weight coefficients respectively, and then summing them up. The expression is: J=min(in rec ·J bra_rec +in f ·J bra_f +in T ·J bra_T ) Among them, J is the optimal braking strategy target value, w rec is the energy recovery weight coefficient, w f is the smoothness weight coefficient, w T is the braking force tracking target weight coefficient.

Images