Electro-hydraulic composite brake control method and system based on component dynamic characteristics prediction

Abstract

The invention discloses an electro-hydraulic composite braking control method and system based on component dynamic characteristic prediction, the method comprising the following steps: constructing a nonlinear MPC model prediction controller based on a vehicle dynamics model and a braking system model considering the component dynamic characteristics; solving the optimal strategy network of the nonlinear MPC model prediction controller based on a reinforcement learning algorithm; deploying the optimal strategy network to an electro-hydraulic composite braking system, calculating an electro-hydraulic braking optimal control sequence, and performing composite braking control based on the electro-hydraulic braking optimal control sequence; the invention can comprehensively consider the dynamic characteristics of hydraulic braking and electric braking systems, establish a relevant model, adopt an MPC strategy based on the model, and solve the optimal control sequence in combination with a DDPG reinforcement learning algorithm, thereby ultimately improving braking performance and safety, being able to adapt to frequently changing braking requirements, and improving solution efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle brake control. Specifically, the present invention is applied to the field of electric vehicles, and in particular to an electro-hydraulic composite brake control method and system based on component dynamic characteristic prediction. Background Art

[0002] The electro-hydraulic compound braking system is an important component of modern automobile braking technology. It realizes the decoupling and independent control of the vehicle braking process by rationally distributing the hydraulic and electric braking forces, thereby significantly improving the efficiency of energy recovery.

[0003] However, this system still faces many challenges in practical applications: due to its complex mechanical structure and physical properties, the hydraulic braking system has multiple nonlinear links, such as pipeline friction and pressure-volume (pV) characteristics. These nonlinear factors lead to obvious delays and inertia effects in the hydraulic braking response. When a braking command is issued, due to the influence of these nonlinear characteristics, the hydraulic braking system often cannot immediately achieve the expected braking effect, thus affecting the response speed and stability of the entire braking system. At the same time, the electric braking performance is also subject to a variety of conditions, including the battery SOC (state of charge), rechargeable power, and motor efficiency. The state of the brake components; changes in these factors may cause the electric braking performance to be unexpectedly weakened and unable to always maintain the best working state; when emergency braking or high-intensity braking is required, if the electric braking system cannot meet the vehicle's immediate demand for braking torque, it will seriously affect the braking effect and may even endanger driving safety.

[0004] In view of the above problems, the prior art has proposed a variety of solutions, which are as follows:

[0005] (1) For example, Chinese patent CN118387064A discloses a fuzzy-controlled electro-hydraulic composite braking torque compensation method, which attempts to solve the difference between the required total braking torque and the actual total braking torque and superimpose it on the torque compensation module for coordinated control. However, this method only considers the compensation of the braking force difference and does not fully consider the dynamic characteristics of the hydraulic brake and the motor brake themselves. Therefore, there is still a significant difference between the target braking control amount and the actual output control amount during the control process.

[0006] (2) For example, Chinese patent CN111497846B discloses an electro-hydraulic composite brake control method and system for an electric vehicle, which controls the early hydraulic intervention by setting the hydraulic intervention trigger strength to reduce the impact of hydraulic brake intervention transition conditions when the motor force is saturated; however, this rule-based algorithm design makes the control process discontinuous, especially at the boundary or transition area of ​​the set threshold, which may lead to unstable braking performance.

[0007] (3) For example, Chinese patent CN117341645A discloses a composite braking hysteresis compensation method for an extended-range electric drive axle semitrailer, which uses an MPC controller to optimize the control of the motor compensation torque and the pneumatic compensation torque during braking. However, after designing the MPC controller, the scheme does not provide a specific control method, and the mathematical methods such as state matrix quadratic programming used have large computational complexity and low efficiency, and have obvious disadvantages when dealing with multi-constrained nonlinear problems.

[0008] In summary, most of the existing electro-hydraulic composite brake system control schemes fail to comprehensively consider the dynamic characteristics of hydraulic braking and electric braking components, resulting in that in the actual control process, when the desired braking pressure is given, the braking system often cannot achieve the current control target; in addition, the existing schemes are mostly designed for brake compensation control under a single braking condition, and in the actual control process, there may be multiple conditions where the braking targets change frequently; for example, when the vehicle first brakes and then suddenly steps on the brake pedal, due to the limitation of the electric braking capacity and the time delay for the hydraulic system to build pressure, the vehicle may not be able to immediately decelerate to the expected speed, thereby affecting the braking effect and driving safety; therefore, how to comprehensively consider the dynamic characteristics of hydraulic braking and electric braking and design an electro-hydraulic composite brake system control scheme that can adapt to changes in multiple braking conditions has become a problem that needs to be solved urgently. Summary of the invention

[0009] The object of the present invention is to provide an electro-hydraulic composite brake control method and system based on component dynamic characteristic prediction, thereby solving all or one of the above problems existing in the prior art.

[0010] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0011] In one aspect, the present invention provides an electro-hydraulic composite brake control method based on component dynamic characteristic prediction, comprising the following steps:

[0012] Construct a brake system model that takes into account the dynamic characteristics of components;

[0013] Constructing a nonlinear MPC model predictive controller based on the vehicle dynamics model and the braking system model considering the dynamic characteristics of the components;

[0014] Solving the optimal policy network of the nonlinear MPC model predictive controller based on a reinforcement learning algorithm;

[0015] The optimal strategy network is deployed to the electro-hydraulic composite brake system, an electro-hydraulic brake optimal control sequence is calculated based on the optimal strategy network, and the electro-hydraulic composite brake system is called to perform composite brake control based on the electro-hydraulic brake optimal control sequence.

[0016] As an improved solution, the construction of a brake system model considering the dynamic characteristics of components includes:

[0017] Based on the relationship between braking torque and brake cylinder pressure, a dynamic characteristic model of the hydraulic braking system is constructed;

[0018] Based on the influencing factors of electric braking torque, a dynamic characteristic model of the motor braking system is established;

[0019] The hydraulic brake system dynamic characteristic model and the motor brake system dynamic characteristic model are used as the brake system model considering component dynamic characteristics.

[0020] As an improved solution, the nonlinear MPC model predictive controller is constructed based on the vehicle dynamics model and the braking system model considering the dynamic characteristics of the components, including:

[0021] Building the vehicle dynamics model based on the vehicle force parameters, vehicle attribute parameters and vehicle state variables;

[0022] Substituting the braking system model considering the dynamic characteristics of components into the vehicle dynamics model to obtain the vehicle longitudinal dynamics equation;

[0023] Converting the vehicle longitudinal dynamics equation into a continuous state space equation, discretizing the continuous state space equation to obtain a system discrete state space equation; using the system discrete state space equation as a nonlinear MPC model predictive controller;

[0024] Setting the vehicle speed as the state variable of the nonlinear MPC model predictive controller, and setting the target hydraulic brake torque and the target electric brake torque as the control variables of the nonlinear MPC model predictive controller;

[0025] Setting constraints of the nonlinear MPC model predictive controller and a target optimization function of the nonlinear MPC model predictive controller;

[0026] The optimization direction of the target optimization function is: the control target of the electro-hydraulic composite brake does not vibrate, the smoothness of the braking process is guaranteed, and the actual vehicle speed can follow the target well through the braking force control.

[0027] As an improved solution, the optimal policy network of the nonlinear MPC model predictive controller is solved based on a reinforcement learning algorithm, including:

[0028] Initializing and configuring the state, action, and reward function of the reinforcement learning algorithm;

[0029] After the initialization configuration, calculating a maximum reward function in the reinforcement learning algorithm that optimizes the objective optimization function of the nonlinear MPC model predictive controller;

[0030] Determine the policy network corresponding to the maximum reward function as the optimal policy network.

[0031] As an improved solution, the dynamic characteristic model of the hydraulic brake system is constructed based on the relationship between the brake torque and the brake cylinder pressure, including:

[0032] Determine a linear function relationship between a target brake torque and a target brake cylinder pressure;

[0033] Determine the pressure model of the brake master cylinder;

[0034] Calculating an analytical expression for transient pressure in the master brake cylinder based on the pressure model of the master brake cylinder;

[0035] Combining the linear function relationship between the target braking torque and the target brake cylinder pressure and the analytical expression of the transient pressure in the brake master cylinder, a first dynamic relationship between the actual master cylinder pressure output and the target braking torque is obtained;

[0036] determining a second relationship between actual hydraulic brake torque and actual brake master cylinder pressure;

[0037] The first dynamic relationship and the second relationship are combined to obtain a third dynamic relationship between the actual braking torque and the target braking torque of the hydraulic brake system; the third dynamic relationship is used as the dynamic characteristic model of the hydraulic brake system.

[0038] As an improved solution, the dynamic characteristic model of the motor braking system is constructed based on the influencing factors of the electric braking torque, including:

[0039] determining the efficiency characteristics of the electric brake system based on the influencing factors of the electric brake torque;

[0040] A fourth relationship between the actual electric braking torque and the target electric braking torque is determined based on the efficiency characteristics of the electric braking system; and the fourth relationship is used as a dynamic characteristic model of the motor braking system.

[0041] As an improved solution, the discrete state space equation of the system is:

[0042] ; Where m is the mass of the vehicle, g is the acceleration of gravity, and f is the rolling resistance coefficient. is the road slope angle, C D is the air resistance coefficient, A is the frontal area of ​​the vehicle, ρ is the air density, δ is the conversion coefficient of the vehicle's rotational mass, is the dynamic efficiency value, is the time constant in the system, is the slope coefficient of the braking torque and the brake master cylinder pressure, is the speed of the vehicle at time k, are the target hydraulic braking torque and target electric braking torque at time k respectively;

[0043] The constraints are: ;

[0044] The objective optimization function is: ;in, is the parameter to adjust the influence weights of the two sub-objectives, and N is the number of cycles of predictive control.

[0045] As an improved solution, the reinforcement learning algorithm includes: DDPG algorithm;

[0046] The initialization configuration of the state, action and reward function of the reinforcement learning algorithm includes:

[0047] Setting the state variable of the nonlinear MPC model predictive controller to the state of the DDPG algorithm;

[0048] Setting the control quantity of the nonlinear MPC model predictive controller to the action of the DDPG algorithm;

[0049] The inverse of the objective optimization function of the nonlinear MPC model predictive controller is set as the reward function of the DDPG algorithm.

[0050] As an improved solution, the reward function is: .

[0051] On the other hand, the present invention also provides an electro-hydraulic composite brake control system based on component dynamic characteristic prediction, comprising:

[0052] Model building module, used to: build a brake system model that takes into account the dynamic characteristics of components;

[0053] A controller building module is used to: build a nonlinear MPC model predictive controller based on a vehicle dynamics model and a braking system model considering component dynamic characteristics;

[0054] A reinforcement learning module, used to solve the optimal strategy network of the nonlinear MPC model predictive controller based on a reinforcement learning algorithm;

[0055] The control intervention module is used to: deploy the optimal strategy network to the electro-hydraulic composite brake system, calculate the electro-hydraulic brake optimal control sequence based on the optimal strategy network, and call the electro-hydraulic composite brake system to perform composite brake control based on the electro-hydraulic brake optimal control sequence.

[0056] The beneficial effects of the technical solution of the present invention are:

[0057] 1. The electro-hydraulic composite braking control method based on component dynamic characteristics prediction described in the present invention can comprehensively consider the dynamic characteristics of the hydraulic braking and electric braking systems, establish a relevant model, and adopt the MPC strategy based on this model. At the same time, the DDPG reinforcement learning algorithm is combined to solve the optimal control sequence, which ultimately improves the braking performance and safety, can adapt to frequently changing braking needs, and improves the solution efficiency and accuracy, making up for the defects of the existing technology and having a high application value.

[0058] 2. The electro-hydraulic composite brake control system based on component dynamic characteristic prediction described in the present invention can realize the electro-hydraulic composite brake control method based on component dynamic characteristic prediction described in the present invention through the mutual cooperation of system modules. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 It is a flow chart of the electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to Embodiment 1 of the present invention;

[0061] Figure 2 It is a logic flow diagram of the electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to Embodiment 1 of the present invention;

[0062] Figure 3 It is a schematic diagram of the relationship between the slope coefficient and the initial braking speed in the electro-hydraulic composite braking control method based on component dynamic characteristic prediction according to Embodiment 1 of the present invention;

[0063] Figure 4 It is a schematic diagram of the training process of step S302 in the electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to embodiment 1 of the present invention;

[0064] Figure 5It is a schematic diagram of the architecture of the electro-hydraulic composite brake control system based on component dynamic characteristic prediction described in Example 2 of the present invention. DETAILED DESCRIPTION

[0065] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0066] In the description of the present invention, it should be noted that the embodiments described in the present invention are only part of the embodiments of the present invention, rather than all of the embodiments; based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present invention.

[0067] The terms "first", "second", etc. in the specification and claims of this article and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this article described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0068] Embodiment 1: This embodiment provides an electro-hydraulic composite brake control method based on component dynamic characteristic prediction, such as Figure 1~Figure 4 As shown, the following steps are included:

[0069] S100: Build a braking system model that takes into account the dynamic characteristics of components, including:

[0070] S101. Build a dynamic characteristic model of the hydraulic brake system, as follows:

[0071] (1) According to the brake control signal, the target brake torque and the target brake cylinder pressure are calibrated to obtain the linear function relationship between the target brake torque and the target brake cylinder pressure: ①; Among them, The target control pressure output by the brake valve, is the target hydraulic braking torque, is the calibration parameter.

[0072] (2) When obtaining the target master cylinder pressure signal, the output of the actual master cylinder pressure is delayed. This dynamic process involves relatively complex physical characteristics. Therefore, the pressure model of the brake master cylinder is expressed as a first-order time delay model, and then the dynamic process is effectively approximated and fitted, which improves the accuracy of the control strategy and obtains the following expression: ②; Among them, is the transient pressure in the brake master cylinder; is the time constant in the system, which is obtained through Laplace transformation based on the brake pressure build-up response time and brake release response time under specified conditions; is the delay time in the braking circuit and is obtained by performing straight-line braking tests on the vehicle at different speeds.

[0073] (3) Solve the above first-order linear non-homogeneous differential equation (i.e., equation ②) to obtain the transient pressure in the brake master cylinder: The analytical expression of; the simultaneous equations ① and The analytical expression of is used to obtain the dynamic relationship between the actual master cylinder pressure output and the target braking torque: ③.

[0074] (4) The relationship between the actual hydraulic brake torque and the actual brake master cylinder pressure is obtained through the hydraulic brake torque system model: ④; In which, this relationship is usually a linear relationship, which can be obtained by fitting the experimental data of the brake. That is the slope coefficient of the braking torque and the brake master cylinder pressure. Affected by the initial braking speed, under a fixed master cylinder pressure, the braking torque generated by different initial braking speeds can be used to obtain the initial braking speed and slope coefficient. Functional relationship: ; In this functional relationship, For two different initial braking velocities, is the slope coefficient at different initial braking speeds; it should also be noted that the braking pressure can usually increase to 20 in a very short time. Therefore, when the brake master cylinder pressure is lower than 20 When the braking speed is different, the influence of different initial braking speeds on the braking torque is small and can be ignored.

[0075] (5) By combining equations ③ and ④, we can obtain the dynamic relationship between the actual braking torque and the target braking torque of the hydraulic braking system. This relationship is used as the dynamic characteristic model of the hydraulic braking system: .

[0076] In this step, since the brake pressure response of the hydraulic brake system is delayed and time-varying, and the physical model of the hydraulic system is relatively complex and difficult to ensure real-time performance during application, a hydraulic brake model based on the dynamic characteristics of the system is constructed according to the dynamic relationship between the target hydraulic brake torque and the actual brake torque.

[0077] S102. Build a dynamic characteristic model of the motor brake system, as follows:

[0078] (1) Considering that the electric braking torque is affected by many factors (such as the power battery SOC, charging power, and the external characteristics of the motor itself), a simplified model is established to describe the efficiency characteristics of the electric braking system, such as: ;in, It is the dynamic efficiency value, which has significant nonlinear characteristics, is difficult to explicitly model, and changes greatly with time and state; is a specific efficiency function, usually in the form of a lookup table; is the battery charge state; is the charging power.

[0079] (2) Based on the efficiency characteristic model of the electric brake system, the relationship between the actual electric brake torque and the target electric brake torque is obtained: .

[0080] In this step, the relationship between the actual electric braking torque and the target electric braking torque is used as a dynamic characteristic model of the motor braking system.

[0081] S200, constructing an electro-hydraulic composite brake nonlinear prediction model controller, including:

[0082] S201, construct the vehicle longitudinal dynamics equation, as follows:

[0083] (1) Building a vehicle dynamics model:

[0084] ;in, is the actual force of the vehicle; in driving condition, The motor drives the wheels to rotate and is provided separately; in braking conditions, It is provided by hydraulic braking and electric braking through the action of the vehicle and the ground; in this dynamic equation, the driving force is expressed as positive and the braking force is expressed as negative; m is the mass of the vehicle, g is the acceleration of gravity, f is the rolling resistance coefficient, is the road slope angle, C D is the air resistance coefficient, A is the frontal area of ​​the vehicle, ρ is the air density, and δ is the conversion coefficient of the vehicle's rotational mass. These parameters are all known parameters; is the first-order derivative of the vehicle speed v, and v is a state variable.

[0085] (2) Substituting the hydraulic brake system dynamic characteristic model and the motor brake system dynamic characteristic model constructed in step S100 for the vehicle dynamics model, the relationship between the vehicle speed and the target electro-hydraulic brake torque during the braking process is obtained as follows:

[0086] ;

[0087] in, is the wheel radius.

[0088] S202: Construct a nonlinear MPC model predictive controller, as follows:

[0089] In this embodiment, the vehicle longitudinal dynamics equation obtained in S201 is converted into a continuous state space equation, and the continuous state space equation is discretized to obtain a system discrete state space equation; the system discrete state space equation is used as a nonlinear MPC model predictive controller, as follows: ;in, is the speed of the vehicle at time k, are the target hydraulic braking torque and target electric braking torque at time k respectively; the vehicle speed is the state quantity of this controller, namely ; The target hydraulic braking torque and the target electric braking torque are the control variables of this controller, that is, .

[0090] S203, setting the constraints of the controller, as follows:

[0091] In this embodiment, in order to ensure that the electro-hydraulic compound torque output by the nonlinear MPC model predictive controller is in line with reality, the following constraints are set: ; Among them, the amplitude and change rate of hydraulic braking torque and electric braking torque are bounded parameters; The value of is determined according to the actual vehicle situation.

[0092] S204, setting the target optimization function of the controller, as follows:

[0093] In this embodiment, in order to allocate the electro-hydraulic braking force in advance according to the braking sequence in the future time period under the premise of considering the dynamic characteristics of the components, the performance optimization index and optimization direction are set as follows:

[0094] (1) The control target of the electro-hydraulic compound brake does not vibrate, ensuring the smoothness of the braking process;

[0095] (2) Through braking force control, the actual vehicle speed can follow the target well;

[0096] In this embodiment, based on the above optimization index and optimization direction, the following MPC controller optimization objective function value is minimized: ;in, To adjust the parameters of the two sub-objective influence weights, N is the number of cycles of predictive control, and the above parameters are adaptively set according to the actual vehicle situation.

[0097] S300, solving the optimal control strategy of electro-hydraulic composite brake based on reinforcement learning algorithm, including:

[0098] S301, determine the state, action and reward function of the DDPG algorithm, as follows:

[0099] (1) Set the state variable of the MPC controller to the state of the DDPG algorithm, which is specifically defined as: ;in, is the state variable in the MPC controller, namely the vehicle speed sequence.

[0100] (2) Set the control variable of the MPC controller to the action of the DDPG algorithm, which is specifically defined as: ;in, is the control variable in the MPC controller, that is, the target electro-hydraulic braking force sequence.

[0101] (3) Set the opposite of the MPC controller’s objective optimization function as the reward function of the DDPG algorithm. The specific reward function is: ; The reward function is the negative of the MPC controller objective function.

[0102] S302, using the DDPG algorithm to calculate the maximum reward function of the MPC controller, so that the objective function of the MPC controller is optimal, as follows:

[0103] (1) Initialize the four networks of DDPG (current policy network, target policy network, current value network, and target value network) and their parameters, expressed as: ;in, For the strategy network, For the value network, , are the parameters of the current policy network and the target policy network respectively, are the current value network and target value network parameters, respectively. is the next state of the interaction between the policy network and the environment.

[0104] (2) The current strategy network is used to obtain the current vehicle speed state s from the environment, calculate the deterministic action a of the target brake electro-hydraulic torque based on s, and transmit it to the environment. The corresponding calculation formula is: ;

[0105] (3) Interact with the environment for one time step and convert the interaction data generated during the braking process into Stored in the experience pool; after the experience pool has enough experience, extract a batch of data samples as training data.

[0106] (4) The target strategy network is used to pass the calculated action results of the next state (vehicle speed and target electro-hydraulic brake torque) to the target value network, which is expressed as: .

[0107] (5) Use the target value network to map the action of the next state to The action value is passed to the current value network and is expressed as: .

[0108] (6) Using the current value network to calculate the current state of action The action value is passed to the current policy network and is expressed as: .

[0109] (7) Use the current value network and the current policy network to calculate the loss function respectively, and pass the calculation results to the optimizer for training to obtain the updated parameters as follows:

[0110] ;

[0111] .

[0112] (8) The current network is used to pass the updated parameters obtained after training to the target network for soft update, which is expressed as: .

[0113] (9) Repeat the training steps (1) to (8) above using the updated parameters until the cumulative reward converges to the maximum, at which point the training is completed.

[0114] (10) After each training session, the previous training result is used to perform multiple iterative training to obtain the optimal training result. The network corresponding to the optimal training result is the trained strategy network, which can be used as a solver for solving the optimal electro-hydraulic composite brake control sequence considering dynamic characteristics.

[0115] S400, policy deployment optimization, including:

[0116] S401: deploying the trained policy network to the electro-hydraulic composite braking system of the vehicle to be applied.

[0117] S402, calling the trained strategy network to solve the optimal control sequence of the electro-hydraulic brake according to the vehicle speed sequence, and the optimal control sequence finally obtained is generally:

[0118] .

[0119] S402, the first item of the optimal control sequence and As the control value at the current moment, it is sent to the brake unit of the electro-hydraulic composite brake system for execution; at the same time, the relevant status data of the vehicle is recorded, and the relevant status data can be used to further optimize the control strategy later.

[0120] In summary, this method can ensure the timely response and effectiveness of the braking force when the braking control target changes suddenly (such as point braking or emergency braking) by accurately modeling the dynamic characteristics of hydraulic braking and electric braking, thereby significantly improving the braking performance and safety of the vehicle; furthermore, this method adopts the MPC controller, which can coordinate the control of the hydraulic braking and electric braking systems in advance according to the vehicle speed sequence in the future, effectively cope with the working conditions where the braking target changes frequently, and ensure the smoothness and continuity of the braking process; at the same time, this method transforms the MPC controller solution problem into a reinforcement learning problem by training the DDPG reinforcement learning algorithm. When faced with complex nonlinear constraints and implicit expressions, it significantly reduces the amount of calculation, improves the solution speed, and ensures the accuracy of the solution, providing strong support for the real-time control of the electro-hydraulic composite braking system.

[0121] It should be noted that the above examples are only for explaining the present invention and cannot limit the protection scope of the present invention.

[0122] Embodiment 2, based on the same inventive concept as the electro-hydraulic composite brake control method based on component dynamic characteristic prediction described in Embodiment 1, provides an electro-hydraulic composite brake control system based on component dynamic characteristic prediction, such as Figure 5 As shown, including:

[0123] Model building module, used to: build a brake system model that takes into account the dynamic characteristics of components;

[0124] A controller building module is used to: build a nonlinear MPC model predictive controller based on a vehicle dynamics model and a braking system model considering component dynamic characteristics;

[0125] A reinforcement learning module, used to solve the optimal strategy network of the nonlinear MPC model predictive controller based on a reinforcement learning algorithm;

[0126] The control intervention module is used to: deploy the optimal strategy network to the electro-hydraulic composite brake system, calculate the electro-hydraulic brake optimal control sequence based on the optimal strategy network, and call the electro-hydraulic composite brake system to perform composite brake control based on the electro-hydraulic brake optimal control sequence.

[0127] Embodiment 3: This embodiment provides a computer-readable storage medium, including:

[0128] The storage medium is used to store computer software instructions used to implement the electro-hydraulic composite braking control method based on the prediction of component dynamic characteristics described in the above-mentioned embodiment 1, which includes a program for executing the above-mentioned electro-hydraulic composite braking control method based on the prediction of component dynamic characteristics; specifically, the executable program can be built into the electro-hydraulic composite braking control system based on the prediction of component dynamic characteristics described in embodiment 2, so that the electro-hydraulic composite braking control system based on the prediction of component dynamic characteristics can implement the electro-hydraulic composite braking control method based on the prediction of component dynamic characteristics described in embodiment 1 by executing the built-in executable program.

[0129] In addition, the computer-readable storage medium of this embodiment may adopt any combination of one or more computer-readable storage media, wherein the computer-readable storage medium includes electrical, optical, electromagnetic, infrared or semiconductor systems, devices or components, or any combination thereof.

[0130] Different from the prior art, the present application adopts an electro-hydraulic composite braking control method and system based on component dynamic characteristics prediction, which can comprehensively consider the dynamic characteristics of hydraulic braking and electric braking systems, establish a relevant model, and adopt an MPC strategy based on this model. At the same time, the DDPG reinforcement learning algorithm is combined to solve the optimal control sequence, which ultimately improves the braking performance and safety, can adapt to frequently changing braking needs, and improves the solution efficiency and accuracy, making up for the defects of the prior art and having high application value.

[0131] It should be understood that in the various embodiments of this document, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0132] It should also be understood that in the embodiments of this article, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0133] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this article.

[0134] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0135] In the several embodiments provided herein, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0136] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of this article.

[0137] In addition, each functional unit in each embodiment of this invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional unit.

[0138] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this article is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of this article. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0139] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An electro-hydraulic composite brake control method based on component dynamic characteristics prediction, characterized in that: The following steps are involved: Construct a brake system model that takes into account the dynamic characteristics of components; Constructing a nonlinear MPC model predictive controller based on the vehicle dynamics model and the braking system model considering the dynamic characteristics of the components; Solving the optimal policy network of the nonlinear MPC model predictive controller based on a reinforcement learning algorithm; Deploying the optimal strategy network to an electro-hydraulic composite brake system, calculating an electro-hydraulic brake optimal control sequence based on the optimal strategy network, and calling the electro-hydraulic composite brake system to perform composite brake control based on the electro-hydraulic brake optimal control sequence; The construction of the brake system model considering the dynamic characteristics of components includes: constructing a dynamic characteristic model of a hydraulic brake system based on the relationship between the brake torque and the brake cylinder pressure; constructing a dynamic characteristic model of a motor brake system based on the influencing factors of the electric brake torque; and using the dynamic characteristic model of the hydraulic brake system and the dynamic characteristic model of the motor brake system as the brake system model considering the dynamic characteristics of components; The dynamic characteristic model of the hydraulic brake system is constructed based on the relationship between the braking torque and the brake cylinder pressure, including: determining the linear function relationship between the target braking torque and the target brake cylinder pressure; determining the pressure model of the brake master cylinder; calculating the analytical expression of the transient pressure in the brake master cylinder based on the pressure model of the brake master cylinder; combining the linear function relationship between the target braking torque and the target brake cylinder pressure and the analytical expression of the transient pressure in the brake master cylinder to obtain a first dynamic relationship between the actual master cylinder pressure output and the target braking torque; determining a second relationship between the actual hydraulic braking torque and the actual brake master cylinder pressure; combining the first dynamic relationship with the second relationship to obtain a third dynamic relationship between the actual braking torque of the hydraulic brake system and the target braking torque; and using the third dynamic relationship as the dynamic characteristic model of the hydraulic brake system.

2. The electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to claim 1, characterized in that: The nonlinear MPC model predictive controller is constructed based on the vehicle dynamics model and the braking system model considering the dynamic characteristics of components, including: Building the vehicle dynamics model based on the vehicle force parameters, vehicle attribute parameters and vehicle state variables; Substituting the braking system model considering the dynamic characteristics of components into the vehicle dynamics model to obtain the vehicle longitudinal dynamics equation; Converting the vehicle longitudinal dynamics equation into a continuous state space equation, discretizing the continuous state space equation to obtain a system discrete state space equation; using the system discrete state space equation as a nonlinear MPC model predictive controller; Setting the vehicle speed as the state variable of the nonlinear MPC model predictive controller, and setting the target hydraulic brake torque and the target electric brake torque as the control variables of the nonlinear MPC model predictive controller; Setting constraints of the nonlinear MPC model predictive controller and a target optimization function of the nonlinear MPC model predictive controller; The optimization direction of the target optimization function is: the control target of the electro-hydraulic composite brake does not vibrate, the smoothness of the braking process is guaranteed, and the actual vehicle speed can follow the target well through the braking force control.

3. The electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to claim 1, characterized in that: The method of solving the optimal strategy network of the nonlinear MPC model predictive controller based on a reinforcement learning algorithm includes: Initializing and configuring the state, action, and reward function of the reinforcement learning algorithm; After the initialization configuration, calculating a maximum reward function in the reinforcement learning algorithm that optimizes the objective optimization function of the nonlinear MPC model predictive controller; Determine the policy network corresponding to the maximum reward function as the optimal policy network.

4. The electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to claim 1, characterized in that: The dynamic characteristic model of the motor braking system is constructed based on the influencing factors of the electric braking torque, including: determining the efficiency characteristics of the electric brake system based on the influencing factors of the electric brake torque; A fourth relationship between the actual electric braking torque and the target electric braking torque is determined based on the efficiency characteristics of the electric braking system; and the fourth relationship is used as a dynamic characteristic model of the motor braking system.

5. The electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to claim 2, characterized in that: The discrete state space equation of the system is: ; Where m is the mass of the vehicle, g is the acceleration of gravity, and f is the rolling resistance coefficient. is the road slope angle, C D is the air resistance coefficient, A is the frontal area of ​​the vehicle, ρ is the air density, δ is the conversion coefficient of the vehicle's rotational mass, is the dynamic efficiency value, is the time constant in the system, is the slope coefficient of the braking torque and the brake master cylinder pressure, is the speed of the vehicle at time k, are the target hydraulic braking torque and target electric braking torque at time k respectively; The constraints are: ; The objective optimization function is: ;in, is the parameter to adjust the influence weights of the two sub-objectives, and N is the number of cycles of predictive control.

6. The electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to claim 3 is characterized in that: The reinforcement learning algorithm includes: DDPG algorithm; The initialization configuration of the state, action and reward function of the reinforcement learning algorithm includes: Setting the state variable of the nonlinear MPC model predictive controller to the state of the DDPG algorithm; Setting the control quantity of the nonlinear MPC model predictive controller to the action of the DDPG algorithm; The inverse of the objective optimization function of the nonlinear MPC model predictive controller is set as the reward function of the DDPG algorithm.

7. The electro-hydraulic composite brake control method based on component dynamic characteristic prediction according to claim 6 is characterized in that: The reward function is: .

8. An electro-hydraulic composite brake control system based on component dynamic characteristics prediction, characterized in that: include: A model building module is used to: build a braking system model that takes into account the dynamic characteristics of components; the model building module builds a hydraulic braking system dynamic characteristic model based on the relationship between braking torque and brake cylinder pressure; the model building module builds a motor braking system dynamic characteristic model based on the influencing factors of electric braking torque; the model building module uses the hydraulic braking system dynamic characteristic model and the motor braking system dynamic characteristic model as the braking system model that takes into account the dynamic characteristics of components; the model building module determines the linear function relationship between the target braking torque and the target brake cylinder pressure; the model building module determines the pressure model of the brake master cylinder; the model building module is based on the brake master cylinder The pressure model calculates the analytical expression of the transient pressure in the brake master cylinder; the model building module combines the linear function relationship between the target braking torque and the target brake cylinder pressure and the analytical expression of the transient pressure in the brake master cylinder to obtain a first dynamic relationship between the actual master cylinder pressure output and the target braking torque; the model building module determines the second relationship between the actual hydraulic braking torque and the actual brake master cylinder pressure; the model building module combines the first dynamic relationship with the second relationship to obtain a third dynamic relationship between the actual braking torque and the target braking torque of the hydraulic braking system; the model building module uses the third dynamic relationship as the dynamic characteristic model of the hydraulic braking system; A controller building module is used to: build a nonlinear MPC model predictive controller based on a vehicle dynamics model and a braking system model considering component dynamic characteristics; A reinforcement learning module, used to solve the optimal strategy network of the nonlinear MPC model predictive controller based on a reinforcement learning algorithm; The control intervention module is used to: deploy the optimal strategy network to the electro-hydraulic composite brake system, calculate the electro-hydraulic brake optimal control sequence based on the optimal strategy network, and call the electro-hydraulic composite brake system to perform composite brake control based on the electro-hydraulic brake optimal control sequence.

Images