An electric vehicle brake-by-wire system wheel cylinder hydraulic pressure control method

Abstract

The application provides a wheel cylinder hydraulic pressure control method of an electric vehicle brake-by-wire system, comprising the following steps: constructing a nonlinear model of an electric vehicle electronic hydraulic brake system; designing an adaptive extended state observer based on an improved particle swarm optimization algorithm to estimate a preliminary electric vehicle wheel cylinder hydraulic pressure; designing a sliding mode attitude controller based on the adaptive extended state observer, taking the estimated preliminary electric vehicle wheel cylinder hydraulic pressure as input; improving a sliding mode surface equation through a target function and a fitness function established by an improved alpha evolution algorithm, and introducing a chaotic disturbance term, an evolution compensation term and an adaptive chattering suppression factor, so as to obtain an improved sliding mode control method, and controlling the electric vehicle wheel cylinder hydraulic pressure through the improved sliding mode control method. The application can realize high-precision hydraulic pressure control and improve the anti-interference ability.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle technology, and more specifically to a method for controlling the hydraulic pressure of wheel cylinders in a brake-by-wire system for electric vehicles. Background Technology

[0002] With the rapid development of intelligent and new energy technologies in the automotive industry, the reliability and safety of automotive braking systems, as a core component of vehicle safety, are becoming increasingly important. A vehicle's braking performance directly affects driving safety, parking safety, and economic efficiency. Electro-hydraulic braking systems may encounter various external disturbances in actual operating conditions, such as sudden changes in road surface adhesion coefficient, load disturbances caused by incline driving, hydraulic oil viscosity shifts due to ambient temperature changes, and unstable hydraulic pressure and nonlinear hysteresis effects. These disturbances can affect the performance, stability, and safety of the hydraulic braking system.

[0003] Current research on algorithms for automotive electro-hydraulic braking systems mostly focuses on single control algorithms, such as PID control, MPC model predictive control, H∞ robust control, and fuzzy control. However, these algorithms can only achieve low-precision hydraulic pressure control to a certain extent and have weak anti-interference capabilities. Summary of the Invention

[0004] In view of this, the present invention provides a method for controlling the hydraulic pressure of wheel cylinders in a brake-by-wire system for electric vehicles, so as to achieve high-precision hydraulic pressure control and improve anti-interference capability.

[0005] A method for controlling the hydraulic pressure of wheel cylinders in a brake-by-wire system for an electric vehicle includes:

[0006] Step S1: Construct a nonlinear model of the electric vehicle's electro-hydraulic braking system, and establish a mathematical model of the braking system, a dynamic balance equation of the braking system, and a dynamic model of the braking system based on the constructed nonlinear model;

[0007] Step S2: Based on the mathematical model of the braking system, the dynamic equilibrium equation of the braking system, and the dynamic model of the braking system established in Step S1, an adaptive extended state observer based on the improved particle swarm optimization algorithm is designed to estimate the initial hydraulic pressure of the electric vehicle wheel cylinder. The improved particle swarm optimization algorithm is based on the particle swarm optimization algorithm by introducing Chebyshev chaotic mapping, random inertia coefficient, and combining the gravitational search algorithm with Gaussian perturbation for optimization.

[0008] Step S3: Based on the adaptive extended state observer designed in step S2 using the improved particle swarm optimization algorithm, and with the estimated preliminary electric vehicle wheel cylinder hydraulic pressure as input, design a sliding mode attitude controller based on the adaptive extended state observer.

[0009] Step S4, based on the adaptive extended state observer based sliding mode attitude controller designed in step S3, the target function and fitness function established by improving the alpha evolution algorithm, the sliding mode surface equation is improved through the target function and fitness function, and the chaos disturbance term, evolution compensation term and adaptive chattering suppression factor are introduced, so that the improved sliding mode control method is obtained, and the wheel cylinder hydraulic pressure of the electric vehicle is controlled through the improved sliding mode control method.

[0010] The electric vehicle brake-by-wire system wheel cylinder hydraulic pressure control method provided by the application has the following beneficial effects:

[0011] (1) The application designs an adaptive extended state observer based on an improved particle swarm optimization algorithm, optimizes the particle swarm optimization algorithm by combining a Chebyshev chaotic mapping, a random inertia coefficient and a gravitational search algorithm and a Gaussian disturbance, and can improve the global search ability of the group.

[0012] (2) Based on the traditional hydraulic pressure control algorithm, the target function and fitness function established by improving the alpha evolution algorithm are used to improve the sliding mode surface equation, and the system dynamic response performance is optimized, so that the electric vehicle hydraulic brake system can still maintain stability when encountering external interference, and the chattering suppression ability is significantly improved.

[0013] (3) In the sliding mode control method, the chaos disturbance term is introduced to enhance the global search ability to avoid local optimization, the evolution compensation term is introduced to further improve the anti-interference ability, and the adaptive chattering suppression factor can effectively weaken the chattering phenomenon of the sliding mode control, and finally realize high-precision and strong-robustness wheel cylinder hydraulic pressure control. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The flowchart of the electric vehicle brake-by-wire system wheel cylinder hydraulic pressure control method provided by the embodiment of the application is shown.

[0015] Figure 2 The comparison chart of the wheel cylinder hydraulic pressure obtained by the method of the application and the hydraulic pressure and ideal hydraulic pressure obtained by the traditional sliding mode control algorithm. DETAILED DESCRIPTION

[0016] The embodiments of the application will be described in detail below, and examples of the embodiments are shown in the drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are intended to explain the embodiments of the application, and cannot be understood as a limitation of the application.

[0017] Please refer to Figure 1The embodiment of the application provides a wheel cylinder hydraulic pressure control method of an electric vehicle regenerative braking system, comprising steps S1-S4:

[0018] In step S1, a nonlinear model of the electric vehicle electronic hydraulic braking system is constructed, and a braking system mathematical model, a braking system dynamics balance equation and a braking system dynamics model are established based on the constructed nonlinear model.

[0019] In step S1, the expression of the braking system mathematical model is as follows:

[0020]

[0021]

[0022]

[0023] wherein, is the equivalent rotational inertia of the rotating part, is the angular acceleration, is the rated torque of the motor, is the torque borne by the primary ball screw, is the axial thrust of the ball screw, is the lead of the ball screw, is the transmission efficiency, is the gear transmission ratio.

[0024] The expression of the braking system dynamics balance equation is as follows:

[0025]

[0026] wherein, is the wheel cylinder mass, is the equivalent damping, is the equivalent spring stiffness of the wheel cylinder, is the piston displacement, is the first-order derivative of , is the second-order derivative of , is the piston area, is the brake wheel cylinder pressure.

[0027] Specifically, the liquid pressure relationship expression in the cavity is as follows:

[0028]

[0029]

[0030] wherein, is the brake wheel cylinder pressure change rate, is the bulk modulus of the brake liquid, is the brake chamber liquid volume change amount, is the brake chamber initial liquid volume, is the brake wheel cylinder output brake liquid flow, is the brake wheel cylinder chamber brake liquid leakage coefficient.

[0031] The expression of the brake system dynamics model is:

[0032]

[0033]

[0034] wherein, is the pressure change rate, is the brake chamber pressure.

[0035] Step S2, based on the brake system mathematical model, the brake system dynamics balance equation and the brake system dynamics model established in step S1, an adaptive extended state observer based on an improved particle swarm optimization algorithm is designed to estimate the preliminary electric vehicle wheel cylinder hydraulic pressure, and the improved particle swarm optimization algorithm is introduced on the basis of the particle swarm optimization algorithm Chebyshev chaos mapping, random inertia coefficient and gravitational search algorithm and Gaussian disturbance combination optimization.

[0036] Firstly, the electric vehicle electronic hydraulic brake system is simplified as a second-order system represented as:

[0037]

[0038] wherein, is the input quantity, is the output quantity second-order derivative, is the total disturbance containing all external disturbance factors, is the control gain.

[0039] Define the intermediate state variable , , , , , , first-order derivative of the output , which can be obtained:

[0040]

[0041]

[0042]

[0043] wherein, is The first derivative, for The first derivative, for The first derivative, It is in a perturbation state.

[0044] Furthermore, the state-space equation can be obtained as follows:

[0045]

[0046]

[0047] in, For input variables, , Indicates transpose. Input variables The first derivative, Here is the state transition matrix. The input coupling matrix, Let be the perturbation coupling matrix. For observation output, The noise coupling matrix is... This is a disturbance signal.

[0048] Therefore, the state equation of the adaptive extended state observer is:

[0049]

[0050]

[0051] in, This is the state estimate. , , , These are the estimated values ​​for the first, second, and third dimensions of the state, respectively. for The first derivative, To output the estimated value, is the observer gain matrix.

[0052] To ensure the observer is stable and has the desired dynamic response, the gain matrix needs to be designed. This ensures that the root of the characteristic equation of the error dynamic equation lies in the left half-plane and has the desired bandwidth.

[0053] Furthermore, the characteristic equation for the adaptive extended state observer gain is:

[0054]

[0055] The characteristic equation of the error dynamic equation is:

[0056] ;

[0057] wherein, is the characteristic equation, is the bandwidth of the observer, is the determinant of a matrix, , , is a system parameter.

[0058] The gain matrix of the observer is obtained as is:

[0059]

[0060] wherein, , , is the observer gain.

[0061] Finally, the expression of the adaptive extended state observer is:

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068] wherein, , , are the estimated values of , , respectively, , , are the estimated values of , , respectively.

[0069] Further, in order to avoid the population aggregation phenomenon caused by traditional random initialization, Chebyshev chaotic mapping is used as the population initialization strategy, and the expression of Chebyshev chaotic mapping is:

[0070]

[0071] wherein, For the first Input for each iteration For the first Input for each iteration The Chebyshev chaotic mapping parameters determine the uniformity of population initialization;

[0072] Adjust the observer bandwidth To minimize the estimation error, The larger the value, the better it is for global search of the optimal value, but the local search capability becomes weaker; The smaller the value, the opposite is true. To improve the dynamics of the inertia coefficient, this paper adopts a random inertia coefficient. Specifically, the expression for the random inertia coefficient is:

[0073]

[0074] in, This is the dynamic frequency value. This is the minimum value of the inertia coefficient. This represents the maximum value of the inertia coefficient. yes Random numbers between It is a value used to measure the degree of deviation between the random inertia coefficient and the expected value. It is a normally distributed random number;

[0075] When the population gets stuck in a local optimum, a combination of gravity search algorithm and Gaussian perturbation is used to improve the global search capability of the swarm intelligence algorithm. The expression for the optimization by combining gravity search algorithm and Gaussian perturbation is:

[0076]

[0077]

[0078]

[0079]

[0080]

[0081] in, For the first The global best position in the nth iteration dimensional components, For the first The global best position in the nth iteration dimensional components, For the first The global best position of the second iteration dimensional components, for To the sine absolute value of the random number within the range, is a random number between 0 and 1, and is the golden section coefficient, is the learning factor, is the proportional parameter, is the time constant, and respectively represent the maximum and minimum values of each dimension of all particles in the current iteration, is a Gaussian distribution function, is a time-varying variance.

[0082] The optimized adaptive extended state observer can estimate the hydraulic pressure of the electric vehicle wheel cylinder in real time to obtain the preliminary electric vehicle wheel cylinder hydraulic pressure.

[0083] Step S3, based on the adaptive extended state observer designed in step S2 based on the improved particle swarm optimization algorithm, and taking the estimated preliminary electric vehicle wheel cylinder hydraulic pressure as input, a sliding mode attitude controller based on the adaptive extended state observer is designed.

[0084] In step S3, the expression of the sliding mode attitude controller based on the adaptive extended state observer is:

[0085]

[0086]

[0087]

[0088]

[0089] wherein, is the observation error, , is the observer gain.

[0090] Step S4, based on the sliding mode attitude controller based on the adaptive extended state observer designed in step S3, the objective function and fitness function established by the improved alpha evolution algorithm, the sliding mode surface equation is improved through the objective function and fitness function, and the chaos disturbance term, evolution compensation term and adaptive chattering suppression factor are introduced, thereby obtaining the improved sliding mode control method, and the hydraulic pressure of the electric vehicle wheel cylinder is controlled through the improved sliding mode control method.

[0091] In step S4, the expression of the improved alpha evolution algorithm is:

[0092]

[0093]

[0094]

[0095] in, For the first The updated values ​​of the vectors in the initial step. Let be a solution vector randomly selected from the current population. This represents the optimal solution vector in the current population. For adaptive crossover probability, The initial exploration probability, The decay rate coefficient, For hyperparameters, To improve the time step index of the Alpha Evolutionary Algorithm, For dynamic disturbance intensity, With a mean of 0 and a variance of Gaussian noise, This represents the maximum boundary value within the range of possible values.

[0096] The evolutionary equation expression for the improved Alpha Evolutionary Algorithm is as follows:

[0097]

[0098]

[0099]

[0100]

[0101]

[0102] in, For particles At any moment The velocity vector, For particles At any moment The velocity vector, For particles The optimal position of an individual For the first Particles in the next iteration Location, This represents the globally optimal position of the particle swarm. For the first Particles in the next iteration Location; This is a chaotic perturbation term used to escape local optima; For chaotic perturbation functions, To be with particles the relevant variable, is the inertia weight, and are the maximum and minimum inertia weight, respectively, is the initial amplitude of the disturbance, is the control parameter, is the decay coefficient, , is the learning factor, , is the adaptive constant.

[0103] wherein the expression of the objective function is:

[0104]

[0105]

[0106]

[0107] wherein, is the objective function, is the norm of the tracking error , is the first derivative of the tracking error , is the sign function, is the desired trajectory, is the actual output trajectory; is the error normalization function, used to suppress overcompensation when the error is large; , , is the dynamic weight constant, is the differential of time , is the total duration of the control process.

[0108] wherein the expression of the fitness function is:

[0109]

[0110] wherein, is the fitness function, is the weight coefficient, is the control cost, is the control input that changes over time .

[0111] wherein the expression of the improved sliding mode surface equation is:

[0112]

[0113]

[0114]

[0115]

[0116]

[0117] in, For improved sliding surface, , For the first Sliding surface parameters at the next iteration , For the first Sliding surface parameters at the next iteration It is an adaptive chatter suppression factor. Let be the error function. , , As a weighting factor, It is the hyperbolic tangent function. It is the composite control rate. For sliding mode control items, For evolutionary compensation terms, For control parameters, This is the stiffness coefficient. This is an estimated value.

[0118] In step S4, the expression for the chaotic perturbation term is:

[0119]

[0120]

[0121] in, For chaotic perturbation terms, The coupling coefficient is... These are state parameters;

[0122] The expression for the evolutionary compensation term is:

[0123]

[0124] in, For the first Each weighting coefficient For feature mapping function, It is a time function. This represents the total number of weight coefficients for the collaborative compensation items;

[0125] The expression for the adaptive chattering suppression factor is:

[0126]

[0127] wherein, and are the minimum and maximum values of the chattering suppression gain respectively, is a natural exponential function, is a parameter for the decay of the control gain with the error.

[0128] Specifically, stability analysis is performed again through Lyapunov function, and the specific formula is as follows:

[0129]

[0130]

[0131] wherein, is a Lyapunov function, is a first derivative of is a sliding mode variable, is an adjustment parameter, is an estimated value, , is a sliding mode convergence coefficient, is an estimated error convergence coefficient. When

[0132] , and , the system is globally asymptotically stable. Finally, the wheel cylinder hydraulic pressure of the electric vehicle is controlled through the improved sliding mode control method.

[0133]

[0134] is a comparison chart of the wheel cylinder hydraulic pressure obtained by the method of the present application, the hydraulic pressure obtained by the traditional sliding mode control algorithm and the ideal hydraulic pressure, and Figure 2 it can be known that the curve of the method proposed in the present application is closest to the ideal value, the response speed is shortened by 30%, the chattering is suppressed to within 0.02 MPA, high-precision hydraulic pressure control can be realized, the anti-interference ability and robustness are effectively improved, and the influence of external interference on the hydraulic brake system is reduced. Figure 2 In summary, the wheel cylinder hydraulic pressure control method of the electric vehicle brake-by-wire system according to the above-mentioned embodiments has the following beneficial effects:

[0135] (1) The adaptive extended state observer based on the improved particle swarm optimization algorithm is designed, the particle swarm optimization algorithm is optimized by combining the Chebyshev chaotic mapping, random inertia coefficient and gravitational search algorithm and Gaussian disturbance optimization, and the global search ability of the group is improved.

[0136]

[0137] ​(2) On the basis of the traditional hydraulic force control algorithm, the target function and the fitness function are established by improving the alpha evolution algorithm, the sliding mode surface equation is improved through the target function and the fitness function, the system dynamic response performance is optimized, the electric vehicle hydraulic brake system can still maintain stability when encountering external interference, and the chattering suppression ability is significantly improved;

[0138] (3) In the sliding mode control method, the chaos disturbance term is introduced to enhance the global search ability to avoid local optimization, the evolution compensation term is introduced to further improve the anti-interference ability, and the adaptive chattering suppression factor can effectively weaken the chattering phenomenon of the sliding mode control, and finally the high-precision and strong-robustness wheel cylinder hydraulic pressure control is realized.

[0139] The above-mentioned embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it cannot be understood as the limitation of the scope of the patent. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for controlling hydraulic pressure of a wheel cylinder of an electric vehicle brake-by-wire system, characterized by, Comprise: Step S1, the nonlinear model of electric vehicle electronic hydraulic brake system is constructed, the mathematical model of brake system, the dynamic balance equation of brake system and the dynamic model of brake system are established based on the constructed nonlinear model; Step S2, based on the mathematical model of brake system, the dynamic balance equation of brake system and the dynamic model of brake system established in step S1, the adaptive extended state observer based on improved particle swarm optimization algorithm is designed to estimate the preliminary electric vehicle wheel cylinder pressure, the improved particle swarm optimization algorithm is introduced into Chebyshev chaotic mapping, random inertia coefficient and gravitational search algorithm and Gaussian disturbance combination optimization based on particle swarm optimization algorithm; Step S3, based on the adaptive extended state observer based on improved particle swarm optimization algorithm designed in step S2, and taking the estimated preliminary electric vehicle wheel cylinder pressure as input, the sliding mode attitude controller based on adaptive extended state observer is designed; Step S4, based on the sliding mode attitude controller based on adaptive extended state observer designed in step S3, the objective function and fitness function are established by improving alpha evolution algorithm, the sliding mode surface equation is improved through the objective function and fitness function, and the chaotic disturbance term, evolution compensation term and adaptive chattering suppression factor are introduced, so that the improved sliding mode control method is obtained, and the electric vehicle wheel cylinder pressure is controlled through the improved sliding mode control method.

2. The electric vehicle regenerative braking system wheel cylinder pressure control method of claim 1, wherein, In step S1, the expression of the mathematical model of brake system is: wherein, is the equivalent moment of inertia of the rotating parts, is the angular acceleration, is the motor rated torque, is the torque on the primary ball screw, is the axial thrust on the ball screw, is the lead of the ball screw, is the transmission efficiency, is the gear ratio; The expression of the dynamic balance equation of brake system is: wherein, is the wheel cylinder mass, is the equivalent damping, is the wheel cylinder equivalent spring rate, is the piston displacement, is the first derivative of is the second derivative of is the piston area, is the brake wheel cylinder pressure; The expression of the dynamic model of brake system is: wherein, is the rate of change of pressure, is the brake chamber pressure, is the brake fluid bulk modulus of elasticity, is the initial fluid volume of the brake chamber.

3. The electric vehicle regenerative braking system wheel cylinder pressure control method of claim 2, wherein, In step S2, the expression of Chebyshev chaotic mapping is: wherein, is the first input to the step iteration, is the first input to the step iteration, is a Chebyshev chaos map parameter; The expression of random inertia coefficient is: wherein, is a dynamic frequency value, is a minimum value of the inertia coefficient, is a maximum value of the inertia coefficient, is is a random number between is a value for measuring the degree of deviation between the random inertia coefficient and the expected value, is a random number of normal distribution; The expression of gravitational search algorithm and Gaussian disturbance combination optimization is: wherein is the global best position of the th iteration, is the th component of the global best position of the th iteration, is the th component of the global best position of the th iteration, is the th component of the global best position of the th iteration, is the sine absolute value of a random number in the range is a random number between is the golden section coefficient, is the learning factor, is the proportionality parameter, is the time constant, denote the maximum and minimum values of each dimension for all particles at the current iteration number, is the Gaussian distribution function, is the time-varying variance.​​​ 4. The electric vehicle regenerative braking system wheel cylinder pressure control method of claim 3, wherein, In step S4, the expression of improved alpha evolution algorithm is: in, For the first The updated values ​​of the vectors in the initial step, Let be a solution vector randomly selected from the current population. This represents the optimal solution vector in the current population. For adaptive crossover probability, The initial exploration probability, The decay rate coefficient, For hyperparameters, To improve the time step index of the Alpha Evolutionary Algorithm, For dynamic disturbance intensity, With a mean of 0 and a variance of Gaussian noise, This represents the maximum boundary value within the range of possible values.

5. The electric vehicle regenerative braking system wheel cylinder pressure control method of claim 4, wherein, In step S4, the evolution equation expression of improved alpha evolution algorithm is: wherein is the position of the particle at time is the velocity vector of the particle at time is the position of the particle at time is the velocity vector of the particle at time is the individual optimal position of the particle at the th iteration, is the global optimal position of the particle swarm, is the position of the particle at the th iteration, is the chaotic disturbance term, is the chaotic disturbance function, is the variable related to the particle , is the inertia weight, and are the maximum and minimum inertia weight, respectively, is the initial amplitude of the disturbance, is the control parameter, is the decay coefficient, , is the learning factor, , is the adaptive constant.

6. The electric vehicle regenerative braking system wheel cylinder pressure control method of claim 5, wherein, In step S4, the expression of objective function is: wherein, is a target function, is a tracking error norm, is a first derivative of the tracking error , is a sign function, is a desired trajectory, is an actual output trajectory, is an error normalization function, , , is a dynamic weight constant, is a differential of time , is a total duration of the control process; In step S4, the expression of fitness function is: in, For the fitness function, These are the weighting coefficients. To control costs, For time Changing control inputs.

7. The electric vehicle regenerative braking system wheel cylinder pressure control method of claim 6, wherein, The expression of improved sliding mode surface equation is: wherein, is the improved sliding surface, , is the first iterative sliding surface parameter, , is the first iterative sliding surface parameter, is the adaptive chattering damping factor, is the error function, , , is the weight factor, is the hyperbolic tangent function, is the compound control rate, is the sliding mode control term, is the evolutionary compensation term, is the control parameter, is the stiffness coefficient, is the estimated value.

8. The electric vehicle regenerative braking system wheel cylinder pressure control method of claim 7, wherein, In step S4, the expression of chaotic disturbance term is: wherein, is a chaotic perturbation term, is a coupling coefficient, is a state parameter; The expression of evolution compensation term is: wherein is a first weight coefficient, is a second weight coefficient, is a feature mapping function, is a time function, is a total number of weight coefficients of the synergistic compensation term; The expression of adaptive chattering suppression factor is: wherein and are the minimum and maximum values of the buffeting suppression gain, respectively, is a natural exponential function, is a parameter controlling the decay of the gain with the error.

Images