An adaptive pressure control method for electronic power-assisted braking system considering multi-dimensional nonlinear disturbances

Abstract

The present invention belongs to the field of automotive technology, specifically, a method for adaptive pressure control of an electronic power-assisted braking system that considers multidimensional nonlinear disturbances. The method comprises the following steps: Step 1: Establishing key models of the motor, hydraulic pressure, and friction; Step 2: Designing a hydraulic-position-current cascade controller to achieve adaptive pressure control of the electronic power-assisted braking system. The hydraulic controller utilizes an adaptive radial basis function neural network and robust sliding mode theory to address time-varying uncertainty disturbances in the hydraulic system; the position controller incorporates friction feedforward compensation and sliding mode control to address nonlinear friction obstruction in the transmission mechanism; and the current controller utilizes Lyapunov theory to design a current decoupling solution to address the dynamic coupling of electromagnetic characteristics of a permanent magnet synchronous motor. The present invention utilizes an adaptive radial basis function neural network and robust sliding mode theory to design a hydraulic controller for the electronic power-assisted braking system; utilizes friction feedforward compensation and sliding mode control theory to design a position controller for the electronic power-assisted braking system; and utilizes Lyapunov theory to design a current controller for the electronic power-assisted braking system. Ultimately, the method achieves adaptive pressure control of the electronic power-assisted braking system, providing a reasonable solution for high-quality active braking pressure control in intelligent vehicles.

Description

Technical Field

[0001] The present invention belongs to the field of automobile technology, and in particular relates to an adaptive pressure control method for an electronic power-assisted braking system taking multi-dimensional nonlinear disturbances into consideration. Background Art

[0002] The increasing safety demands of smart electric vehicles require chassis braking systems to possess precise and rapid active braking capabilities. To this end, major research institutions both domestically and internationally have conducted extensive research on electronic power-assisted braking systems (EPBRs) that utilize a motor and transmission mechanism as their core while retaining a traditional hydraulic electronic control unit. During active braking, these electromechanical-hydraulic coupled EPBs face complex multidimensional nonlinear perturbations, posing significant challenges to precise pressure control.

[0003] Currently, some researchers have conducted research on precise pressure control in electronic power-assisted braking systems. Brake pressure control in electronic power-assisted braking systems can typically be achieved by fitting the dynamic mapping characteristic (PV characteristic) between the actual pressure in the master cylinder and the input pushrod. However, due to factors such as brake fluid leakage and friction between the brake fluid and the pipeline, electronic power-assisted braking systems are subject to complex time-varying hydraulic uncertainty disturbances. Fitting a single PV characteristic cannot accurately reflect the operating state of the electronic power-assisted braking system. In contrast, accurately modeling the hydraulic system through mathematical models can help electronic power-assisted braking systems achieve better pressure control performance. However, these complex hydraulic models, used to represent time-varying hydraulic uncertainty disturbances, have numerous internal parameters and are difficult to accurately obtain. Furthermore, the servo motors within electronic power-assisted braking systems require a transmission mechanism to achieve brake pressure control, which presents significant nonlinear friction interference such as static friction, Coulomb friction, and viscous friction. Furthermore, the servo motors typically used in electronic power-assisted braking systems are high-performance permanent magnet synchronous motors. The dynamic coupling of the electromagnetic characteristics of the motor's electromagnetic shaft and torque shaft during high-speed rotation of permanent magnet synchronous motors also affects the pressure control performance of electronic power-assisted braking systems. Summary of the Invention

[0004] To solve the above problems, the present invention provides an adaptive pressure control method for an electronic power-assisted braking system taking into account multi-dimensional nonlinear disturbances. An adaptive radial basis function neural network and robust sliding mode theory are adopted to design a hydraulic controller for the electronic power-assisted braking system. Friction feedforward compensation and sliding mode control theory are adopted to design a position controller for the electronic power-assisted braking system. Lyapunov theory is adopted to design a current controller for the electronic power-assisted braking system. Ultimately, adaptive pressure control of the electronic power-assisted braking system is achieved, providing a reasonable solution for high-quality active braking pressure control of smart cars.

[0005] The technical solution of the present invention is described as follows in conjunction with the accompanying drawings:

[0006] An adaptive pressure control method for an electronic power-assisted braking system considering multi-dimensional nonlinear disturbances comprises the following steps:

[0007] Step 1: Establish key models of motor, hydraulic pressure and friction;

[0008] Step 2: Design a hydraulic-position-current cascade controller to achieve adaptive pressure control of the electronic power-assisted braking system; the hydraulic controller uses an adaptive radial basis function neural network and robust sliding mode theory to solve the problem of time-varying uncertainty disturbance of the hydraulic system; the position controller introduces friction feedforward compensation and sliding mode control to solve the problem of nonlinear friction obstruction of the transmission mechanism; the current controller uses Lyapunov theory to design a current decoupling solution to solve the problem of dynamic coupling of the electromagnetic characteristics of the permanent magnet synchronous motor.

[0009] Furthermore, the specific method of step one is as follows:

[0010] 11) According to the torque balance relationship of the permanent magnet synchronous motor output shaft, the motor motion equation is established:

[0011]

[0012] Where, f Indicates the motor load torque; T e Represents the electromagnetic torque of the motor; Represents the motor's mechanical angular acceleration; J represents the motor's moment of inertia;

[0013] 12) Establish hydraulic system model;

[0014] Assuming that the brake pressures of the master cylinder and wheel cylinder are consistent, the hydraulic compensation term f(P) is introduced to characterize the time-varying uncertainty disturbance of the hydraulic system. The final hydraulic system model is:

[0015]

[0016] Where, Indicates the master cylinder push rod speed; A indicates the effective area of ​​the master cylinder piston; Indicates the actual pressure change rate of the brake master cylinder; P indicates the actual pressure of the brake master cylinder; K c Represents the brake fluid bulk modulus; V m Indicates the volume of brake fluid in the master cylinder; V w represents the volume of brake fluid in the wheel cylinder; f(P) represents the hydraulic compensation term; x represents the actual master cylinder push rod displacement;

[0017] 13) Modeling static friction, Coulomb friction, and viscous friction:

[0018]

[0019] Where, T f represents the friction torque; ρ v represents the viscous friction coefficient; ω m Represents the motor mechanical angular velocity; ρ c represents the Coulomb friction factor; T c Represents the Coulomb friction torque when the system is unloaded; ε n represents the motor speed threshold; n represents the gear ratio; h represents the ball screw lead; T s Represents the static friction torque.

[0020] Furthermore, the specific method for designing a hydraulic controller based on an adaptive radial basis function neural network is as follows:

[0021] 211) Let the target master cylinder push rod speed be the hydraulic control law, that is, Convert the hydraulic system model into

[0022]

[0023] Where, α represents the first parameter of the hydraulic system model; β represents the second parameter of the hydraulic system model; u represents the hydraulic control law;

[0024] 212) According to the target pressure P of the brake master cylinder * The difference between the actual pressure P of the brake master cylinder and the actual pressure P is p Establish the sliding surface of the hydraulic controller:

[0025] s p =σ1e p +∫e p dt

[0026] Where s p represents the hydraulic controller sliding surface; σ1 represents the first parameter of the hydraulic sliding controller, and σ1>0; e p Indicates the difference between the target pressure of the brake master cylinder and the actual pressure of the brake master cylinder;

[0027] 213) Correspondingly, the differential of the sliding surface of the hydraulic controller is solved as follows:

[0028]

[0029] Where, represents the differential of the sliding surface of the hydraulic controller; Indicates the differential between the target pressure of the brake master cylinder and the actual pressure of the brake master cylinder; Indicates the target pressure differential of the brake master cylinder;

[0030] 214) Use the exponential reaching law containing saturation function:

[0031]

[0032]

[0033] Where c represents the second parameter of the hydraulic sliding mode controller, and c>0; q represents the third parameter of the hydraulic sliding mode controller, and q>0; sat(s p ) represents the inclusion function; ξ represents the saturation function threshold;

[0034] 215) Substituting the exponential reaching law containing the saturation function into the differential of the sliding surface of the hydraulic controller, the hydraulic control law is simplified to:

[0035]

[0036] Where, sat(s p ) indicates the included function;

[0037] 216) Use radial basis function neural network to approximate and fit it; take the actual pressure of the brake master cylinder as the network input, and the estimated value of the hydraulic compensation term As the network output; the radial basis neural network algorithm is designed as:

[0038]

[0039] Where, represents the estimated value of the hydraulic compensation term; Represents the estimated network weights; h(P) represents the Gaussian basis function network output; c j represents the center value of the Gaussian basis function of the jth neuron in the hidden layer of the network; b j is the width of the Gaussian basis function of the jth neuron in the hidden layer;

[0040] 217) Ideal value of hydraulic compensation term f(P) * Expressed as:

[0041] f(P) * =W *T h(P)+ε p

[0042] Where f(P) * represents the ideal value of the hydraulic compensation term; W *T represents the ideal network weight; h(P) represents the Gaussian basis function network output; ε p represents the network approximation error, and ε p ≤|T|, T represents a bounded real number;

[0043] 218) Set the difference between the ideal network weight and the estimated network weight to be get:

[0044]

[0045] Where, It represents the difference between the ideal hydraulic compensation term and the estimated hydraulic compensation term; Represents the difference between the ideal network weight and the estimated network weight;

[0046] 219) Estimated value of hydraulic compensation Into the hydraulic control law, establish a new hydraulic control law u p :

[0047]

[0048] Where u p represents the new hydraulic control law; represents the estimated value of the hydraulic compensation term; sat(s p ) indicates the included function; s p represents the sliding surface of the hydraulic controller;

[0049] 2110) The new hydraulic control law u p Substituting the differential of the sliding surface of the hydraulic controller, we can simplify it to get:

[0050]

[0051] 2111) Design of hydraulic layer Lyapunov function The differential form is:

[0052]

[0053] Where, represents the differential of the Lyapunov function of the hydraulic layer; μ represents the coefficient of the Lyapunov function of the hydraulic layer; represents the estimated network weight differential;

[0054] 2112) According to Lyapunov's direct method, the adaptive law of the radial basis neural network is designed as follows:

[0055]

[0056] 2113) Further simplify the differential form of the hydraulic layer Lyapunov function:

[0057]

[0058] Considering the network approximation error ε p ≤|T| is bounded, so we choose When , the hydraulic layer Lyapunov function can be differentiated Ensure the closed-loop robustness of the hydraulic control system;

[0059] Finally, the target master cylinder push rod displacement is:

[0060] x * =∫u p dt

[0061] Where x * Indicates the target master cylinder push rod displacement; u p Represents the new hydraulic control law.

[0062] Furthermore, the specific method for designing a position controller based on friction model compensation is as follows:

[0063] 221) Use angle proportional-integral control to establish the target motor mechanical angular velocity

[0064]

[0065] Where, Indicates the target motor mechanical angular velocity; k p Indicates the position control proportional gain; k i Indicates the position control integral gain; x * Indicates the target master cylinder push rod displacement;

[0066] 222) In the speed sliding mode control, the target torque axis current is selected as the control input, that is, Then the state variable equation of the speed system is established as:

[0067]

[0068] Where, represents the differential of the motor's mechanical angular acceleration; J represents the motor's moment of inertia; P n Indicates the number of motor poles; ψ f Represents the motor rotor flux; represents the differential of the speed sliding mode control law;

[0069] 223) Set the difference between the target motor mechanical angular velocity and the actual motor mechanical angular velocity As the speed sliding mode tracking error, the speed controller sliding mode surface is established:

[0070]

[0071] Where σ2 represents the first parameter of the speed sliding mode controller, and σ2>0; s ω represents the sliding surface of the speed controller; e ω Indicates the difference between the target motor mechanical angular velocity and the actual motor mechanical angular velocity; Indicates the differential between the target motor mechanical angular velocity and the actual motor mechanical angular velocity;

[0072] 224) The exponential approach function is selected as the speed sliding mode approach law:

[0073]

[0074] Where s ω represents the sliding surface of the speed controller; represents the differential of the sliding surface of the speed controller; k represents the second parameter of the speed sliding controller, and k>0; d represents the third parameter of the speed sliding controller, and d>0; sgn(s ω ) represents the sign function related to the rotation speed;

[0075] 225) By combining the speed controller sliding mode surface, the speed sliding mode reaching law and the speed system state variable equation, the motor torque shaft feedback current is obtained as:

[0076]

[0077] Where, Indicates the motor torque shaft feedback current; u s represents the speed sliding mode control law; k represents the second parameter of the speed sliding mode controller, and k>0; Indicates the differential between the target motor mechanical angular velocity and the actual motor mechanical angular velocity;

[0078] 226) The motor torque axis feedforward compensation current based on the friction model is introduced into the position controller:

[0079]

[0080] Where: K t Indicates the motor torque coefficient; T f represents friction torque; Indicates the motor torque axis feedforward compensation current;

[0081] 227) Finally, the output target motor torque shaft current is:

[0082]

[0083] Where, Indicates the motor torque axis feedforward compensation current; Indicates the motor torque axis feedback current; Indicates the target motor torque shaft current.

[0084] Furthermore, the specific method of designing a current controller based on Lyapunov theory is as follows:

[0085] 231) Establish the current tracking error as:

[0086]

[0087]

[0088] Where i d Indicates the actual motor excitation shaft current; i q Indicates the actual motor excitation torque shaft current; Indicates the target motor excitation shaft current, and sets Indicates the target motor torque shaft current; e d Indicates the difference between the target motor excitation shaft current and the actual motor excitation shaft current; e q Indicates the difference between the target motor torque axis current and the actual motor torque axis current;

[0089] 232) The corresponding current tracking error differential is:

[0090]

[0091] Where, Indicates the differential between the target motor excitation shaft current and the actual motor excitation shaft current; Indicates the differential difference between the target motor torque shaft current and the actual motor torque shaft current; u d Indicates the motor excitation shaft voltage; u q Represents the motor torque shaft voltage; R is the motor stator resistance; L d Indicates the motor excitation shaft inductance; L q Represents the motor torque shaft inductance; ψ f Represents the motor rotor flux; ω e Indicates the motor electrical angular velocity;

[0092] 233) Define the Lyapunov function of the motor excitation axis current controller as The Lyapunov function of the motor torque axis current controller is defined as The differential form is:

[0093]

[0094] Where, represents the differential of the Lyapunov function of the motor excitation axis current controller; represents the differential of the Lyapunov function of the motor torque axis current controller;

[0095] 234) According to Lyapunov direct method, the motor excitation axis current control law and the motor torque axis current control law are designed as follows:

[0096]

[0097] Where, Represents the motor excitation shaft current control law; Represents the motor torque axis current control law; γ1 represents the motor excitation axis control parameter, and setting γ1>-R / L d ; γ1 represents the motor torque axis control parameter, and setting γ2>-R / L q .

[0098] The beneficial effects of the present invention are:

[0099] 1) This paper adopts adaptive radial basis function neural network and robust sliding mode theory to design the hydraulic controller of electronic power-assisted braking system, which solves the problem of time-varying uncertainty disturbance of hydraulic pressure.

[0100] 2) The present invention adopts friction feedforward compensation and sliding mode control theory to design an electronic power-assisted braking system position controller, overcoming the significant nonlinear friction interference such as static friction, Coulomb friction and viscous friction within the mechanism;

[0101] 3) The present invention adopts Lyapunov theory to design a current controller for an electronic power-assisted braking system, which solves the problem of dynamic coupling of electromagnetic characteristics between the electromagnetic shaft and the torque shaft of a permanent magnet synchronous motor during high-speed rotation. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0103] Figure 1 It is a schematic diagram of the architecture of the present invention;

[0104] Figure 2 This is a schematic diagram of the braking system structure based on the electronic power-assisted braking system;

[0105] Figure 3 Schematic diagram of the control effect. DETAILED DESCRIPTION

[0106] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0107] Example 1

[0108] This embodiment provides an adaptive pressure control method for an electronic power-assisted braking system taking into account multi-dimensional nonlinear disturbances.

[0109] Figure 1 This is a diagram of the architecture of an adaptive pressure control method for an electronic power-assisted brake system designed by the present invention considering multi-dimensional nonlinear disturbances. The specific description is as follows: (1) In the hydraulic controller, an adaptive radial basis function neural network is used to calculate the actual pressure P of the target brake master cylinder. * The hydraulic compensation term f(P) is dynamically estimated based on the actual brake master cylinder pressure P, which helps the robust sliding mode control output the target master cylinder push rod horizontal displacement x * (2) In the position controller, the angle proportional-integral control is based on the target master cylinder push rod horizontal displacement x * And the actual master cylinder push rod horizontal displacement x, output the target motor mechanical angular velocity Combined with the actual motor mechanical angular velocity ω m Helps speed sliding mode control generate initial motor torque shaft current On this basis, a feedforward compensation current based on the friction model is added Final output target motor torque shaft current (3) In the current controller, the target motor torque shaft current and the actual torque axis current i q , target excitation shaft current and the actual excitation shaft current i d As input, the excitation shaft voltage u is obtained through the current decoupling control designed by Lyapunov theory. d , torque shaft voltage u q , as follows:

[0110] Step 1: Establish key models for motor, hydraulics, and friction, as follows:

[0111] The structure of the brake system based on the electronic power-assisted brake system is as follows: Figure 2 As shown, it consists of brake pedal, electronic power-assisted braking system, brake master cylinder, hydraulic control unit and other components.

[0112] The specific working principle is as follows:

[0113] (1) Based on the top-level braking requirements, the permanent magnet synchronous motor inside the electronic power-assisted braking system converts the motor's output torque into horizontal thrust for the power-assisted valve body through a secondary transmission mechanism consisting of gears and a ball screw. (2) Under this force, the power-assisted valve body overcomes the preload of the master cylinder return spring, pushing the master cylinder push rod and piston horizontally, driving the brake fluid in the reservoir and master cylinder through the hydraulic electronic control unit into the brake wheel cylinder. (3) The brake fluid in the wheel cylinder is squeezed to generate braking pressure, causing the brake caliper to clamp the disc brake, realizing the active braking function.

[0114] Assuming that the permanent magnet synchronous motor is an ideal motor and ignoring eddy current and hysteresis losses, the stator voltage equation is established on the excitation axis-torque axis:

[0115]

[0116] Where u d Indicates the motor excitation shaft voltage; u q Represents the motor torque shaft voltage; R is the motor stator resistance; i d Indicates the actual motor excitation shaft current; i q Indicates the actual motor excitation torque shaft current; L d Indicates the motor excitation shaft inductance; L q Represents the motor torque shaft inductance; ψ f Represents the motor rotor flux; ω e Indicates the motor electrical angular velocity;

[0117] From the stator voltage equation, we can see that the permanent magnet synchronous motor faces the problem of motor electromagnetic characteristics coupling. d Affected by the actual motor torque shaft current i q Influence, motor torque shaft voltage u q Affected by the actual motor excitation shaft current i d The influence of the permanent magnet synchronous motor is e Rising and getting bigger.

[0118] The electromagnetic torque equation of the surface-mounted permanent magnet synchronous motor is:

[0119] T e =1.5P n i q ψ f (2)

[0120] Where, T e Represents the electromagnetic torque of the motor; P n Indicates the number of motor pole pairs; i q Indicates the actual motor excitation torque shaft current; ψ f Represents the motor rotor flux.

[0121] According to the torque balance relationship of the permanent magnet synchronous motor output shaft, the motor motion equation is established:

[0122]

[0123] Where, T L Indicates the motor load torque; T e Indicates the electromagnetic torque of the motor; represents the motor's mechanical angular acceleration; J represents the motor's moment of inertia.

[0124] Subsequently, a hydraulic system model was established. Due to factors such as brake fluid leakage and friction between the brake fluid and the pipes, the electronic power-assisted braking system suffers from complex time-varying hydraulic uncertainty disturbances. This paper assumes that the brake pressures of the master cylinder and wheel cylinders are consistent and introduces a hydraulic compensation term f(P) to characterize the time-varying hydraulic uncertainty disturbances in the hydraulic system. The resulting hydraulic system model is:

[0125]

[0126] Where, represents the master cylinder push rod speed; x represents the master cylinder push rod displacement; A represents the effective area of ​​the master cylinder piston; Indicates the actual pressure change rate of the brake master cylinder; P indicates the actual pressure of the brake master cylinder; K c Represents the brake fluid bulk modulus; V m Indicates the volume of brake fluid in the master cylinder; V w represents the volume of wheel cylinder brake fluid; f(P) represents the hydraulic compensation term.

[0127] In addition, the servo motor built into the electronic power-assisted braking system needs to control the brake pressure through a transmission mechanism. Nonlinear friction obstacles such as static friction, Coulomb friction, and viscous friction will also have a certain impact on the pressure control performance of the electronic power-assisted braking system. To this end, the present invention uses the following friction model to model these friction forces:

[0128]

[0129] Where, T f represents the friction torque; ρ v represents the viscous friction coefficient; ω m Represents the motor mechanical angular velocity; ρ c represents the Coulomb friction coefficient; A represents the effective area of ​​the master cylinder piston; P represents the actual pressure of the brake master cylinder; T c Represents the Coulomb friction torque when the system is unloaded; ε n is the motor speed threshold; T L represents the motor load torque; n is the gear ratio; h is the ball screw lead; T s is the static friction torque.

[0130] Step 2: Design a hydraulic-position-current cascade controller to achieve adaptive pressure control of the electronic power-assisted braking system. The hydraulic controller uses an adaptive radial basis function neural network and robust sliding mode theory to address the time-varying uncertainty disturbance of the hydraulic system. The position controller introduces friction feedforward compensation and sliding mode control to address the nonlinear friction obstruction of the transmission mechanism. The current controller uses Lyapunov theory to design a current decoupling solution to address the dynamic coupling of the electromagnetic characteristics of the permanent magnet synchronous motor. The details are as follows:

[0131] The specific method of designing a hydraulic controller based on an adaptive radial basis function neural network is as follows:

[0132] 211) Let the target master cylinder push rod speed be the hydraulic control law, that is, Convert the hydraulic system model into

[0133]

[0134]

[0135] Where, α represents the first parameter of the hydraulic system model; β represents the second parameter of the hydraulic system model; u represents the hydraulic control law;

[0136] 212) According to the target pressure P of the brake master cylinder * The difference between the actual pressure P of the brake master cylinder and the actual pressure P is p Establish the sliding surface of the hydraulic controller:

[0137] s p =σ1e p +∫e p dt (9)

[0138] Where s p represents the hydraulic controller sliding surface; σ1 represents the first parameter of the hydraulic sliding controller, and σ1>0; e p Indicates the difference between the target pressure of the brake master cylinder and the actual pressure of the brake master cylinder;

[0139] 213) Correspondingly, the differential of the sliding surface of the hydraulic controller is solved as follows:

[0140]

[0141] Where, represents the differential of the sliding surface of the hydraulic controller; Indicates the differential between the target pressure of the brake master cylinder and the actual pressure of the brake master cylinder; Indicates the target pressure differential of the brake master cylinder;

[0142] 214) Use the exponential reaching law containing saturation function:

[0143]

[0144] Where c represents the second parameter of the hydraulic sliding mode controller, and c>0; q represents the third parameter of the hydraulic sliding mode controller, and q>0; sat(s p ) represents the inclusion function; ξ represents the saturation function threshold;

[0145] 215) Substituting the exponential reaching law containing the saturation function into the differential of the sliding surface of the hydraulic controller, the hydraulic control law is simplified to:

[0146]

[0147] Where, sat(s p ) indicates the included function;

[0148] 216) Considering that the hydraulic compensation term f(P) in the hydraulic control law (13) is difficult to measure accurately, the present invention uses a radial basis function neural network with advantages such as simple structure and low computing power to approximate and fit it. The actual pressure of the brake master cylinder is used as the network input, and the estimated value of the hydraulic compensation term is As the network output. The radial basis neural network algorithm is designed as follows:

[0149]

[0150] Where, represents the estimated value of the hydraulic compensation term; Represents the estimated network weights; h(P) represents the Gaussian basis function network output; c j represents the center value of the Gaussian basis function of the jth neuron in the hidden layer of the network; b j is the width of the Gaussian basis function of the jth neuron in the hidden layer;

[0151] 217) Correspondingly, the ideal value of the hydraulic compensation term f(P) * Expressed as:

[0152] f(P) * =W *T h(P)+ε p (16)

[0153] Where f(P) * represents the ideal value of the hydraulic compensation term; W *T represents the ideal network weight; h(P) represents the Gaussian basis function network output; ε p represents the network approximation error, and ε p ≤|T|, T represents a bounded real number;

[0154] 218) Set the difference between the ideal network weight and the estimated network weight to be get:

[0155]

[0156] Where, It represents the difference between the ideal hydraulic compensation term and the estimated hydraulic compensation term; Represents the difference between the ideal network weight and the estimated network weight;

[0157] 219) Estimated value of hydraulic compensation Into the hydraulic control law, establish a new hydraulic control law u p :

[0158]

[0159] Where u p represents the new hydraulic control law; represents the estimated value of the hydraulic compensation term; sat(s p ) indicates the included function; s p represents the sliding surface of the hydraulic controller;

[0160] 2110) The new hydraulic control law u p Substituting the sliding surface differential (9) of the hydraulic controller, we can simplify it to obtain:

[0161]

[0162] 2111) Design of hydraulic layer Lyapunov function The differential form is:

[0163]

[0164] Where, represents the differential of the Lyapunov function of the hydraulic layer; μ represents the coefficient of the Lyapunov function of the hydraulic layer; represents the estimated network weight differential;

[0165] 2112) According to Lyapunov's direct method, the adaptive law of the radial basis neural network is designed as follows:

[0166]

[0167] 2113) Further simplify the differential form of the hydraulic layer Lyapunov function:

[0168]

[0169] Considering the network approximation error ε p ≤|T| is bounded, so we choose When , the hydraulic layer Lyapunov function can be differentiated Ensure the closed-loop robustness of the hydraulic control system;

[0170] Finally, the target master cylinder push rod displacement is:

[0171] x * =∫u p dt (23)

[0172] Where x * Indicates the target master cylinder push rod displacement; u p Represents the new hydraulic control law.

[0173] The specific method for designing a position controller based on friction model compensation is as follows:

[0174] 221) Use angle proportional-integral control to establish the target motor mechanical angular velocity

[0175]

[0176] Where, Indicates the target motor mechanical angular velocity; k p Indicates the position control proportional gain; k i Indicates the position control integral gain; x * Indicates the target master cylinder push rod displacement;

[0177] 222) In the speed sliding mode control, the target torque axis current is selected as the control input, that is, Then the state variable equation of the speed system is established as:

[0178]

[0179] Where, represents the differential of the motor's mechanical angular acceleration; J represents the motor's moment of inertia; P n Indicates the number of motor poles; ψ f Represents the motor rotor flux; represents the differential of the speed sliding mode control law;

[0180] 223) Set the difference between the target motor mechanical angular velocity and the actual motor mechanical angular velocity As the speed sliding mode tracking error, the speed controller sliding mode surface is established:

[0181]

[0182] Where σ2 represents the first parameter of the speed sliding mode controller, and σ2>0; s ω represents the sliding surface of the speed controller; e ω Indicates the difference between the target motor mechanical angular velocity and the actual motor mechanical angular velocity; Indicates the differential between the target motor mechanical angular velocity and the actual motor mechanical angular velocity;

[0183] 224) The exponential approach function is selected as the speed sliding mode approach law:

[0184]

[0185] Where s ω represents the sliding surface of the speed controller; represents the differential of the sliding surface of the speed controller; k represents the second parameter of the speed sliding controller, and k>0; d represents the third parameter of the speed sliding controller, and d>0; sgn(s ω ) represents the sign function related to the rotation speed;

[0186] 225) By combining the speed controller sliding mode surface, the speed sliding mode reaching law and the speed system state variable equation, the motor torque shaft feedback current is obtained as:

[0187]

[0188] Where, Indicates the motor torque shaft feedback current; u s represents the speed sliding mode control law; k represents the second parameter of the speed sliding mode controller, and k>0; Indicates the differential between the target motor mechanical angular velocity and the actual motor mechanical angular velocity;

[0189] 226) Considering that the nonlinear friction of the mechanism affects the pressure control accuracy of the electronic power-assisted braking system, the motor torque axis feedforward compensation current based on the friction model is introduced into the position controller:

[0190]

[0191] Where K t Indicates the motor torque coefficient; T f represents friction torque; Indicates the motor torque axis feedforward compensation current;

[0192] 227) Finally, the output target motor torque shaft current is:

[0193]

[0194] Where, Indicates the motor torque axis feedforward compensation current; Indicates the motor torque axis feedback current; Indicates the target motor torque shaft current.

[0195] The specific method for designing a current controller based on Lyapunov theory is as follows:

[0196] 231) Establish the current tracking error as:

[0197]

[0198]

[0199] Where i d Indicates the actual motor excitation shaft current; i qIndicates the actual motor excitation torque shaft current; Indicates the target motor excitation shaft current, and sets Indicates the target motor torque shaft current; e d Indicates the difference between the target motor excitation shaft current and the actual motor excitation shaft current; e q Indicates the difference between the target motor torque axis current and the actual motor torque axis current;

[0200] 232) The corresponding current tracking error differential is:

[0201]

[0202] Where, Indicates the differential between the target motor excitation shaft current and the actual motor excitation shaft current; Indicates the differential difference between the target motor torque shaft current and the actual motor torque shaft current; u d Indicates the motor excitation shaft voltage; u q Represents the motor torque shaft voltage; R is the motor stator resistance; L d Indicates the motor excitation shaft inductance; L q Represents the motor torque shaft inductance; ψ f Represents the motor rotor flux; ω e Indicates the motor electrical angular velocity;

[0203] 233) Define the Lyapunov function of the motor excitation axis current controller as The Lyapunov function of the motor torque axis current controller is defined as The differential form is:

[0204]

[0205] Where, represents the differential of the Lyapunov function of the motor excitation axis current controller; represents the differential of the Lyapunov function of the motor torque axis current controller;

[0206] 2234) According to Lyapunov direct method, the motor excitation axis current control law and the motor torque axis current control law are designed as follows:

[0207]

[0208] Where, Represents the motor excitation shaft current control law; Represents the motor torque axis current control law; γ1 represents the motor excitation axis control parameter, and setting γ1>-R / L d ; γ1 represents the motor torque axis control parameter, and setting γ2>-R / L q .

[0209] In summary, this embodiment provides an electronic power-assisted braking system hydraulic controller designed using an adaptive radial basis function neural network and robust sliding mode theory, an electronic power-assisted braking system position controller designed using friction feedforward compensation and sliding mode control theory, and an electronic power-assisted braking system current controller designed using Lyapunov theory, ultimately achieving adaptive pressure control of the electronic power-assisted braking system.

[0210] Example 2

[0211] This embodiment uses a simulation platform built on MATLAB / Simulink to test the adaptive pressure control method for the electronic power-assisted braking system designed by the present invention. A simulation experiment is carried out with a sinusoidal pressure of 2.5MPa offset, 2.5MPa amplitude and 1Hz frequency as the control target of the electronic power-assisted braking system. The test results are shown in the figure. Figure 3 As shown. Figure 3 It can be seen that under the patented control strategy, the master cylinder pressure tracking error of the electronic power-assisted braking system is small, and the steady-state error is controlled within 0.1 MPa. This demonstrates that the invention's comprehensive application of adaptive radial basis function neural networks, robust sliding mode theory, friction feedforward compensation, and current decoupling control effectively ensures the brake pressure control performance of the electronic power-assisted braking system.

[0212] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the scope of protection of the present invention is not limited to the specific details of the above embodiments. Within the technical concept of the present invention, any technician familiar with the technical field can make equivalent replacements or changes based on the technical solution and inventive concept of the present invention within the technical scope disclosed by the present invention. These simple variations all fall within the scope of protection of the present invention.

[0213] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any appropriate manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0214] In addition, the various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. An adaptive pressure control method for an electronic power-assisted braking system considering multi-dimensional nonlinear disturbances, characterized in that: The following steps are involved: Step 1: Establish key models of motor, hydraulic pressure and friction; Step 2: Design a hydraulic-position-current cascade controller to achieve adaptive pressure control of the electronic power-assisted braking system. The hydraulic controller uses an adaptive radial basis function neural network and robust sliding mode theory to address the time-varying uncertainty disturbance of the hydraulic system. The position controller introduces friction feedforward compensation and sliding mode control to solve the nonlinear friction problem of the transmission mechanism; the current controller uses Lyapunov theory to design a current decoupling solution to solve the dynamic coupling problem of the electromagnetic characteristics of the permanent magnet synchronous motor; The specific method of step one is as follows: 11) According to the torque balance relationship of the permanent magnet synchronous motor output shaft, the motor motion equation is established: Where, f represents the motor load torque; T e Represents the electromagnetic torque of the motor; Represents the motor's mechanical angular acceleration; J represents the motor's moment of inertia; 12) Establish hydraulic system model; Assuming that the brake pressures of the master cylinder and wheel cylinder are consistent, the hydraulic compensation term f(P) is introduced to characterize the time-varying uncertainty disturbance of the hydraulic system. The final hydraulic system model is: Where, Indicates the master cylinder push rod speed; A indicates the effective area of ​​the master cylinder piston; Indicates the actual pressure change rate of the brake master cylinder; P indicates the actual pressure of the brake master cylinder; K c Represents the brake fluid bulk modulus; V m Indicates the volume of brake fluid in the master cylinder; V w represents the volume of brake fluid in the wheel cylinder; f(P) represents the hydraulic compensation term; x represents the actual master cylinder push rod displacement; 13) Modeling static friction, Coulomb friction, and viscous friction: Where, T f represents the friction torque; ρ v represents the viscous friction coefficient; ω m Represents the motor mechanical angular velocity; ρ c represents the Coulomb friction factor; T c Represents the Coulomb friction torque when the system is unloaded; ε n represents the motor speed threshold; n represents the gear ratio; h represents the ball screw lead; T s Represents the static friction torque.

2. The method for adaptive pressure control of an electronic power-assisted braking system considering multi-dimensional nonlinear disturbances according to claim 1, characterized in that: The specific method of designing a hydraulic controller based on an adaptive radial basis function neural network is as follows: 211) Let the target master cylinder push rod speed be the hydraulic control law, that is, Convert the hydraulic system model into Where, α represents the first parameter of the hydraulic system model; β represents the second parameter of the hydraulic system model; u represents the hydraulic control law; 212) According to the target pressure P of the brake master cylinder * The difference between the actual pressure P of the brake master cylinder and the actual pressure P is p Establish the sliding surface of the hydraulic controller: s p =σ1e p +∫e p dt Where s p represents the hydraulic controller sliding surface; σ1 represents the first parameter of the hydraulic sliding mode controller, and σ1>0; e p Indicates the difference between the target pressure of the brake master cylinder and the actual pressure of the brake master cylinder; 213) Correspondingly, the differential of the sliding surface of the hydraulic controller is solved as follows: Where, represents the differential of the sliding surface of the hydraulic controller; Indicates the differential between the target pressure of the brake master cylinder and the actual pressure of the brake master cylinder; Indicates the target pressure differential of the brake master cylinder; 214) Use the exponential reaching law containing saturation function: Where c represents the second parameter of the hydraulic sliding mode controller, and c>0; q represents the third parameter of the hydraulic sliding mode controller, and q>0; sat(s p ) represents the inclusion function; ξ represents the saturation function threshold; 215) Substituting the exponential reaching law containing the saturation function into the differential of the sliding surface of the hydraulic controller, the hydraulic control law is simplified to: Where, sat(s p ) indicates the included function; 216) Use radial basis function neural network to approximate and fit it; take the actual pressure of the brake master cylinder as the network input, and the estimated value of the hydraulic compensation term As the network output; the radial basis neural network algorithm is designed as: Where, represents the estimated value of the hydraulic compensation term; Represents the estimated network weights; h(P) represents the Gaussian basis function network output; c j represents the center value of the Gaussian basis function of the jth neuron in the hidden layer of the network; b j is the width of the Gaussian basis function of the jth neuron in the hidden layer; 217) Ideal value of hydraulic compensation term f(P) * Expressed as: f(P) * =W *T h(P)+ε p Where f(P) * represents the ideal value of the hydraulic compensation term; W *T represents the ideal network weight; h(P) represents the Gaussian basis function network output; ε p represents the network approximation error, and ε p ≤|T|, T represents a bounded real number; 218) Set the difference between the ideal network weight and the estimated network weight to be get: Where, It represents the difference between the ideal hydraulic compensation term and the estimated hydraulic compensation term; Represents the difference between the ideal network weight and the estimated network weight; 219) Estimated value of hydraulic compensation Substitute into the hydraulic control law and establish a new hydraulic control law u p : Where u p represents the new hydraulic control law; represents the estimated value of the hydraulic compensation term; sat(s p ) indicates the included function; s p represents the sliding surface of the hydraulic controller; 2110) The new hydraulic control law u p Substituting the differential of the sliding surface of the hydraulic controller, we can simplify it to get: 2111) Design of hydraulic layer Lyapunov function The differential form is: Where, represents the differential of the Lyapunov function of the hydraulic layer; μ represents the coefficient of the Lyapunov function of the hydraulic layer; represents the estimated network weight differential; 2112) According to Lyapunov's direct method, the adaptive law of the radial basis neural network is designed as follows: 2113) Further simplify the differential form of the hydraulic layer Lyapunov function: Considering the network approximation error ε p ≤|T| is bounded, so we choose When , the hydraulic layer Lyapunov function can be differentiated Ensure the closed-loop robustness of the hydraulic control system; Finally, the target master cylinder push rod displacement is: x * =∫u p dt Where x * Indicates the target master cylinder push rod displacement; u p Represents the new hydraulic control law.

3. The method for adaptive pressure control of an electronic power-assisted braking system considering multi-dimensional nonlinear disturbances according to claim 1, characterized in that: The specific method for designing a position controller based on friction model compensation is as follows: 221) Use angle proportional-integral control to establish the target motor mechanical angular velocity Where, Indicates the target motor mechanical angular velocity; k p Indicates the position control proportional gain; k i Indicates the position control integral gain; x * Indicates the target master cylinder push rod displacement; 222) In the speed sliding mode control, the target torque axis current is selected as the control input, that is, Then the state variable equation of the speed system is established as: Where, represents the differential of the motor's mechanical angular acceleration; J represents the motor's moment of inertia; P n Indicates the number of motor poles; ψ f Represents the motor rotor flux; represents the differential of the speed sliding mode control law; 223) Set the difference between the target motor mechanical angular velocity and the actual motor mechanical angular velocity As the speed sliding mode tracking error, the speed controller sliding mode surface is established: Where σ2 represents the first parameter of the speed sliding mode controller, and σ2>0; s ω represents the sliding surface of the speed controller; e ω Indicates the difference between the target motor mechanical angular velocity and the actual motor mechanical angular velocity; Indicates the differential between the target motor mechanical angular velocity and the actual motor mechanical angular velocity; 224) The exponential approach function is selected as the speed sliding mode approach law: Where s ω represents the sliding surface of the speed controller; represents the differential of the sliding surface of the speed controller; k represents the second parameter of the speed sliding mode controller, and k>0; d represents the third parameter of the speed sliding mode controller, and d>0; sgn(s ω ) represents the sign function related to the rotation speed; 225) By combining the speed controller sliding mode surface, the speed sliding mode reaching law and the speed system state variable equation, the motor torque shaft feedback current is obtained as: Where, Indicates the motor torque shaft feedback current; u s represents the speed sliding mode control law; k represents the second parameter of the speed sliding mode controller, and k>0; Indicates the differential between the target motor mechanical angular velocity and the actual motor mechanical angular velocity; 226) The motor torque axis feedforward compensation current based on the friction model is introduced into the position controller: Where: K t Indicates the motor torque coefficient; T f represents friction torque; Indicates the motor torque axis feedforward compensation current; 227) Finally, the output target motor torque shaft current is: Where, Indicates the motor torque axis feedforward compensation current; Indicates the motor torque axis feedback current; Indicates the target motor torque shaft current.

4. The method for adaptive pressure control of an electronic power-assisted braking system considering multi-dimensional nonlinear disturbances according to claim 1, characterized in that: The specific method for designing a current controller based on Lyapunov theory is as follows: 231) Establish the current tracking error as: Where i d Indicates the actual motor excitation shaft current; i q Indicates the actual motor excitation torque shaft current; Indicates the target motor excitation shaft current, and sets Indicates the target motor torque shaft current; e d Indicates the difference between the target motor excitation shaft current and the actual motor excitation shaft current; e q Indicates the difference between the target motor torque axis current and the actual motor torque axis current; 232) The corresponding current tracking error differential is: Where, Indicates the differential between the target motor excitation shaft current and the actual motor excitation shaft current; Indicates the differential difference between the target motor torque shaft current and the actual motor torque shaft current; u d Indicates the motor excitation shaft voltage; u q Represents the motor torque shaft voltage; R is the motor stator resistance; L d Indicates the motor excitation shaft inductance; L q Represents the motor torque shaft inductance; ψ f Represents the motor rotor flux; ω e Represents the motor electrical angular velocity; ω m Indicates the mechanical angular velocity of the motor; 233) Define the Lyapunov function of the motor excitation axis current controller as The Lyapunov function of the motor torque axis current controller is defined as The differential form is: Where, represents the differential of the Lyapunov function of the motor excitation axis current controller; represents the differential of the Lyapunov function of the motor torque axis current controller; 234) According to Lyapunov direct method, the motor excitation axis current control law and the motor torque axis current control law are designed as follows: Where, Represents the motor excitation shaft current control law; Represents the motor torque axis current control law; γ1 represents the motor excitation axis control parameter, and setting γ1>-R / L d ; γ1 represents the motor torque axis control parameter, and setting γ2>-R / L q .

Images