Parameterization multi-objective performance optimization design method and system for vehicle active suspension system

Abstract

The invention discloses a parameterized multi-objective performance optimization design method and system for a vehicle active suspension system, and the method comprises the steps: firstly, carrying out the parameter identification of a controlled system, obtaining a mathematical model of the vehicle active suspension system, and constructing a vehicle active suspension system control frame based on a generalized equivalent input interference GEID method; then, a closed-loop state space equation of the vehicle active suspension control system is derived through the control framework, parameterized representation is carried out on parameters to be designed and control targets, and a control problem is converted into a multi-target optimization problem related to controller parameters; and finally, designing an efficient multi-objective optimization algorithm, and comprehensively optimizing unmeasurable performance output estimation, disturbance suppression, noise suppression, input constraint and robustness performance of the control system to obtain optimal controller parameters. According to the method, the robustness and the stability of the system can be better guaranteed under the complex working condition, and coordination and balance between performances are fully considered during control parameter design.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobile dynamics control, and in particular relates to a parametric multi-objective performance optimization design method and system for a vehicle active suspension system. Background Art

[0002] The vehicle suspension system is an important component that affects the ride comfort, handling stability and safety of the vehicle. Due to its fixed structural parameters, the traditional passive suspension system is difficult to meet multiple performance requirements under complex and changeable driving conditions. In recent years, active suspension systems have attracted widespread attention by dynamically adjusting suspension characteristics through controllers to improve vehicle performance. However, there are many challenges in the control design of active suspension systems. First, there are many unknown disturbances during vehicle operation, such as disturbances caused by uneven road surfaces and measurement noise, which have a significant impact on the performance of the control system. Second, due to the complexity of the vehicle system and the variability of the operating environment, problems such as parameter disturbances, modeling errors and linearization errors are prevalent, making it difficult to ensure the robustness of the system design. Third, most current control design methods usually focus on meeting stability conditions, but rarely consider the overall optimization of system performance, such as vehicle acceleration estimation, disturbance suppression, noise suppression, input constraints and robustness improvement. In order to deal with the above problems, control strategies based on the equivalent input disturbance (EID) method have gradually attracted attention. The EID method can effectively deal with matched and unmatched disturbances and simplify system design by estimating and compensating for equivalent input disturbances on the input channel. However, the existing EID control design is mostly based on numerical simulation for parameter adjustment, which fails to systematically optimize multiple performance objectives and is difficult to adapt to complex practical application scenarios. Therefore, how to design a performance-oriented parametric multi-objective optimization method based on both computational complexity and practical feasibility to comprehensively optimize the performance of the vehicle active suspension system has become an urgent problem to be solved in the current technical field. Summary of the invention

[0003] In view of the problems of insufficient performance optimization, limited anti-disturbance capability and high complexity of parametric design in the design of vehicle active suspension system in the prior art, the present invention proposes a performance-oriented parametric multi-objective optimization design method for vehicle active suspension system. Through this method, comprehensive optimization of multiple performance targets of active suspension system can be achieved, including vehicle acceleration estimation, disturbance suppression, noise suppression, input optimization and robustness design, thereby improving the ride comfort and handling stability of the vehicle. Compared with the existing methods, the main features of the present invention are as follows: a method based on equivalent input disturbance is adopted, and the unmeasurable output estimation, disturbance suppression, noise suppression, input constraint and robustness performance of active suspension system are comprehensively optimized through multi-objective optimization strategy. Clear parametric design conditions are proposed, and complex performance optimization problems are converted into specific controller parameter design tasks, which significantly reduces the design difficulty. At the same time, the present invention provides an efficient multi-objective optimization algorithm, which realizes the comprehensive optimization of various performance targets through optimization solution, and ensures that the control parameters meet the performance requirements under complex working conditions.

[0004] The technical solution adopted by the present invention is a parametric multi-objective performance optimization design method and system for a vehicle active suspension system, comprising the following steps:

[0005] Step 1: Perform parameter identification on the controlled system to obtain a mathematical model of the vehicle active suspension system; construct a vehicle active suspension system control framework based on the general equivalent input disturbance (GEID) method, wherein the control framework includes a mathematical model of the controlled object, a state observer, a state feedback controller and a GEID estimator.

[0006] Step 2: Based on the control framework described in step 1, an equivalent description diagram and performance-oriented multi-performance control objectives are given, the closed-loop state space equation of the vehicle active suspension control system is derived, and then a complete parameterized representation of the controller parameters to be designed is given.

[0007] Step 3: Based on the complete parameterized representation given in step 2, each control objective is converted into a parameterized form related to the controller parameters, wherein the control objectives include: vehicle acceleration estimation performance, disturbance suppression performance, noise suppression performance, input constraint performance and robust performance, and the control problem is converted into an optimization problem related to the controller parameters. Furthermore, the performance index weights are set to reduce the mutual coupling between the various performances and reduce the computational complexity. Finally, the control problem is converted into a multi-objective optimization problem.

[0008] Step 4: Provide an efficient multi-objective optimization algorithm to optimize and solve various performance objectives, comprehensively optimize the performance of the control system, and obtain the optimal controller parameters to ensure that the control parameters meet all performance requirements under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a flow chart of the present invention;

[0010] Figure 2 is a structural block diagram of an automobile active suspension control system according to an embodiment of the present invention;

[0011] Figure 3 is an equivalent structural block diagram of the system structural block diagram of an embodiment of the present invention;

[0012] Figure 4 is a flow chart of an optimization algorithm for an automobile active suspension system according to an embodiment of the present invention; DETAILED DESCRIPTION

[0013] Specifically, the design method process of the present invention is as follows: Figure 1 As shown, the following steps are included:

[0014] Step 1: Perform parameter identification on the controlled system to obtain the mathematical model of the 1 / 4 vehicle active suspension system.

[0015] In this implementation case, the following system state space equations are obtained through parameter identification:

[0016]

[0017] in, and are the state variables of the system, the control input, the unknown external disturbance of the system, the unmeasurable performance output, the measurable output, and the measurement noise, A, B, B d , C, D and C y is a real matrix of appropriate dimension.

[0018] At the same time, assume that the system (A, B, C y ) is controllable and observable.

[0019] Build as Figure 2 The structural block diagram of the 1 / 4 vehicle active suspension control system shown in FIG. 1 is composed of four parts: a mathematical model of the controlled object, a state observer, a state feedback controller and a generalized equivalent input disturbance estimator.

[0020] First, since the state information of the vehicle active suspension system is difficult to measure directly or the measurement cost is high, the Lumberg state observer is used to estimate the system state without disturbance:

[0021]

[0022] in, is the observer state, u f (t), and are the input, estimated performance output and estimated measurement output, and L is the observer gain to be determined.

[0023] Secondly, a state feedback controller is designed to stabilize the system. The state feedback controller can be expressed as:

[0024]

[0025] Among them, K p is the parameter to be determined for the feedback controller.

[0026] Then, in order to eliminate the influence of external disturbances on the performance of the control system, a generalized equivalent input disturbance estimator is constructed in the inner loop to estimate the equivalent input disturbance d corresponding to the actual mismatch disturbance d(t) to be compensated. e (t).

[0027] definition

[0028]

[0029] One estimate of the equivalent input disturbance can be expressed as:

[0030]

[0031] Among them, K e is the desired control gain of the estimator.

[0032] Given a filter F(s) to solve The state space of the autocorrelation problem can be expressed as:

[0033]

[0034] in, is a state variable, yes The filtered value.

[0035] definition

[0036] A e =A q +B q C q , B e =B q K e , C e =C q (7).

[0037] According to the state space equation of the filter F(s), the state space equation of the generalized equivalent input disturbance estimator can be expressed as:

[0038]

[0039] in, is the estimator state.

[0040] Finally, the control rate including disturbance estimation and compensation can be obtained as:

[0041]

[0042] So far we have completed step one and established Figure 2 The control framework of the vehicle active suspension system based on the generalized equivalent input disturbance estimator is shown in Fig. 2. For (2), it is to observe the state of the system without external disturbance d(t) in order to obtain the output error The interference information in (8) is written as a state space expression to facilitate the subsequent step 2. Figure 2 Derivation of equivalent description diagram and state space expression of closed-loop system.

[0043] Step 2: Based on the control framework obtained in step 1, we can give Figure 2 The equivalent description of the control framework is shown in the figure Figure 3 shown.

[0044] according to Figure 3 From the equivalent description diagram, we can see that the control system consists of subsystem 1 and subsystem 2. Therefore, the control system gain (K e ,L,K p ) can be separated into the design of subsystem 1 and K in subsystem 2 p The control system is designed to meet the following control objectives:

[0045] 1) Stability: The poles of the closed-loop system are configured to the desired positions.

[0046] 2) Estimation performance of unmeasured performance output: In the presence of disturbance d(t), the estimated output of the observer It can approximate the unmeasurable performance output z(t) very well.

[0047] 3) Disturbance suppression performance: The impact of disturbance d(t) on system output z(t) is as small as possible.

[0048] 4) Noise suppression performance: The impact of measurement noise n(t) on control input u(t) is as small as possible.

[0049] 5) Control input constraints: To avoid input commands causing actuator saturation or pulse inputs having adverse effects on the system, constraints should be imposed on the control inputs.

[0050] 6) Robustness to parameter disturbances: The poles of the closed-loop system should be as insensitive to parameter disturbances as possible.

[0051] Thus, the equivalent control structure description diagram and various performance-oriented control objectives are obtained.

[0052] Below, in order to facilitate calculation, the state space descriptions of the two subsystems are given respectively, and the state space description equations of the closed-loop control system are established.

[0053] According to (1) and (2), we can get the error system as:

[0054]

[0055] Combining (10) and (8), the state space expression of subsystem 1 can be expressed as:

[0056]

[0057] in,

[0058] According to (2) and (9), the state space expression of subsystem 2 can be expressed as:

[0059]

[0060] definition Combining (2), (8), and (10), the state space representation of the closed-loop system can be expressed as:

[0061]

[0062] in,

[0063] Next, the design parameters of subsystem 1 are Parameterized representation: First, the system matrix of subsystem 1 is The eigenvalue diagonal matrix of is expressed as:

[0064]

[0065] when When controllable, there exists a generalized left eigenvalue matrix Make right Perform right coprime decomposition so that:

[0066]

[0067] in,

[0068] because Available

[0069]

[0070] in

[0071] Finally, the design parameters of subsystem 1 are The complete parameterization of can be expressed as:

[0072]

[0073] is a free variable, satisfying

[0074] Similarly, we have the design parameter K of subsystem 2 p Parameterized representation: First, the system matrix K of subsystem 2 is p The eigenvalue diagonal matrix of is expressed as:

[0075]

[0076] Since (A,B) is controllable, there exists a generalized left eigenvalue matrix Make A k V=VΛ k . For (sI-A) -1 B is right coprime decomposed so that:

[0077] (sI-A)N(s)=BΓ(s) (21)

[0078] in,

[0079] Because A k =A+BK p , can be obtained

[0080] AV+BW=VΛ k (twenty two)

[0081] Where W = K p V.

[0082] Finally, the design parameters of subsystem 1 are The complete parameterization of can be expressed as:

[0083]

[0084] is a free variable, satisfying

[0085] At this point, a complete parameterized representation of the controller parameters to be designed is obtained, completing all steps in step 2.

[0086] Step 3: Using the complete parameterized representation of the controller parameters to be designed obtained in step 2, various control performance requirements can be transformed into the following optimization problems related to controller parameters:

[0086] In order to ensure that the entire closed-loop system is stable, the desired poles of the two subsystems need to be and Both lie in the left half of the complex plane.

[0087] The optimization of the estimated performance of the closed-loop system can be solved by the following minimization problem:

[0088]

[0089] in,

[0090] The optimization of the disturbance rejection performance of the closed-loop system can be solved by the following minimization problem:

[0091]

[0092] in,

[0093] The optimization of the noise suppression performance of the closed-loop system can be transformed into the following minimization problem to solve:

[0094]

[0095] in,

[0096] The optimization of the closed-loop system control input performance can be transformed into the following minimization problem to solve:

[0097]

[0098] in,

[0099] The optimization of the robust performance of the closed-loop system can be transformed into the following minimization problem:

[0100]

[0101] in,

[0102] Finally, all performance-oriented control objectives obtained in step 2 are transformed into multi-objective optimization problems related to controller parameters.

[0103] Furthermore, the weight parameters of each indicator are set as: α z ,α d ,α n ,α u and α s .

[0104] By using the weight parameters, the above multi-objective optimization problem is transformed into a single-objective optimization problem, and we can get:

[0105]

[0106] At this point, all the contents of step 3 are completed.

[0107] Step 4: Provide an efficient multi-objective optimization algorithm to comprehensively optimize the performance of the control system and obtain the optimal controller parameters to ensure that the control parameters meet all performance requirements under complex working conditions.

[0108] Finally, the multi-objective design algorithm can be expressed as:

[0109] Step 1: Select the filter F(s) by the disturbance suppression frequency range so that Controllable; according to the expected performance, select the expected closed-loop system pole range S d and S k .

[0110] Step 2: According to (17), calculate the right coprime decomposition matrix H(s), R(s); according to (21), calculate N(s), Γ(s); then establish the gain and K p The complete parameterized representation of (19) and (23).

[0111] Step 3: Choose the appropriate weight α z ,α d ,α n ,α u and α s , transforming the multi-objective optimization problem into a single-objective optimization problem of (29).

[0112] Step 4: Solve the optimization problem and obtain the optimal solution

[0113] Step 5: Calculate through (19) and (23) and Finally, according to (7) and (12), the optimal controller parameter L is obtained * ,

[0114] Step 6: Check whether the closed-loop system is stable and whether its performance meets the requirements. If so, stop; if not, return to the first step and redesign / select the pole range.

[0115] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A parametric multi-objective performance optimization design method for a vehicle active suspension system, characterized in that: The following steps are involved: Step 1: perform parameter identification on the controlled system to obtain a mathematical model of the vehicle active suspension system; construct a vehicle active suspension system control framework based on the generalized equivalent input disturbance GEID method, wherein the control framework includes a mathematical model of the controlled object, a state observer, a state feedback controller and a GEID estimator; Step 2: Based on the control framework described in step 1, an equivalent description diagram and a performance-oriented multi-performance control target are given, and a closed-loop state space equation of the vehicle active suspension control system is derived, and then a complete parameterized representation of the controller parameters to be designed is given; Step 3: Based on the complete parameterized representation given in step 2, each control objective is converted into a parameterized form related to controller parameters, wherein the control objectives include: vehicle acceleration estimation performance, disturbance suppression performance, noise suppression performance, input constraint performance and robust performance, and the control problem is converted into a multi-objective optimization problem related to controller parameters; Set the weights of performance indicators to reduce the mutual coupling between various performances and reduce the computational complexity; ultimately, transform the control problem into a multi-objective optimization problem; Step 4: Provide an efficient multi-objective optimization algorithm to optimize and solve various performance objectives, comprehensively optimize the performance of the control system, and obtain the optimal controller parameters to ensure that the control parameters meet all performance requirements under complex working conditions.

2. The parametric multi-objective performance optimization design method for a vehicle active suspension system according to claim 1, characterized in that: In step two, the equivalent description diagram of the vehicle active suspension system control framework and the complete parameterized representation of the controller parameters to be designed are obtained: the equivalent description diagram separates the closed-loop control system into two subsystems, so that the controller parameters can be designed separately, reducing the complexity of control system design; at the same time, the complete parameterized representation of the controller parameters to be designed links the controller parameters to be designed with the control system poles, and introduces free variables to provide the possibility for controller parameter optimization.

3. The parametric multi-objective performance optimization design method for a vehicle active suspension system according to claim 1, characterized in that: In step three, each control objective is converted into a parameterized form related to the controller parameters: multiple control objectives are converted into a parameterized form related to the controller parameters, a connection is established between the control performance of the control system and the controller parameters, and the control problem is converted into a multi-objective optimization problem related to the controller parameters, so that the controller can provide the possibility for comprehensive performance optimization of the vehicle active suspension system; and by giving performance indicator weights, the computational complexity of the optimization problem is reduced, providing a guarantee for the actual feasibility of the control system.

4. The parametric multi-objective performance optimization design method for a vehicle active suspension system according to claim 1, characterized in that: In step 4, an efficient multi-objective optimization algorithm is proposed to comprehensively optimize the performance of the control system and obtain the optimal controller parameters to ensure that the control parameters meet all performance requirements under complex working conditions.

Images