阅读说明：本技术 一种基于模型参考自适应控制的车辆isd悬架主动控制方法 (Vehicle ISD suspension active control method based on model reference adaptive control ) 是由 杨晓峰 颜龙 沈钰杰 刘雁玲 刘昌宁 杨艺 宋航 何涛 于 2020-03-27 设计创作，主要内容包括：本发明公开了一种基于模型参考自适应控制的车辆ISD悬架主动控制方法,该方法包含如下步骤：(1)建立悬架被控模型与参考模型；(2)定义被控模型与理想参考模型广义状态误差e；(3)改写系数矩阵；(4)自适应控制系统稳定性判定；(5)确定自适应律。本发明能使结构简单的被控ISD悬架系统跟踪复杂的理想ISD悬架模型的输出,提高悬架的综合性能。该方法为车辆ISD悬架设计及应用提供了一种新的研究思路与方向,尤其是对于可控ISD悬架的研究。(The invention discloses a vehicle ISD suspension active control method based on model reference adaptive control, which comprises the following steps: (1) establishing a suspension controlled model and a reference model; (2) defining a generalized state error e between a controlled model and an ideal reference model; (3) rewriting the coefficient matrix; (4) judging the stability of the self-adaptive control system; (5) an adaptation law is determined. The invention can lead the controlled ISD suspension system with simple structure to track the output of the complex ideal ISD suspension model and improve the comprehensive performance of the suspension. The method provides a new research idea and direction for the design and application of the vehicle ISD suspension, in particular to the research on the controllable ISD suspension.)

1. A vehicle ISD suspension active control method based on model reference adaptive control comprises the following steps: step 1): establishing a controlled model and an ideal reference model of an ISD suspension of the vehicle based on model reference adaptive control; step 2): defining a control system generalized state error e related to a controlled model and an ideal reference model; step 3): rewriting a coefficient matrix related to a control system state error; step 4): judging the stability of the self-adaptive control system; step 5): and determining the adaptive law of the controlled model of the vehicle ISD suspension in the model reference adaptive control system.

2. The method for actively controlling the ISD suspension of the vehicle based on the model-reference adaptive control as claimed in claim 1, wherein the step 1) is specifically as follows: the kinematic equation of the controlled model is:

wherein m is_sIs sprung mass, m_uIs unsprung mass, k is suspension spring rate, k_tIs equivalent stiffness of the tire, b is inertia coefficient of the inertia container, F_bControlling force, z, for the inerter of the controlled model_rIs the vertical input displacement of the road surface, z_uFor vertical displacement of unsprung mass of controlled model, z_sFor the vertical displacement of the sprung mass of the controlled model,the sprung mass vertical velocity of the controlled model,is the sprung mass vertical acceleration of the controlled model,for unsprung mass vertical velocity of the controlled model,the unsprung mass vertical acceleration of the controlled model;

the kinematic equation for the ideal reference model is:

wherein m is_sIs sprung mass, m_uIs an unsprung mass, k_tIs the equivalent stiffness of the tire, k_rFor the suspension spring rate of the reference model, T(s) is the equivalent impedance of the specific structure, c_pIs a semi-active damping coefficient, F_pAdjustable damper control force for reference model, z_rIs the vertical input displacement of the road surface, z_urVertical displacement of unsprung mass for reference model，z_srFor vertical displacement of the sprung mass of the reference model,to refer to the sprung mass vertical velocity of the model,to refer to the sprung mass vertical acceleration of the model,to refer to the unsprung mass vertical velocity of the model,is the unsprung mass vertical acceleration of the reference model.

3. The method for actively controlling the ISD suspension of the vehicle based on the model-reference adaptive control as claimed in claim 2, wherein the step 2) is specifically as follows: rewriting the state variables of the controlled model and the ideal reference model in the step 1), selecting the speed and the displacement of the sprung mass as state variables, selecting the speed and the displacement of the unsprung mass as input variables, and writing the state equations of the controlled model and the ideal reference model into the following equations:

wherein, F_bAnd F_pControl forces, X, of the controlled model and the reference model, respectively_bAnd X_pRespectively state variables of the controlled model and the reference model,andderivatives, y, of the state variables of the controlled model and the reference model, respectively_rFor controlling system variables, A_b，A_p，B_b，B_p，C_b，C_pIs a coefficient matrix;

wherein the control force F of the controlled model_bComprises the following steps:

F_b＝K₁X_b+K₂F_p+K₃y_r，

K₁feedback regulator for controlled model, K₂Control vector gain, K, for an ideal reference model₃Is a feed-forward regulator;

wherein the generalized state error e and its derivativesThe error equation is satisfied:

e＝X_p-X_b，

4. the method for actively controlling the ISD suspension of the vehicle based on the model-reference adaptive control method as claimed in claim 3, wherein the step 3) is specifically as follows: rewriting a matrix of coefficients a relating to control system state errors_p、B_p、C_p：

By adjusting the feedback regulator K₁Control vector gain K of reference model₂And a feedforward regulator K₃Setting the generalized state error e to zero, rewriting the coefficient matrix A_p、B_p、C_pComprises the following steps:

wherein, K₁ ^*、K₂ ^*And K₃ ^*Respectively representing K when the controlled model is consistent with the reference model₁、K₂And K₃Is taken as the steady state value.

5. The method for actively controlling the ISD suspension of the vehicle based on the model-reference adaptive control method as claimed in claim 4, wherein the step 4) is specifically as follows: and (3) judging the stability of the self-adaptive control system:

selecting a corresponding Lyapunov function V:

wherein, P ∈ R^4×4、R₁、R₂And R₃All represent positive definite symmetric matrices; tr represents the trace of the matrix; ^Trepresenting a transpose of the respective matrix;

the method can be obtained according to the fact that the trace of the unit element matrix is equal to the element value of the unit element matrix and the property of the matrix trace:

due to A_pFor the stabilization of the matrix, there is a positive definite symmetric matrix Q such that A_p ^TP+PA_pThe derivative of the lyapunov function V is derived and substituted into the error equation to yield:

wherein the content of the first and second substances,andare respectively asAndthe derivative of the transposed matrix is then inverted, are respectively asAndthe derivative of (c).

6. The method for actively controlling the ISD suspension of the vehicle based on the model-reference adaptive control method as claimed in claim 5, wherein the step 5) is specifically as follows: determining an adaptive law:

wherein, K₁(t)、K₂(t) and K₃(t) are each K₁、K₂And K₃A function with respect to time t;K₁(0)、K₂(0) and K₃(0) Are each K₁、K₂And K₃A value at time t ═ 0;

the control force of the controlled model is rewritten as:

7. the model-reference-based adaptive control of claim 2The active control method of the vehicle ISD suspension is characterized by further comprising the vertical input speed of the road surface unevennessSatisfies the following conditions:

wherein the content of the first and second substances,vertical input speed of road surface unevenness, v running speed, w (t) white noise signal, G_q(n₀) Is the road surface unevenness coefficient.

Technical Field

The invention belongs to the field of vehicle suspension system control, and particularly relates to vehicle ISD (inertial-Spring-Damper) suspension system control applying an Inerter-Spring device. The invention relates to a control method of an ISD suspension of a vehicle, in particular to a control method of the ISD suspension of the vehicle by applying an electromechanical inertial container.

Background

The concept of inerter was proposed in 2002 by professor Smith of cambridge university. As a mass element with two end points, the inertial container breaks through the limitation of grounding of the mass element with the single end point, makes up the vacancy that the mass corresponds to the capacitor in the electromechanical similarity theory, and can be effectively applied to the design of a vibration isolation system. The ISD suspension of the vehicle is a novel suspension vibration isolation system consisting of an Inerter-Spring-Damper (Inerter-Spring-Damper). The vehicle ISD suspension breaks the constraint of the traditional two-element parallel frame on the vibration isolation performance of the suspension, so that the research of the vehicle suspension is advanced into a new field.

In recent years, controllable ISD suspensions have become an increasingly dominant direction in suspension development and research. Chinese patent CN109334378A discloses a vehicle ISD suspension active control method based on single neuron PID control, which utilizes the learning characteristic of single neuron to improve the problem that the traditional PID control is difficult to deal with nonlinearity and improve the performance of ISD suspension. But the control effect of the method depends on the learning rule and the learning rate, and has higher requirements on the experience of engineering personnel.

The adaptive control system has the characteristics of wire accumulation information, controller adjustability and automatic adaptability. The controlled system reduces the self-existing uncertainty through direct or indirect accumulation of auxiliary signals and periodic identification, and the uncertainty determines the adjustability of the controller, and the controlled system can automatically adjust through changing parameters according to the requirements of performance indexes, so that the controlled system has the self-adaptive capacity for the change of internal and external environments. The reference model in the model reference adaptive control can be an ideal model and is not necessarily a practical and feasible practical system, and the excellent control effect can be achieved only by being similar to the controlled model and enabling the spatial dimensions to be the same.

In the design of suspension control systems, a variety of control methods are used. The prediction control can feed forward the state of the front road surface as a prediction variable, take the future target value or interference of the system into consideration, and reduce the control energy peak value and the control energy consumption of the system, but the calculation amount is large, and the method is not favorable for the online implementation of a fast time-varying system. Aiming at the uncertainty of a suspension control system, control methods such as neural network control and fuzzy control can better deal with complex nonlinear environments, but the neural network control learning speed is slow and the difficulty is high, and the decision speed and the real-time control speed of the fuzzy control are relatively unstable. Adaptive control is a method for adjusting a controller in real time, and can automatically monitor system parameters, so that the system has good performance. By adopting the self-adaptive control, the influence of the random disturbance of the model on the system can be effectively solved, and the dynamic performance of the whole vehicle is improved.

Disclosure of Invention

Based on the reasons, the invention provides the vehicle ISD suspension active control method based on model reference adaptive control, so that the controlled ISD suspension model can track the output of an ideal model, and the control effect and the comprehensive performance of the controllable ISD suspension are effectively improved.

In order to achieve the purpose, the technical scheme adopted by the scheme is as follows: a vehicle ISD suspension active control method based on model reference adaptive control comprises the following steps: step 1): establishing a controlled model and an ideal reference model of an ISD suspension of the vehicle based on model reference adaptive control; step 2): defining a control system generalized state error e related to a controlled model and an ideal reference model; step 3): rewriting a coefficient matrix related to a control system state error; step 4): judging the stability of the self-adaptive control system; step 5): and determining the adaptive law of the controlled model of the vehicle ISD suspension in the model reference adaptive control system.

Further, the step 1) is specifically as follows: the kinematic equation of the controlled model is:

wherein m is_sIs sprung mass, m_uIs unsprung mass, k is suspension spring rate, k_tIs equivalent stiffness of the tire, b is inertia coefficient of the inertia container, F_bControlling force, z, for the inerter of the controlled model_rIs vertical to the road surfaceInput displacement, z_uFor vertical displacement of unsprung mass of controlled model, z_sFor the vertical displacement of the sprung mass of the controlled model,the sprung mass vertical velocity of the controlled model,is the sprung mass vertical acceleration of the controlled model,for unsprung mass vertical velocity of the controlled model,the unsprung mass vertical acceleration of the controlled model;

the kinematic equation for the ideal reference model is:

wherein m is_sIs sprung mass, m_uIs an unsprung mass, k_tIs the equivalent stiffness of the tire, k_rFor the suspension spring rate of the reference model, T(s) is the equivalent impedance of the specific structure, c_pIs a semi-active damping coefficient, F_pAdjustable damper control force for reference model, z_rIs the vertical input displacement of the road surface, z_urFor unsprung mass vertical displacement of the reference model, z_srFor vertical displacement of the sprung mass of the reference model,to refer to the sprung mass vertical velocity of the model,to refer to the sprung mass vertical acceleration of the model,to refer to the unsprung mass vertical velocity of the model,is the unsprung mass vertical acceleration of the reference model.

Further, the step 2) is specifically as follows: rewriting the state variables of the controlled model and the ideal reference model in the step 1), selecting the speed and the displacement of the sprung mass as state variables, selecting the speed and the displacement of the unsprung mass as input variables, and writing the state equations of the controlled model and the ideal reference model into the following equations:

wherein, F_bAnd Fp is the control force, X, of the controlled model and the reference model, respectively_bAnd Xp are the state variables of the controlled model and the reference model respectively,andderivatives, y, of the state variables of the controlled model and the reference model, respectively_rFor controlling system variables, A_b，A_p，B_b，B_p，C_b，C_pIs a coefficient matrix;

wherein the control force F of the controlled model_bComprises the following steps:

F_b＝K₁X_b+K₂F_p+K₃y_r，

K₁feedback regulator for controlled model, K₂Control vector gain, K, for an ideal reference model₃Is a feed-forward regulator;

wherein the generalized state error e and its derivativesThe error equation is satisfied:

e＝X_p-X_b，

further, the step 3) is specifically as follows: rewriting coefficient matrix A_p、B_p、C_p：

By adjusting the feedback regulator K₁Control vector gain K of reference model₂And a feedforward regulator K₃Setting the generalized state error e to zero, rewriting the coefficient matrix A_p、B_p、C_pComprises the following steps:

wherein, K₁ ^*、K₂ ^*And K₃ ^*Respectively representing K when the controlled model is consistent with the reference model₁、K₂And K₃Is taken as the steady state value.

Further, the step 4) is specifically as follows: and (3) judging the stability of the self-adaptive control system:

selecting a corresponding Lyapunov function V:

wherein, P ∈ R^4×4、R₁、R₂And R₃All represent positive definite symmetric matrices; tr represents the trace of the matrix; ^Trepresenting a transpose of the respective matrix;

the method can be obtained according to the fact that the trace of the unit element matrix is equal to the element value of the unit element matrix and the property of the matrix trace:

due to A_pFor the stabilization of the matrix, there is a positive definite symmetric matrix Q such that A_p ^TP+PA_pThe derivative of the lyapunov function V is derived and substituted into the error equation to yield:

wherein the content of the first and second substances,andare respectively asAndthe derivative of the matrix is transposed.

Further, the step 5) is specifically as follows: determining an adaptive law:

wherein, K₁(t)、K₂(t) and K₃(t) are each K₁、K₂And K₃A function with respect to time t;K₁(0)、K₂(0) and K₃(0) Are each K₁、K₂And K₃A value at time t ═ 0;

the control force of the controlled model is rewritten as:

further, the vertical input speed of the road surface unevenness is includedSatisfies the following conditions:

wherein the content of the first and second substances,vertical input speed of road surface unevenness, v running speed, w (t) white noise signal, G_q(n₀) Is the road surface unevenness coefficient.

The invention has the beneficial effects that: the invention relates to a vehicle ISD suspension active control method based on model reference adaptive control, which selects the speed and displacement of an unsprung mass as input signals, calculates an error kinetic equation of a controlled model and a reference model, and realizes the effect of tracking an ideal model and outputting the ideal model by a controlled suspension through a designed adaptive control law. The invention can effectively improve the comprehensive performance of the ISD suspension and the comprehensive vibration isolation performance of the suspension system, has strong robustness and simple physical realization, and is beneficial to engineering application. The method provides a new research idea for the research of the controllable ISD suspension of the vehicle.

Drawings

FIG. 1 is a flow chart of a method for actively controlling an ISD suspension of a vehicle based on model-reference adaptive control;

FIG. 2 is a diagram of an ISD suspension configuration in the example;

wherein (a) is a controlled suspension diagram and (b) is an ideal reference suspension diagram;

FIG. 3 is a model diagram of an ideal suspension structure;

FIG. 4 is a schematic diagram of model reference adaptive control;

fig. 5 is a time domain comparison graph of suspension performance obtained by the method, wherein (a) is a vehicle body acceleration response graph, (b) is a suspension dynamic stroke response graph, and (c) is a tire dynamic load response graph.

FIG. 6 is a frequency domain comparison graph of suspension performance obtained by the method, wherein (a) is a vehicle body acceleration response graph, (b) is a suspension dynamic stroke response graph, and (c) is a tire dynamic load response graph.

Detailed Description

The invention will be further described with reference to the drawings and the specific examples, but the scope of the invention is not limited thereto.

Fig. 1 is a flow chart of the method for actively controlling the ISD suspension of the vehicle based on model reference adaptive control, fig. 2 is a diagram of an equivalent suspension model in the example of the method, and the inerter device in fig. 2 is preferably an electromechanical inerter or a hydroelectric coupling type vehicle suspension impedance control device disclosed in chinese patent CN 204526713U.

Referring to fig. 1, the method for actively controlling the ISD suspension of the vehicle based on the model-reference adaptive control of the present invention includes: step 1): establishing a controlled model and an ideal reference model of an ISD suspension of the vehicle based on model reference adaptive control; step 2): defining a control system generalized state error e related to a controlled model and an ideal reference model; step 3): rewriting a coefficient matrix related to a control system state error; step 4): judging the stability of the self-adaptive control system; step 5): determining the self-adaptive law of a controlled model of an ISD suspension of a vehicle in a model reference self-adaptive control system;

wherein, the step 1) is specifically as follows: establishing kinematic equations of a controlled model and an ideal reference model according to the suspension model shown in FIG. 2;

the kinematic equation of the controlled model is as follows:

wherein m is_sIs sprung mass, m_uIs unsprung mass, k is suspension spring rate, k_tIs equivalent stiffness of the tire, b is inertia coefficient of the inertia container, F_bControlling force, z, for the inerter of the controlled model_rIs the vertical input displacement of the road surface, z_uFor vertical displacement of unsprung mass of controlled model, z_sFor the vertical displacement of the sprung mass of the controlled model,the sprung mass vertical velocity of the controlled model,is the sprung mass vertical acceleration of the controlled model,for unsprung mass vertical velocity of the controlled model,the unsprung mass vertical acceleration of the controlled model;

further, the air conditioner is provided with a fan,(z_s-z_u)、k_t(z_u-z_r) The vertical acceleration of the sprung mass, the suspension dynamic travel and the tire dynamic load are three dynamic performance indexes of the suspension system;

the kinematic equation of the ideal reference model is as follows:

wherein m is_sIs sprung mass, m_uIs an unsprung mass, k_tIs the equivalent stiffness of the tire, k_rFor the suspension spring rate of the reference model, T(s) is the equivalent impedance of the specific structure, c_pIs a semi-active damping coefficient, F_pAdjustable damper control force for reference model, z_rIs the vertical input displacement of the road surface, z_urFor unsprung mass vertical displacement of the reference model, z_srFor vertical displacement of the sprung mass of the reference model,to refer to the sprung mass vertical velocity of the model,to refer to the sprung mass vertical acceleration of the model,to refer to the unsprung mass vertical velocity of the model,unsprung mass vertical acceleration for the reference model;

further, the specific structure of the ideal reference suspension is shown in fig. 3;

wherein, the step 2) is specifically as follows: defining the generalized state error e of the controlled suspension and an ideal reference suspension system:

rewriting the state variables of the controlled model and the ideal reference model in the step 1), selecting the speed and the displacement of the sprung mass as state variables, and the speed and the displacement of the unsprung mass as input variables, so that the state equations of the controlled model and the ideal reference model can be written as follows:

wherein, F_bAnd F_pControl forces, X, of the controlled model and the reference model, respectively_bAnd X_pRespectively state variables of the controlled model and the reference model,andderivatives, y, of the state variables of the controlled model and the reference model, respectively_rFor controlling system variables, A_b，A_p，B_b，B_p，C_b，C_pIs a matrix of coefficients.

Feedback regulator K using controlled model₁Control vector gain K of ideal reference model₂And a feedforward regulator K₃Resulting in a tunable system as shown in fig. 4.

Further, the method can be used for preparing a novel materialControl force F of ground controlled model_bComprises the following steps:

F_b＝K₁X_b+K₂F_p+K₃y_r，

further, the generalized state error e of the model and its derivativesComprises the following steps:

e＝X_p-X_b，

wherein, the step 3) is specifically as follows: rewriting coefficient matrix A_p、B_p、C_p：

Further, in an adaptive control system, the coefficient matrix cannot be directly adjusted by adjusting the feedback regulator K₁Control vector gain K of reference model₂And a feedforward regulator K₃The generalized state error e is made to be zero, namely the consistency of the controlled model and the ideal reference model is satisfied, and the coefficient matrix A is rewritten_p、B_p、C_pComprises the following steps:

wherein, K₁ ^*、K₂ ^*And K₃ ^*Respectively representing K when the controlled model is consistent with the reference model₁、K₂And K₃Is taken as the steady state value.

Wherein, the step 4) is specifically as follows: and (3) judging the stability of the self-adaptive control system:

selecting a corresponding Lyapunov function V:

wherein, P ∈ R^4×4、R₁、R₂And R₃Are all shown asPositively determining a symmetric matrix; tr represents the trace of the matrix; ^Trepresenting the transpose of the corresponding matrix.

Further, the following can be obtained according to the fact that the trace of the single element matrix is equal to the element value of the trace and the property of the matrix trace:

further, due to A_pFor the stabilization of the matrix, there is a positive definite symmetric matrix Q such that A_p ^TP+PA_pThe derivative of the lyapunov function is derived and substituted into the error equation to yield:

wherein the content of the first and second substances,andare respectively asAndthe derivative of the matrix is transposed.

Further, the lyapunov function is greater than zero and its derivative is less than zero, and according to the lyapunov stability theorem, the adaptive control system is globally stable when the input of the adaptive control law is continuous.

Wherein, the step 5) is specifically as follows: determining an adaptive law:

wherein, K₁(t)、K₂(t) and K₃(t) are each K₁、K₂And K₃A function with respect to time t;K₁(0)、K₂(0) and K₃(0) Are each K₁、K₂And K₃A value at time t equal to 0.

Further, the control force of the controlled model is rewritten as:

the model reference adaptive control schematic diagram is shown in fig. 4, the speed and displacement of the sprung mass are selected as state variables, the speed and displacement of the unsprung mass are selected as input signals, an error equation e is defined through a controlled model and an ideal reference model, an adaptive law is designed based on the error equation, equivalent adaptive control can be obtained, and the adaptive control is fed back to a controlled suspension in a form of main power, so that the purpose of tracking the output of the ideal model by the controlled suspension is achieved, and the comprehensive vibration isolation performance of the ISD suspension is improved.

Simulation verification is performed as follows:

simulating the vehicle ISD suspension under the vertical input displacement of the road surface to obtain a suspension performance curve, and comparing the suspension performance curve with a passive suspension, wherein simulation parameters are shown in a table 1;

further, the vertical input speed of the road surface unevennessComprises the following steps:

wherein the content of the first and second substances,vertical input speed of road surface unevenness, v running speed, w (t) white noise signal, G_q(n₀) Is the road surface unevenness coefficient.

Table 1 simulation parameters table:

fig. 5 is a time domain comparison graph of the suspension performance of the embodiment of the present invention, in which (a) is a vehicle body acceleration response graph, (b) is a suspension dynamic stroke response graph, and (c) is a tire dynamic load response graph.

Fig. 6 is a frequency domain comparison graph of the suspension performance of the embodiment of the present invention, in which (a) is a vehicle body acceleration response graph, (b) is a suspension dynamic stroke response graph, and (c) is a tire dynamic load response graph.

The results show that the method has a remarkable effect of improving the suspension performance. The suspension dynamic stroke and the tire dynamic load are obviously reduced, and the riding comfort of the vehicle is further improved.

The examples are preferred embodiments of the present invention, but the present invention is not limited to the embodiments, and modifications, variations and substitutions by those skilled in the art may be made without departing from the spirit of the present invention.