Composite control method capable of adjusting control target of intelligent suspension

Abstract

The invention discloses an intelligent suspension control target adjustable compound control method, which belongs to the technical field of suspension control, and comprises the following steps: S1, constructing a sprung control system and an unsprung control system; s2, expected control rates are determined for the sprung control system and the unsprung control system respectively, and a composite control rate is generated; s3, three-axis acceleration and angular velocity signals are collected from a chassis domain IMU, and sprung acceleration is calculated; and S4, according to the sprung acceleration, determining a sprung control system state and an unsprung control system state based on composite control rate control.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of suspension control, and particularly relates to a composite control method with adjustable intelligent suspension control targets. BACKGROUND

[0002] As an important component of an automobile chassis, a suspension system plays a key role in improving user driving experience. The suspension system includes two subsystems, namely, a spring-up subsystem and a spring-down subsystem, and is respectively associated with vehicle performance indexes of smoothness and handling stability. However, the performance indexes of the suspension system have inherent characteristics of mutual conflict, and therefore, the control algorithm thereof is required to have adjustability to realize the emphasis on a certain performance index under different working conditions.

[0003] A common method is to establish a control system as a whole and use a control algorithm with multi-target control effect. However, since the optimal weights of different performance indexes are located in parameter intervals with large magnitude differences, the process of selecting a compromise weight parameter has nonlinearity, and the parameter setting process is relatively difficult.

[0004] In addition, in order to obtain the state of the control system, a state observer needs to be additionally designed, and a subsystem can also be combined to establish an observation system as a whole. However, if a multidimensional observation system is directly constructed, for example, an observer based on LMI, the solution of the observer gain is mainly affected by the channel with a large disturbance magnitude, that is, it is difficult to ensure that the observation errors of all states are small. SUMMARY

[0005] The application proposes a composite control method with adjustable intelligent suspension control targets to solve the above problems.

[0006] The technical scheme of the application is as follows: A composite control method with adjustable intelligent suspension control targets comprises the following steps:

[0007] S1, constructing a spring-up control system and a spring-down control system;

[0008] S2, determining expected control rates for the spring-up control system and the spring-down control system respectively, and generating a composite control rate;

[0009] S3, collecting three-axis acceleration and angular velocity signals from an IMU in a chassis domain, and calculating spring-up acceleration;

[0010] S4, determining the state of the spring-up control system and the state of the spring-down control system controlled based on the composite control rate according to the spring-up acceleration.

[0011] Further, in S1, the expression of the state space equation of the spring-up control system is as follows:

[0012] ;

[0013] in, Let be the derivative of the state variables of the sprung control system. For the state variables of the sprung control system, This is the first steady system matrix of the on-spring control system. Let be the desired control quantity of the sprung control system. This is the second steady input matrix of the sprung control system. For the known terms of the sprung control system, This refers to the disturbance term in the on-spring control system.

[0014] Furthermore, in S1, the state-space equation of the unsprung control system is expressed as:

[0015] ;

[0016] in, Let be the derivative of the state variables of the unsprung control system. This is the first steady system matrix of the unsprung control system. For the state variables of the unsprung control system, This is the second steady input matrix of the unsprung control system. Let be the desired control quantity of the unsprung control system. This is the first known term in the unsprung control system. This is the second known term of the unsprung control system. This refers to the disturbance term in the unsprung control system.

[0017] Furthermore, in S2, the composite control rate The expression is:

[0018] ;

[0019] in, The normalized sprung compound control ratio (the compound control ratio is...) Normalization ), The desired control law of the sprung control system, This is the normalized unsprung composite control ratio. Let be the desired control law of the unsprung control system.

[0020] Furthermore, S3 includes the following sub-steps:

[0021] S31. Acquire triaxial acceleration and angular velocity signals from the chassis domain IMU;

[0022] S32. Differentiate the angular velocity signal to obtain the angular acceleration;

[0023] S33. Based on the triaxial acceleration and angular acceleration, calculate the triaxial acceleration at the center of gravity of the vehicle body to determine the sprung acceleration.

[0024] Furthermore, in S33, the triaxial acceleration at the vehicle's center of gravity... The expression is:

[0025] ;

[0026] in, The triaxial acceleration at the Inertial Measurement Unit (IMU) The angular acceleration at the IMU. The radius vector calibrated from the IMU to the vehicle's center of gravity. This is the angular velocity signal.

[0027] Furthermore, in S4, the expression for the state of the sprung control system is:

[0028] ;

[0029] in, The measured values ​​of the observation system, The steady system matrix of the measurement system, For the state variables of the sprung control system, The constant input matrix of the measurement system, For the control quantity of the control system, The steady disturbance input matrix of the measurement system is given. This refers to external disturbances to the observation system.

[0030] Furthermore, in S4, the expression for the state of the unsprung control system is:

[0031]

[0032] in, This is an estimate of the state of the unsprung system. For suspension dynamic deflection The estimated value, acceleration of unsprung mass The estimated value, This is the matrix transpose.

[0033] The beneficial effects of this invention are:

[0034] (1) This invention proposes a composite control framework, which transforms the control objectives of smoothness and handling stability into independent control problems of the on-spring / off-spring systems, respectively. Controllers can be designed for the two systems and the desired control rates can be obtained. At the same time, a composite control ratio is introduced, and the two desired control rates are combined by the ratio superposition to obtain the actual control quantity, i.e., the composite control rate.

[0035] (2) Regarding the selection of control methods, compared with the traditional control method that designs the up-sprocket and unsprocket as a control system, the design of independent controllers for the up-sprocket / unsprocket system is smaller in dimension and the control objective is more singular. This avoids the need to solve multi-objective control problems in the traditional control method, making the control method easier to design and suitable for adaptation to commonly used robust control algorithms.

[0036] (3) The present invention focuses on adjusting the smoothness and handling stability. Compared with the traditional control method that designs the oversprung and unsprung parts as a control system, the composite control ratio uses the method of linear superposition of two expected control laws to achieve weight adjustment, which avoids the nonlinearity of the traditional control method through the controller weight adjustment process, and makes the adjustment process of performance indicators more intuitive.

[0037] (4) This invention addresses the acquisition of the control state of the on-spring system by obtaining the on-spring acceleration through an IMU and designing a state observer to estimate the state of the on-spring system.

[0038] (15) The present invention aims to obtain the control state of the unsprung system by designing a disturbance observer for the sprung system to estimate the unsprung mass velocity and combining the suspension dynamic deflection results of the sprung system state observer to estimate the state of the unsprung system. Attached Figure Description

[0039] Figure 1 A flowchart of a composite control method for intelligent suspension control with adjustable target;

[0040] Figure 2 Diagram of a 1 / 4-car intelligent active suspension system;

[0041] Figure 3 This is a structural diagram of the composite control algorithm;

[0042] Figure 4 A comparison of the control effects of different control algorithms on a Class C random road surface;

[0043] Figure 5 The state estimation effect diagram for composite control. Detailed Implementation

[0044] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0045] like Figure 1As shown, the present invention provides a composite control method for intelligent suspension control with adjustable target, comprising the following steps:

[0046] S1. Construct the sprung control system and the unsprung control system;

[0047] S2. Determine the desired control law for the sprung control system and the unsprung control system respectively, and generate the composite control law;

[0048] S3. Acquire triaxial acceleration and angular velocity signals from the chassis domain IMU and calculate the sprung acceleration;

[0049] S4. Based on the sprung acceleration, determine the state of the sprung control system and the unsprung control system based on the composite control law.

[0050] This invention first establishes the state-space equations of the control system for the spring-loaded and unloaded subsystems respectively, and then designs control methods for these two systems to obtain the desired control law. and Then, a normalized composite control ratio is introduced. The actual composite control rate is obtained by linearly superimposing the two desired control rates. Regarding the state acquisition of the sprung control system, the sprung mass acceleration can be obtained from the signal of the chassis domain IMU, and then a state observer can be designed for the sprung control system to obtain its system state. Regarding the state acquisition of the unsprung control system, the unsprung mass velocity is estimated by designing a disturbance observer for the sprung system, and the state estimate of the unsprung control system is obtained by combining the results of the state observer.

[0051] In this embodiment of the invention, in S1, the expression for the state-space equation of the spring-loaded control system is:

[0052] ;

[0053] in, Let be the derivative of the state variables of the sprung control system. For the state variables of the sprung control system, This is the first steady system matrix of the on-spring control system. Let be the desired control quantity of the sprung control system. This is the second steady input matrix of the sprung control system. For the known terms of the sprung control system, This refers to the disturbance term in the on-spring control system.

[0054] In this embodiment of the invention, in S1, the state-space equation of the unsprung control system is expressed as:

[0055] ;

[0056] in, Let be the derivative of the state variables of the unsprung control system. This is the first steady system matrix of the unsprung control system. For the state variables of the unsprung control system, This is the second steady input matrix of the unsprung control system. Let be the desired control quantity of the unsprung control system. This is the first known term in the unsprung control system. This is the second known term of the unsprung control system. This refers to the disturbance term in the unsprung control system.

[0057] In this embodiment of the invention, in S2, the composite control rate The expression is:

[0058] ;

[0059] in, The normalized sprung compound control ratio (the compound control ratio is...) Normalization ), The desired control law of the sprung control system, This is the normalized unsprung composite control ratio. Let be the desired control law of the unsprung control system.

[0060] In this embodiment of the invention, S3 includes the following sub-steps:

[0061] S31. Acquire triaxial acceleration and angular velocity signals from the chassis domain IMU;

[0062] S32. Differentiate the angular velocity signal to obtain the angular acceleration;

[0063] S33. Based on the triaxial acceleration and angular acceleration, calculate the triaxial acceleration at the center of gravity of the vehicle body to determine the sprung acceleration.

[0064] In this embodiment of the invention, in S33, the triaxial acceleration at the center of gravity of the vehicle body... The expression is:

[0065] ;

[0066] in, The triaxial acceleration at the Inertial Measurement Unit (IMU) The angular acceleration at the IMU. The radius vector calibrated from the IMU to the vehicle's center of gravity. This is the angular velocity signal.

[0067] In this embodiment of the invention, in S4, the expression for the state of the spring control system is:

[0068] ;

[0069] in, The measured values ​​of the observation system, The steady system matrix of the measurement system, For the state variables of the sprung control system, The constant input matrix of the measurement system, For the control quantity of the control system, The steady disturbance input matrix of the measurement system is given. This refers to external disturbances to the observation system.

[0070] In this embodiment of the invention, in S4, the expression for the state of the unsprung control system is:

[0071]

[0072] in, This is an estimate of the state of the unsprung system. For suspension dynamic deflection The estimated value, acceleration of unsprung mass The estimated value, This is the matrix transpose.

[0073] In this embodiment of the invention, the suspension system, as an important component of the vehicle chassis, plays a crucial role in improving the user's driving experience. The suspension system includes two subsystems: the sprung and unsprung components, which are respectively linked to vehicle performance indicators such as ride comfort and handling stability. A 1 / 4 scale model equipped with an intelligent active suspension system is shown below. Figure 2 As shown, its dynamic equation is expressed as:

[0074] ;

[0075] In the formula, the two sets of equations are the dynamic equations for the sprung and unsprung systems, respectively; and These represent the vertical displacements of the sprung / unspecified masses, respectively. Road surface elevation; This refers to the dynamic deflection of the suspension. For wheel dynamic deformation; The control force of the actuator; the ride comfort of the suspension is mainly related to the sprung system. Relatedly, handling stability is mainly related to the unsprung system. It should be relevant, and at the same time, it should have appropriate To ensure sufficient working space for the suspension.

[0076] Active suspension systems generate control forces through actuators, directly influencing the dynamic characteristics of subsystems and improving their performance indicators. However, because the actuators are positioned between two subsystems, the control force applied to one subsystem can act as an additional excitation, simultaneously affecting the stability of the other subsystem. In other words, suspension systems inherently possess conflicting performance indicators. Therefore, the control algorithm of the suspension system must be adjustable to prioritize certain performance indicators under different operating conditions.

[0077] Furthermore, the controller implementation requires additional information about the control system's state. However, due to limitations in actual sensor deployment, the directly obtainable chassis signals are the three-axis acceleration and angular velocity signals provided by the chassis domain IMU. Obtaining unsprung acceleration requires additional acceleration sensors installed at locations such as the suspension steering knuckles, and acquiring most velocity and displacement signals requires secondary processing through integration or filtering, or the use of more complex and expensive sensors. Therefore, to reduce costs, a state observer or other methods can be used to obtain the necessary state for the control system implementation solely from IMU information.

[0078] In terms of control system design, the on-spring / off-spring subsystems are usually combined to establish a control system as a whole, and control algorithms with multi-objective control effects are used, such as designing weights in LQR and MPC control, and sliding surfaces in sliding mode control.

[0079] In terms of acquiring the state of the control system, similar to the design of the control system, the subsystems can be merged to form an overall observation system.

[0080] In terms of control system design, although establishing the overall system is relatively intuitive, the process of selecting compromise weight parameters is nonlinear because the optimal weights of different performance indicators are located in parameter ranges with large differences in magnitude, making parameter tuning difficult.

[0081] In terms of control system state acquisition, the observability of a system is not equivalent to a small observation error. Taking the design of an observation system based on LMI as an example, it can be equivalent to a robust control problem of an error system under noise. If a multi-dimensional observation system is directly constructed, the solution of the observer gain is mainly affected by the larger magnitude of the channel state and the disturbance magnitude represented by noise. That is, it is difficult to guarantee that the observation error of all states is small.

[0082] In terms of control system design, this invention establishes an independent control system for each subsystem, obtains the optimal expected control rate for each single performance index, and combines these control rates through proportional superposition to obtain the actual composite control rate. This method enables decoupled control of the subsystems, allows for independent controller design for each single performance index, making parameter tuning relatively easy, and the composite control rate is a linear superposition of different control rates, making the trade-off between performance indicators relatively intuitive.

[0083] Regarding the acquisition of control system states, based on the structure of composite control, the disturbance of the actual sprung system is constituted by the state of the unsprung system, and this disturbance can be estimated through the disturbance observer of the sprung system. Therefore, a state observer can be constructed only for the small-dimensional sprung system, and the unsprung state can be extrapolated through the disturbance observer, making full use of the control system structure while avoiding the problems of establishing an overall observation system.

[0084] The results of the composite control algorithm proposed in this invention are as follows: Figure 3 As shown.

[0085] The state-space equation of the on-spring control system is expressed as:

[0086] ;

[0087] In the formula, For state variables; This refers to the actual control quantity; For desired control quantity; For the disturbance term, ; These are known terms; , , ; ; ;definition , To match the perturbation term, .

[0088] The state-space equation of the unsprung control system is expressed as:

[0089] ;

[0090] In the formula, For state variables; This refers to the actual control quantity; For desired control quantity; For the disturbance term, The known items include and ; , , ; .

[0091] For the spring-load / spring-unload control system, different control methods can be used to design the desired control rate. and Define the composite control ratio. Normalization The design yields the actual control quantity. .

[0092] Without considering the deployment of additional height sensors or unsprung acceleration sensors, only the triaxial acceleration obtained directly from the chassis domain IMU is considered. and angular velocity Signals. Based on these signals, the easily obtainable chassis state is the sprung acceleration. The calculation method is as follows: Differentiation yields angular acceleration Combined with the radius vector from the IMU to the vehicle's center of gravity calibration The triaxial acceleration at the center of mass of the vehicle body is calculated using the acceleration composition formula for a point on a rigid body. .

[0093] Since the angular velocity and angular acceleration are the same at different positions on the rigid body, that is, the three-axis angular velocity at the center of mass of the vehicle body is the same. angular acceleration Therefore, by combining the vehicle's wheelbase and track width, the sprung accelerations of the vehicle body at the four wheels can be calculated using the acceleration synthesis method. .

[0094] Regarding the state of the sprung control system Considering the easily obtainable accelerations on the front and rear axles springs Using a state observer on the spring system, the system state is obtained. The estimated value This includes estimates of front and rear axle dynamic deflection and sprung speed. and The observation system is described as follows:

[0095] In the design process of the above state observer, the observation system is defined as follows, and can be expressed as:

[0096] ;

[0097] In the formula, For measurement; defined ,but , , .

[0098] Regarding the state of the unsprung control system Using a disturbance observer on the spring system, the matched disturbance is obtained. The estimated value and obtain The estimated value ,and That is, the estimated unsprung speed. Finally, combined with... and The state of the unsprung system is approximately obtained. The estimated value .

[0099] The key aspect of this invention lies in the control system design. It proposes a composite control framework that transforms the control objectives of smoothness and handling stability into independent control problems for the sprung / unspecified systems. Controllers can be designed separately for each system, yielding the desired control law. A composite control proportional gain is introduced, and the two desired control laws are combined through proportional superposition to obtain the actual control quantity, i.e., the composite control law.

[0100] In terms of acquiring the control system state, the sprung acceleration is obtained through an IMU, and a state observer is designed to estimate the state of the sprung system. A disturbance observer is designed for the sprung system to estimate the unsprung mass velocity, and the state of the unsprung system is estimated by combining the results from the state observer.

[0101] The following description is based on specific embodiments.

[0102] This paper presents a comparative analysis of four control algorithms on a Class C random road surface: passive suspension (no control), LMI control (control system is an oversprung subsystem), LMI-IUDE control (control system is an oversprung subsystem, IUDE algorithm is a disturbance observer algorithm), and composite control (oversprung subsystem uses LMI-IUDE control, unsprung subsystem uses basic LMI control). Furthermore, for the three control algorithms other than passive suspension, a state observer based on LMI is designed for the oversprung system to obtain the control system's state information.

[0103] In a Class C random road surface, the road surface excitation Defined as:

[0104] ;

[0105] In the formula, It is the spatial frequency power spectral density; It is the lower limit cutoff frequency in space; It is the reference spatial frequency; It is Gaussian white noise.

[0106] The suspension parameters are defined as shown in Table 1.

[0107] Table 1

[0108] Notations and parameters Meanings Values Sprung mass 345 Unsprung mass 40.5 Spring rate 18000 Tire rate 192000 Damper damping 1000 Tire damping 50 Actuator control force [-2500,2500] Vehicle speed 20 Time --

[0109] The control effects of each control algorithm on performance indicators, including sprung mass acceleration, wheel dynamic deformation, and suspension dynamic deflection, are as follows: Figure 4 The three sub-figures from top to bottom illustrate the performance indicators. The smaller images in the lower left corner of each sub-figure represent magnified views of the blue transparent boxes. The performance indicators shown in each sub-figure, from top to bottom, are sprung mass acceleration, wheel dynamic deformation, and suspension dynamic deflection. The control algorithms used for comparison are passive suspension, LMI control, LMI-IUDE control, and composite control, respectively. The root mean square values ​​of the performance indicators are shown in Table 2.

[0110] Regarding sprung mass acceleration control related to ride comfort, the improvement effect compared to passive suspension is significant in the order of basic LMI control, composite control, and LMI-IUDE control. Due to the introduction of the disturbance observer IUDE algorithm, disturbances can be compensated to enhance the algorithm's robustness; therefore, LMI-IUDE control is superior to basic LMI control. However, due to the inherent conflict between ride comfort and handling stability, composite control weakens its effect on ride comfort compared to LMI-IUDE control.

[0111] Regarding wheel dynamic deformation control, which is related to handling stability, all three algorithms showed a deterioration compared to passive suspension. However, the composite control exhibited less deterioration in handling stability compared to LMI-IUDE control, demonstrating the feasibility of balancing the two indicators. Furthermore, concerning suspension dynamic deflection control, which is related to the suspension workspace, although all three control algorithms showed a deterioration compared to passive suspension, it remained within acceptable limits.

[0112] Taking composite control as an example, the state estimation effect of the control system includes suspension dynamic deflection, sprung mass velocity, and unsprung mass velocity, respectively as follows: Figure 5 The three sub-graphs, from top to bottom, show the system states as follows: suspension dynamic deflection, sprung mass velocity, and unsprung mass velocity, respectively. The estimation error is defined as... , , The root mean square values ​​of the estimation errors are shown in Table 2. It can be seen that the estimation errors for all three states are relatively small, demonstrating the feasibility of the designed state observer and disturbance observer in estimating the state of the spring / spindle control system.

[0113] Table 2

[0114] Algorithm Passive suspension LMI control LMI-IUDE control Composite control 0.574 0.352(38.7%) 0.208(63.7%) 0.261(54.6%) 1.761 2.014(-14.4%) 2.822(-60.3%) 2.491(-41.5%) 5.169 5.171(-0.02%) 6.268(-21.3%) 5.899(-14.1%) — 0.451 0.468 0.461 — 0.675 0.343 0.795 — — — 0.890

[0115] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A smart suspension control target adjustable compound control method, characterized in that, The method comprises the following steps: S1, constructing a sprung control system and an unsprung control system; S2, determining a desired control rate for the sprung control system and the unsprung control system respectively, and generating a composite control rate; S3, collecting three-axis acceleration and angular velocity signals from the chassis domain IMU, and calculating the sprung acceleration; S4, determining the state of the sprung control system and the state of the unsprung control system controlled based on the composite control rate according to the sprung acceleration.

2. The intelligent suspension control target adjustable compound control method according to claim 1, characterized in that, In the S1, the expression of the state space equation of the sprung control system is: ； wherein is a derivative of a state variable of the on-spring control system, is a state variable of the on-spring control system, is a first constant system matrix of the on-spring control system, is a desired control variable of the on-spring control system, is a second constant input matrix of the on-spring control system, is a known term of the on-spring control system, is a disturbance term of the on-spring control system.

3. The intelligent suspension control target adjustable compound control method according to claim 1, wherein, In the S1, the state space equation of the unsprung control system is expressed as: ； wherein is a derivative of a state variable of the unsprung control system, is a first constant system matrix of the unsprung control system, is a state variable of the unsprung control system, is a second constant input matrix of the unsprung control system, is a desired control variable of the unsprung control system, is a first known term of the unsprung control system, is a second known term of the unsprung control system, is a disturbance term of the unsprung control system.

4. The intelligent suspension control target adjustable compound control method according to claim 1, wherein, In the S2, the composite control rate The expression is: ； wherein, is a normalized on-spring compound control ratio, is a desired control rate of the on-spring control system, is a normalized off-spring compound control ratio, is a desired control rate of the off-spring control system.

5. The intelligent suspension control target adjustable compound control method according to claim 1, wherein, The S3 comprises the following sub-steps: S31, collecting three-axis acceleration and angular velocity signals from the chassis domain IMU; S32, differentiating the angular velocity signal to obtain angular acceleration; S33, calculating the three-axis acceleration at the center of mass of the vehicle body according to the three-axis acceleration and the angular acceleration, and determining the sprung acceleration.

6. The intelligent suspension control target adjustable compound control method according to claim 5, wherein, In the S33, the three-axis acceleration at the vehicle body mass center The expression is: ； wherein, is the three-axis acceleration at the inertial measurement unit, is the angular acceleration at the IMU, is the vector radius of the IMU to the body center of mass calibration, is the angular velocity signal.

7. The intelligent suspension control target adjustable compound control method according to claim 1, wherein, In the S4, the expression of the state of the sprung control system is: ； wherein is a measurement of the observation system, is a constant system matrix of the measurement system, is a state variable of the on-spring control system, is a constant input matrix of the measurement system, is a control variable of the control system, is a constant disturbance input matrix of the measurement system, is an external disturbance of the observation system.

8. The intelligent suspension control target adjustable compound control method according to claim 1, wherein, In the S4, the expression of the state of the unsprung control system is: ； wherein is an estimate of the unsprung system state, is an estimate of the suspension deflection , is an estimate of the unsprung mass acceleration , is the matrix transpose.

Images