Load adaptive control method based on CDC semi-active suspension
By switching the nonlinear semi-active suspension model and the load adaptive control method designed by segmented Lyapunov theory, the problem of unstable response of CDC semi-active suspension system under load changes and road disturbance is solved, and the stability of the system is achieved and improve comfort, suitable for embedded electronic control system applications.
Patent Information
- Application Number
- CN202411257593.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-11-12
AI Technical Summary
The response of CDC semi-active suspension system is unstable under load changes and road disturbances. The existing load adaptive control algorithm is difficult to effectively apply, and it is easy to lead to performance losses and frequent saturation of control outputs.
Adaptive load control method based on switching nonlinear semi-active suspension model and segmented Lyapunov theory is adopted. By establishing a switching T-S system model, a variable gain controller is designed to ensure that the control output is within the feasible domain of driving current and avoid current shearing. performance degradation.
It achieves improved system stability and comfort under load changes and road disturbance conditions, reduces design conservatism and control volume change rate, avoids frequent saturation output, and is suitable for embedded electronic control system applications.
Smart Images

Figure CN118927897A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of vehicle vibration reduction control, in particular to a load adaptive control method based on a CDC semi-active suspension. Background Art
[0002] The vehicle suspension system should be able to attenuate the vibrations transmitted from the road surface to the vehicle body and reduce the dynamic loads applied to the wheel axles under various operating conditions. The adverse effects of human vibration become apparent when the exposure time exceeds the recommended standards set by the International Organization for Standardization (ISO) 2631. Integrating continuous damping control (CDC) shock absorbers into the suspension system can significantly improve the system's comfort performance. The CDC shock absorbers developed by ZF SACHS have been successfully used in commercial vehicles, including VW-Phaeton, Ford-Mondeo and GM-Lacrosse. CDC shock absorbers provide continuously variable damping characteristics by adjusting the orifice size in the hydraulic flow valve. Compared with magnetorheological and electrorheological shock absorbers, CDC shock absorbers have a simpler and more economical design, and also have the advantages of good reliability and durability.
[0003] However, on the one hand, the vehicle mass varies due to factors such as the number of passengers and the load carried, and these fluctuations may lead to undesirable responses even if the controller is fine-tuned based on the nominal model. In the past two decades, a large number of studies have been devoted to investigating load-adaptive active suspension systems. However, it is impractical to directly apply the load-adaptive controller designed for active suspension systems to CDC semi-active suspension systems due to the inherent nonlinear characteristics of CDC semi-active suspension systems; on the other hand, the load-adaptive controller needs to avoid the performance loss caused by the clipping strategy commonly used in semi-active suspensions. Therefore, designing and developing a load-adaptive control algorithm under the premise of ensuring the CDC damper driving current constraint is of great significance for improving the comfort and handling of vehicles in practical applications. Summary of the invention
[0004] In order to address the shortcomings of the background technology, the present invention provides a load adaptive control method based on CDC semi-active suspension, which can effectively handle load changes and road disturbances. On the premise that the vehicle weight can be accurately estimated, the nonlinear characteristics and dissipative constraints of the actuator are fully considered, and the load adaptive capability of the CDC semi-active suspension system is realized within the feasible domain of the driving current.
[0005] To achieve the above object, the present invention adopts the following technical solution: a load adaptive control method based on CDC semi-active suspension, comprising the following steps:
[0006] Step 1: Establish a switched nonlinear semi-active suspension model with zero symmetry constraints
[0007] The CDC damper characteristic data obtained from the test is calculated based on the suspension extension and contraction speed signal. The segments are as follows:
[0008]
[0009] Where α is the partition index, and is the speed boundary of the partition;
[0010] The value range of the damper response current I is in, definition have d is the time constant, σ des is the control input of the converted system, and its value range is The damping force model is expressed as:
[0011]
[0012] in,
[0013]
[0014] Take the state quantity Road disturbance Among them, L b is the vertical displacement of the sprung mass vibration, L t is the vertical displacement of the unsprung mass vibration, L g For the road excitation displacement, according to the partition of the CDC damper characteristic data, the switching nonlinear semi-active suspension model is established as follows:
[0015]
[0016] z=C α (M b ,σ)x+δ α
[0017] When α=1,3,
[0018]
[0019]
[0020]
[0021]
[0022] When α=2,
[0023]
[0024]
[0025]
[0026] a α =[0 0 0 0 0] T , δ α =[0 0 0] T
[0027] in, is the performance output of the system, and is the weight coefficient, B u =[0 0 0 0 d] T , B d =[0 -1 0 0 0] T , F d is the input damping force of the CDC semi-active suspension system, M b is the time-varying sprung mass, M t is the unsprung mass, k s is the spring stiffness, k t is the vertical stiffness of the tire, c p is the equivalent passive damping, and are all constants;
[0028] Step 2: Introduce the vehicle weight parameter into the antecedent variable cluster and convert the switching nonlinear semi-active suspension model into a vehicle weight-dependent switching TS system model
[0029] Take the antecedent variable The velocity origin partition is not included to obtain the antecedent variable ξ α,2 =ψ(σ),ξ α,3 =Γ(σ), including the velocity origin partition The switching nonlinear semi-active suspension model is converted into the switching TS system model as follows:
[0030]
[0031]
[0032] In the formula, is the number of TS subsystems in the α partition, δ(x) is the switching signal, h αj (ξ) is the weight of the jth TS subsystem in the α partition, A αj and C αj are the coefficient matrix and output matrix of the j-th TS subsystem in the α partition respectively;
[0033] Step 3: List all possible switching index pairs in the system
[0034] Using Lyapunov function as V(x k )=x k T P α x k , all possible switching index pairs of the switching nonlinear semi-active suspension model are as follows:
[0035]
[0036] Among them, β is the partition that the system enters at the next moment;
[0037] Step 4: Based on the piecewise Lyapunov theory, the process of obtaining the variable gain controller is given
[0038] The system is discretized as follows:
[0039]
[0040]
[0041] In the formula, and is the constant matrix after discretization;
[0042] The control quantity is expressed as:
[0043]
[0044] In the formula, K αS is the control gain of the sth TS subsystem in the α partition;
[0045] r is H ∞ Performance index, so that the following inequality holds:
[0046]
[0047] Partition information Through S-Procedure transformation:
[0048]
[0049] In the formula, λ αj is a positive scalar, ε is the calibration parameter;
[0050] definition use The following inequality holds:
[0051]
[0052] In the formula, κ αj is a positive scalar,
[0053] Apply Schur's lemma and define Y αS =K αs Q α , the above formula is equivalent to:
[0054]
[0055] In the formula,
[0056] Step 5: Give the final control input expression based on the actual semi-active suspension system
[0057]
[0058] In the formula, I des is the control current of the damper.
[0059] Furthermore, it is assumed that the vehicle weight change satisfies the following conditions:
[0060] M b,low ≤M b ≤M b,high
[0061] Then the antecedent variable ξ in step 2 α,1 According to the nonlinear idea of sectors:
[0062]
[0063] in:
[0064]
[0065] In the formula, and is the membership function.
[0066] Furthermore, in step 4, the calibration parameter ε takes a value between 0 and 1.
[0067] Compared with the prior art, the present invention has the following beneficial effects:
[0068] 1. The method of the present invention converts the control problem based on asymmetric driving current constraint into a switching nonlinear system with symmetric control quantity constraint, which is convenient for the design of the controller;
[0069] 2. The method of the present invention designs the controller based on the piecewise Lyapunov theory, which reduces the design conservatism caused by the increase of subsystems due to the introduction of vehicle weight parameters into the variable weight of the antecedent table;
[0070] 3. The method of the present invention will not produce frequently saturated control outputs, and the control quantity change rate is relatively smooth compared to the existing algorithms, which is conducive to the application in embedded electronic control systems;
[0071] 4. Compared with a load adaptive controller that uses damping force as a control variable, the method of the present invention avoids the expected performance degradation caused by current "shearing". BRIEF DESCRIPTION OF THE DRAWINGS
[0072] Figure 1 is a system schematic diagram of a quarter suspension dynamics model in the control method of the present invention;
[0073] Figure 2 It is a segmented schematic diagram of the characteristic data of the CDC damper in the control method of the present invention;
[0074] Figure 3 is a schematic diagram of switching index pairs in the control method of the present invention;
[0075] Figure 4 is a comparison diagram of the system FR response results under overload conditions obtained from the actual vehicle test in the embodiment;
[0076] Figure 5 It is a comparison chart of the system FR response results under underload conditions obtained from the actual vehicle test in the embodiment. DETAILED DESCRIPTION
[0077] The technical solution of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0078] like Figure 1 to Figure 3 As shown, a load adaptive control method based on a CDC semi-active suspension comprises the following steps:
[0079] Step 1: Establish a switched nonlinear semi-active suspension model with zero symmetry constraints
[0080] Combination Figure 1 As shown, the quarter suspension dynamics model is given as follows:
[0081]
[0082]
[0083]
[0084] Here we assume that the vehicle weight change satisfies the following conditions:
[0085] M b,low ≤M b ≤M b,high
[0086] Where M b is the time-varying sprung mass, L b is the vertical displacement of the sprung mass vibration, k s is the spring stiffness, L t is the vertical displacement of the unsprung mass vibration, c p is the equivalent passive damping, F d is the input damping force of the CDC semi-active suspension system, M t is the unsprung mass, k t is the vertical stiffness of the tire, L g is the road excitation displacement, I is the response current of the damper, d is the time constant, I des is the control current of the damper.
[0087] Combination Figure 2 As shown in the figure, the CDC damper characteristic data obtained from the test is calculated according to the extension and contraction speed signal of the suspension. The segments are as follows:
[0088]
[0089] Where α is the partition index, and is the velocity boundary of the partition. Correspondingly, the damping force model adopts the least square fitting as follows:
[0090]
[0091] in,
[0092]
[0093] Where b α 、c α ,d α 、e α 、f α , g α 、h α and i α are all constants.
[0094] The value range of the damper response current I is in, definition have σ des is the control input of the converted system, and its value range is Then F d It can be further expressed as:
[0095]
[0096] in,
[0097]
[0098] In the formula, and are all constants.
[0099] Take the state quantity Road disturbance According to the partition of CDC damper characteristic data, the switching nonlinear semi-active suspension model is established as follows:
[0100]
[0101] z=C α (M b ,σ)x+δ α
[0102] Where:
[0103] When α=1,3,
[0104]
[0105]
[0106]
[0107]
[0108] When α=2,
[0109]
[0110]
[0111]
[0112] a α =[0 0 0 0 0] T , δ α =[0 0 0] T
[0113] in, is the performance output of the system, and is the weight coefficient, B u =[0 0 0 0 d] T , B d =[0 -1 0 0 0] T ;
[0114] Step 2: Introduce the vehicle weight parameter into the antecedent variable cluster and convert the switching nonlinear semi-active suspension model into a vehicle weight-dependent switching TS system model
[0115] Take the antecedent variable According to the nonlinear idea of sectors, we have:
[0116]
[0117] in, and is the membership function, and the expression is:
[0118]
[0119] In the partition that does not contain the velocity origin, take the antecedent variable ξ α,2 =ψ(σ),ξ α,3 =Γ(σ), in the partition containing the velocity origin, The switched nonlinear semi-active suspension model is further converted into a switched TS system model as follows:
[0120]
[0121]
[0122] In the formula, is the number of TS subsystems in the α partition, δ(x) is the switching signal, which is determined by the extension and retraction speed signal of the suspension Decision, h αj (ξ) is the weight of the jth TS subsystem in the α partition, A αj and C αj are the coefficient matrix and output matrix of the j-th TS subsystem in the α partition respectively;
[0123] Step 3: List all possible switching index pairs in the system
[0124] Using Lyapunov function as V(x k )=x k T P α x k , since the controller needs to ensure the switching stability between partitions where switching may occur, combined with Figure 3 As shown in Figure 2, all possible switching index pairs of the switching nonlinear semi-active suspension model are as follows:
[0125]
[0126] Among them, β is the partition that the system enters at the next moment;
[0127] Step 4: Based on the piecewise Lyapunov theory, the process of obtaining the variable gain controller is given
[0128] The system is discretized as follows:
[0129]
[0130]
[0131] In the formula, and is the constant matrix after discretization.
[0132] The control quantity is expressed as:
[0133]
[0134] In the formula, K αs is the control gain of the sth TS subsystem in the α partition.
[0135] If the designed controller can stabilize the closed-loop system under the road disturbance input, and the L2 gain from the disturbance to the performance output is The following inequality must hold:
[0136]
[0137] In the formula, H ∞ Performance indicators.
[0138] Partition information Through S-Procedure transformation:
[0139]
[0140] In the formula,
[0141] λ αj is a positive scalar, ε is a calibration parameter, which takes a value between 0 and 1.
[0142] definition use The following inequality holds:
[0143]
[0144] In the formula, κ αj is a positive scalar,
[0145] Apply Schur's lemma and define Y αs =K αs Q α , the above formula is equivalent to:
[0146]
[0147] In the formula,
[0148] The controller derivation including the velocity origin partition can be obtained based on the above process without considering the partition information;
[0149] Step 5: Give the final control input expression based on the actual semi-active suspension system
[0150]
[0151] Example
[0152] This embodiment is based on actual vehicle tests on standard bad roads. Figure 4 (a) and (b) are the gains of the road excitation to the vehicle body vibration acceleration and tire dynamic load when the vehicle weight is overloaded, Figure 5 (a) and (b) are the gains of the road excitation to the vehicle body vibration acceleration and tire dynamic load when the vehicle weight is underloaded. It can be seen that under load changes, compared with the damping force algorithm and the algorithm based on constant vehicle weight, the present invention further improves the overall performance of the system. Figure 4 (c) and (d) show that compared with the damping force controller, the control current output by the controller of the present invention is always within the executable range of the solenoid valve drive current.
[0153] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other forms of assembly without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations within the meaning and range of equivalents of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
[0154] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.
Claims
1. A load adaptive control method based on CDC semi-active suspension, Features: The following steps are involved: Step 1: Establish a switched nonlinear semi-active suspension model with zero symmetry constraints The CDC damper characteristic data obtained from the test is calculated based on the suspension extension and contraction speed signal. The segments are as follows: In the formula, α is the partition index, and is the speed boundary of the partition; The value range of the damper response current I is in, definition have is the time constant, σ des is the control input of the converted system, and its value range is The damping force model is expressed as: in, Take the state Road disturbance Among them, L b is the vertical displacement of the sprung mass vibration, L t is the vertical displacement of the unsprung mass vibration, L g For the road excitation displacement, according to the partition of the CDC damper characteristic data, the switching nonlinear semi-active suspension model is established as follows: z=C α (M b ,σ)x+δ α When α=1,3, When α=2, a α =[0 0 0 0 0] T ,d α =[0 0 0] T in, is the performance output of the system, and is the weight coefficient, B d =[0 -1 0 0 0] T , F d is the input damping force of the CDC semi-active suspension system, M b is the time-varying sprung mass, M t is the unsprung mass, k s is the spring stiffness, k t is the vertical stiffness of the tire, c p is the equivalent passive damping, and are all constants; Step 2: Introduce the vehicle weight parameter into the antecedent variable cluster and convert the switching nonlinear semi-active suspension model into a vehicle weight-dependent switching TS system model Take the antecedent variable The velocity origin partition is not included to obtain the antecedent variable ξ α,2 =ψ(σ),ξ α,3 =Γ(σ), including the velocity origin partition The switching nonlinear semi-active suspension model is converted into the switching TS system model as follows: In the formula, is the number of TS subsystems in the α partition, δ(x) is the switching signal, h αj (ξ) is the weight of the jth TS subsystem in the α partition, A αj and C αj are the coefficient matrix and output matrix of the j-th TS subsystem in the α partition respectively; Step 3: List all possible switching index pairs in the system Using Lyapunov function as V(x k )=x k T P α x k , all possible switching index pairs of the switching nonlinear semi-active suspension model are as follows: Among them, β is the partition that the system enters at the next moment; Step 4: Based on the piecewise Lyapunov theory, the process of obtaining the variable gain controller is given The system is discretized as follows: In the formula, and is the constant matrix after discretization; The control quantity is expressed as: In the formula, K αs is the control gain of the sth TS subsystem in the α partition; H ∞ Performance index, so that the following inequality holds: Partition information Through S-Procedure transformation: In the formula, λ αj is a positive scalar, ε is the calibration parameter; definition use The following inequality holds: In the formula, κ αj is a positive scalar, Apply Schur's lemma and define Y αs =K αs Q α , the above formula is equivalent to: In the formula, Step 5: Give the final control input expression based on the actual semi-active suspension system In the formula, I des is the control current of the damper.
2. A load adaptive control method based on CDC semi-active suspension according to claim 1, Features: Assume that the vehicle weight changes satisfy the following conditions: M b,low ≤M b ≤M b,high Then the antecedent variable ξ in step 2 α,1 According to the nonlinear idea of sectors: in: In the formula, and is the membership function.
3. The load adaptive control method based on CDC semi-active suspension according to claim 1, Features: In the step 4, the calibration parameter ε takes a value between 0 and 1.
