Movatterモバイル変換


[0]ホーム

URL:


CN120171234B - Self-adaptive suspension system posture adjusting method - Google Patents

Self-adaptive suspension system posture adjusting method

Info

Publication number
CN120171234B
CN120171234BCN202510644958.2ACN202510644958ACN120171234BCN 120171234 BCN120171234 BCN 120171234BCN 202510644958 ACN202510644958 ACN 202510644958ACN 120171234 BCN120171234 BCN 120171234B
Authority
CN
China
Prior art keywords
planet
planetary
suspension system
optimization algorithm
optimal solution
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202510644958.2A
Other languages
Chinese (zh)
Other versions
CN120171234A (en
Inventor
王霞
张文硕
韩启凤
冯增雷
江城城
李婕
张状
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Engineering Vocational and Technical University
Original Assignee
Shandong Engineering Vocational and Technical University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong Engineering Vocational and Technical UniversityfiledCriticalShandong Engineering Vocational and Technical University
Priority to CN202510644958.2ApriorityCriticalpatent/CN120171234B/en
Publication of CN120171234ApublicationCriticalpatent/CN120171234A/en
Application grantedgrantedCritical
Publication of CN120171234BpublicationCriticalpatent/CN120171234B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

Translated fromChinese

本发明公开了一种自适应悬挂系统姿态调节方法,属于PID技术优化领域,具体为:步骤1、采集悬挂系统的车辆姿态数据,包括车身的垂向位移、俯仰角、横摆角等数据,建立自适应悬挂系统姿态调节的动态模型;步骤2、改进行星优化算法,引入拓扑绝缘体维度敏感性机制与自适应的搜索调节策略改进行星优化算法;步骤3、通过改进行星优化算法(IPOA)优化悬挂系统姿态调节的PID控制器的Kp、Ki、Kd系数;步骤4、根据实时采集到的姿态变化数据,采用优化后的PID控制器实时调节悬挂系统的刚度和阻尼参数;本发明方法有效克服了传统PID控制方法在悬挂系统姿态调节中存在的响应迟缓、超调量大及稳态精度不足的问题,确保车辆姿态符合预期要求。

The present invention discloses a method for adaptive suspension system posture adjustment, which belongs to the field of PID technology optimization. Specifically, the method comprises the following steps: step 1, collecting vehicle posture data of the suspension system, including vertical displacement, pitch angle, yaw angle and other data of the vehicle body, and establishing a dynamic model for adaptive suspension system posture adjustment; step 2, improving the planetary optimization algorithm, introducing a topological insulator dimensional sensitivity mechanism and an adaptive search adjustment strategy to improve the planetary optimization algorithm; step 3, optimizing the Kp, Ki, and Kd coefficients of the PID controller for suspension system posture adjustment by using an improved planetary optimization algorithm (IPOA); step 4, using the optimized PID controller to adjust the stiffness and damping parameters of the suspension system in real time based on the posture change data collected in real time. The method of the present invention effectively overcomes the problems of slow response, large overshoot and insufficient steady-state accuracy existing in traditional PID control methods in suspension system posture adjustment, thereby ensuring that the vehicle posture meets the expected requirements.

Description

Self-adaptive suspension system posture adjusting method
Technical Field
The invention belongs to the technical field of PID control optimization, and particularly relates to a self-adaptive suspension system posture adjusting method.
Background
The suspension system is one of key components of the vehicle, and aims to relieve ground impact force, ensure that the jolt of the vehicle body is reduced when the road surface of the vehicle is uneven or the load is changed, and further improve the driving stability. With the increasing demands of the automotive industry for vehicle performance, modern suspension systems are evolving towards intellectualization and adaptation. For example, intelligent suspension systems dynamically adjust suspension elements by monitoring vehicle conditions and road conditions in real time via sensors and Electronic Control Units (ECU). The self-adaptive suspension system can adjust suspension characteristics according to road conditions and driving styles, and automatically adjust the performance of the vehicle under various working conditions.
In adaptive suspension systems, attitude adjustment is an important control objective to maintain vehicle body stability. The main purpose of the vehicle is to ensure the stable running of the vehicle under various road conditions, avoid the occurrence of excessive pitching, rolling or jumping of the vehicle body, and further improve the operability and the comfort of the vehicle. Adaptive suspension systems typically employ active or semi-active suspension technology in combination with sensors and control algorithms to adjust body attitude.
In order to fully play the advantages of the self-adaptive suspension system, the parameter optimization design of the suspension system is particularly important. In recent years, with the rapid development of computer technology, intelligent optimization algorithms such as genetic algorithm, particle swarm optimization algorithm, and gray wolf optimization algorithm have been widely used for parameter optimization of suspension systems due to their excellent global searching capability. The Planetary Optimization Algorithm (POA) is an emerging global optimization algorithm, the inspiration of which derives from the principle of gravitational interaction in celestial mechanics. The algorithm regards candidate solutions in the optimization problem as "planets" by simulating the gravitational relationship between the sun and the planets, and searches and optimizes through the gravitational effects between the "planets". The POA algorithm has a simple structure, high efficiency and strong global searching capability, and is therefore widely focused. However, due to the global nature of the gravitational effects, POA algorithms may converge prematurely to a locally optimal solution, thereby losing the ability to continue to explore globally optimal solutions.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a self-adaptive suspension system posture adjusting method, which optimizes a PID controller for suspension system posture adjustment by improving an epicyclic optimization algorithm (IPOA) and aims at improving the control precision and response speed of the posture adjusting controller, combines a traditional PID control algorithm and an intelligent optimization algorithm, the proportional coefficient Kp, the integral coefficient Ki and the differential coefficient Kd of the PID controller are optimized through the IPOA algorithm, so that the suspension system can be dynamically and adaptively adjusted according to real-time road conditions and vehicles, the control precision is improved, the overshoot is effectively reduced, the accurate adjustment of the vehicle posture is realized in the adaptive suspension system, and the running stability and the comfort of the vehicles under complex road conditions are obviously improved.
In order to achieve the aim, the invention adopts the following technical scheme that the self-adaptive suspension system posture adjusting method comprises the following specific steps:
And step 1, collecting vehicle attitude data of a suspension system, and establishing a dynamic model of attitude adjustment of the self-adaptive suspension system for optimization and adjustment of a subsequent control strategy.
Step 2, improving a planetary optimization algorithm, wherein the specific improvement strategy is as follows:
im1, introducing a topological insulator dimension sensitivity mechanism to improve a mass calculation equation in the POA algorithm;
Im2, in order to break through the limitation of the traditional gravitation model in the original algorithm, a space-time nonlinear mapping and topology control mechanism is introduced, so that the planet individuals can perform self-adaptive global search in the search space;
Im3, introducing a self-organizing criticality mechanism, so that the planetary individuals conduct self-organizing evolution under the guidance of dynamic cognitive feedback and local disturbance, and conduct more flexible local search in a complex space.
And 3, performing parameter optimization on the PID controller with the posture adjusted by improving an planetary optimization algorithm (IPOA), and optimizing a proportional coefficient Kp, an integral coefficient Ki and a differential coefficient Kd in the PID controller by simulating the law of planetary orbit motion by the IPOA algorithm.
And 4, carrying out real-time adjustment on the posture of the suspension system by adopting an optimized PID controller, and adjusting the vertical supporting force applied on the suspension element in the suspension system by the PID controller according to the posture data acquired in real time to ensure that the posture of the vehicle meets the target requirement.
Further, in the step 1, vehicle posture data of the suspension system is collected, including vertical displacement, pitch angle, roll angle, etc. of the vehicle body, and dynamic information such as road surface condition, load change, etc., and an adaptive suspension system posture control model is built for subsequent control optimization and adjustment.
Further, a vehicle dynamics model is established, wherein the vehicle dynamics model can be described by a three-degree-of-freedom system, namely vertical displacement, pitch angle and yaw angle, and the mathematical model of the vertical displacement is as follows:
(1);
In the formula (1), the components are as follows,For the quality of the car body,Is the vertical displacement of the mass center of the vehicle body,The vertical support force applied for the ith suspension;
Further, the mathematical model of pitch angle is:
(2);
in the formula (2), the amino acid sequence of the compound,For the moment of inertia of the vehicle body about the transverse axis,Is the pitching angle of the vehicle body,AndThe distances from the front axle and the rear axle to the mass center of the vehicle body are respectively,-Vertical support forces applied for the four suspensions;
Still further, the mathematical model of yaw motion is:
(3);
in the formula (3), the amino acid sequence of the compound,For the moment of inertia of the vehicle body about the longitudinal axis,Is the transverse swing angle of the vehicle body,Is the lateral distance from the center of mass to the left and right wheels.
Further, in order to implement posture adjustment, an error of each posture amount needs to be defined, the error is defined as a difference between an expected posture value and an actual posture value, and the defined error is:
(4);
in the formula (4), the amino acid sequence of the compound,Respectively represent a vertical displacement error, a pitch angle error and a yaw angle error,Respectively representing a target vertical displacement, a target pitch angle and a yaw angle,Is the actual vertical displacement of the car body,Is the actual pitch angle of the car body,Is the actual yaw angle of the vehicle body;
Further, the vertical displacement error will beThe input PID controller unit processes the error through the PID controller which is optimized on line by the IPOA algorithm and outputs a control signalSuspension element acting on an adaptive suspension system, adjusting in real time the 4 vertical support forces exerted on the suspension element-Thereby adjusting the vertical displacement, pitch angle and yaw angle of the vehicle body and ensuring the stable posture of the vehicle body suspension system;
Further, a control signal is establishedThe dynamic mapping relation with the vertical supporting force is as follows:
(5);
In the formula (5), the amino acid sequence of the compound,-For vertical supporting force in 4 suspension elements, A is a pitch angle compensation coefficient of the suspension elements, and B is a yaw angle compensation coefficient of the suspension elements;
Further, during the attitude adjustment, the Kp, ki, kd coefficients of the PID controller unit are mapped online to individual positions in the search space of the improved planetary optimization algorithm, and three dimensional values of the planetary individual positions represent [ Kp Ki Kd ], respectively.
Further, introducing a topological insulator dimension sensitivity mechanism in the steps 2 and Im1 improves a quality calculation equation in the POA algorithm, wherein the mechanism is used for introducing sensitivity activation coefficients of each dimensionThe high dimension and the low dimension of the individual are distinguished, and the effective distinction of the high dimension and the low dimension individual is realized, so that the influence of the distance change on different dimensions on the quality value has differential response. When the sensitivity of a certain dimension is higher, the contribution of the dimension to quality evaluation can be obviously enhanced, so that a planet individual is guided to accelerate the convergence speed in the key dimension, the optimization capacity of an algorithm in a high-dimensional search space is improved, and an improved quality calculation equation is as follows:
(6);
In the formula (6), the amino acid sequence of the compound,For the mass of the ith planet,For the mass of the j-th planet, i=1,.. nPop, j=1,.. nPop, nPop is the algorithm population number,The fitness of the ith or jth planet,For the latitude value of the search space,For the sensitivity activation coefficient of the d-th dimension,For the parameter value of the ith planet in the d dimension,For the value of the current optimal solution in the d-th dimension,For the attractive force parameters between the planet and the optimal solution,Being extremely small constant, prevents denominator zero from causing calculation errors.
Furthermore, a space-time nonlinear mapping and topology control mechanism is introduced in the steps 2 and Im2, so that the planetary individuals can perform self-adaptive global search in the search space, and the improved global search mathematical model is as follows:
(7);
in the formula (7), the amino acid sequence of the compound,For the updated individual position of the ith planet, t is the current iteration number,For the individual position of the current iteration of the ith planet,As the gravitational acceleration factor,AndTo take the value of the random number between 0 and 1,For the optimal solution position in the current iteration,For the attractive force parameters between the planet and the optimal solution,For the purpose of the diversity of the search parameters,As a function of the state of the planet,In order to adjust the parameters of the influence of the neighborhood,Is the neighborhood of the ith planet in the current iteration, wherein,The mathematical model of (a) is:
(8);
in the formula (8), t is the current iteration number,For the fitness of the ith planet in the kth generation,For the fitness of the optimal solution,For the individual position of the ith planet at the kth generation,Individual positions at the kth generation are the optimal solutions.
Furthermore, a self-organizing criticality mechanism is introduced in the steps 2 and Im3, so that the planetary individuals are subjected to self-organizing evolution under the guidance of dynamic cognitive feedback and local disturbance, and are guided to perform more flexible searching in a complex space, and the improved local searching mathematical model is as follows:
(9);
In the formula (9), the amino acid sequence of the compound,For the updated individual position of the ith planet,For the individual position of the current iteration of the ith planet, t is the current iteration number,Critical locations evolved for the system's self-organization,Is a locally optimal solution in the planetary neighborhood,For the critical disturbance intensity to be a critical disturbance intensity,The intensity is guided for the neighborhood.
Further, in the step 3, parameter optimization is performed on the PID controller with posture adjustment by improving an planetary optimization algorithm (IPOA), the IPOA algorithm performs optimization by simulating a law of planetary orbit motion, and a proportional coefficient Kp, an integral coefficient Ki and a differential coefficient Kd in the PID controller are optimized, and the specific steps are as follows:
S1, initializing a population scale nPop of an Improved Planetary Optimization Algorithm (IPOA), a maximum iteration number MaxIter, a search space dimension Dim, and search space upper and lower bounds [ Ub, lb ], wherein Ub and Lb are unit vectors of Dim order;
s2, initializing b, c, G and Rmin parameters of the improved planetary optimization algorithm, initializing individual positions of the improved planetary optimization algorithm, generating initial positions of population individuals, and initializing a mathematical model of the individual positions to be:
(10);
In the formula (10), the amino acid sequence of the compound,For a randomly generated initial position of the planet,For random numbers with values between [0,1], ub and Lb are the same as the meaning;
S3, calculating fitness of each planet, updating individual positions and fitness of all the planets in a greedy selection mode, and selecting the individual position with the minimum fitness in the population as an optimal solutionAnd recordIs the best fitness in the current iteration;
S4, calculating and updating a moment parameter M, wherein a mathematical model of the moment parameter M is as follows:
(11);
In the formula (11), the amino acid sequence of the compound,AndFor the mass of the ith planet and the mass of the jth planet,The Cartesian distance between two planets, G is the gravitational parameter;
S5, calculating Cartesian distance between the planet and the optimal solution,The mathematical model of (a) is:
(12);
in the formula (12), each parameter has the same meaning as above;
s6, when>The algorithm enters a global exploration stage, simulates a planet construction motion track far away from the sun to carry out global search, updates individual positions, and adopts the improved mathematical model of global exploration as above,The mathematical model of (a) is:
(13);
In the formula (13), the amino acid sequence of the compound,The initial Cartesian distance between the planet and the optimal solution is the same as the other parameters;
s7, when<The algorithm enters a local development stage, a planet construction motion track close to the sun is simulated to perform local search, the individual position is updated, and an improved mathematical model for local exploration is the same as the above;
S8, checking whether the current iteration times t is larger than MaxIter, if so, outputting an optimal solution of the improved planetary optimization algorithm, decoding the optimal solution into a proportional coefficient Kp, an integral coefficient Ki and a differential coefficient Kd in the PID controller, if not, executing t=t+1, and returning to S3 to continue iterative optimization.
The invention provides a self-adaptive suspension system posture adjustment method, which optimizes a PID controller for posture adjustment in a suspension system by improving an planetary optimization algorithm (IPOA), and has the following beneficial effects compared with the prior art:
P1, by introducing a topological insulator dimension sensitivity mechanism, a mass calculation equation in a Planetary Optimization Algorithm (POA) is improved, so that a planetary individual can be adaptively adjusted according to sensitivity characteristics of search space dimension change, and the sensing and adaptation capacity of an optimization process to a complex non-uniform space structure is enhanced, so that the accuracy and dynamic response capacity of PID controller parameter optimization in the suspension system attitude adjustment process are improved;
p2, by introducing a space-time nonlinear mapping and topology control mechanism, a single linear evolution mode of position updating in a traditional gravity model is broken through, self-adaptive distortion searching and topology perception migration of a planet individual in a search space are realized, a global exploration range is effectively expanded, and searching efficiency and adaptability to complex working condition changes in a gesture adjustment PID parameter optimization process are improved;
And P3, by introducing a self-organizing criticality mechanism, the planetary individuals form self-organizing evolution behaviors under the dynamic cognitive feedback and the local disturbance guidance, and the local jump and the global adjustment can be adaptively carried out according to the system state change in the posture adjustment of the suspension system, so that the adjustment precision of the PID controller is obviously improved, and the posture stability and the comfort of the suspension system under the complex road condition are improved.
Drawings
FIG. 1 is a roadmap of an adaptive suspension attitude adjustment technique.
FIG. 2 is a flowchart of an improved planetary optimization algorithm.
FIG. 3 is a graph comparing fitness of a modified planetary optimization algorithm with a conventional planetary optimization algorithm.
FIG. 4 is a graph comparing vertical displacement of a modified planetary optimization algorithm and a conventional planetary optimization algorithm.
Fig. 5 is a pitch angle comparison graph of the improved planetary optimization algorithm and the ordinary planetary optimization algorithm.
Fig. 6 is a graph comparing yaw angles of the improved planetary optimization algorithm and the ordinary planetary optimization algorithm.
Detailed Description
The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments, and all other embodiments obtained by those skilled in the art without making creative efforts based on the embodiments of the present invention are included in the protection scope of the present invention.
The invention provides a self-adaptive suspension system posture adjusting method, which is shown in figure 1 and comprises the following specific steps:
and 1, acquiring vehicle attitude data of a suspension system, including vertical displacement, pitch angle, yaw angle and the like of a vehicle body, monitoring road surface conditions and dynamic response of the vehicle, and establishing a dynamic model of the attitude adjustment of the self-adaptive suspension system based on the data.
Specifically, in the embodiment of the invention, mathematical models are respectively built for vertical displacement, pitch angle and yaw angle through Matlab, wherein the vertical displacement mathematical model is as follows:
function dz = heave_model(t, z, F_z_func, m)
% vertical displacement equation
Dz1=z (2);% velocity is equal to the first derivative of displacement
Dz2=sum (f_z_func (t))/m%acceleration is determined by the sum of the four suspension forces
dz = [dz1; dz2];
end。
Further, the mathematical model of pitch angle is:
function dtheta = pitch_model(t, theta, F1_func, F2_func, F3_func, F4_func, l_f, l_r, I_x)
% pitch equation
Dtheta =theta (2);% angular velocity is equal to the first derivative of pitch angle
Dtheta = (f1_func (t) ×l_f-f2_func (t) ×l_f-f3_func (t) x l_r+f4_func (t) x l_r)/i_x;% angular acceleration
dtheta = [dtheta1; dtheta2];
end。
Still further, the mathematical model of yaw motion is:
function dphi = roll_model(t, phi, F1_func, F2_func, F3_func, F4_func, l_w, I_y)
% yaw equation
Dphi =phi (2);% angular velocity is equal to the first derivative of the yaw angle
Dphi = (f1_func (t) ×l/u w+F3_func (t) ×l_w-f2_func (t) x l_w-f4_func (t) x l_w)/i_y;% angular acceleration
dphi = [dphi1; dphi2];
end。
Further, in order to implement posture adjustment, an error of each posture amount needs to be defined, the error is defined as a difference between an expected posture value and an actual posture value, and the defined error is:
(4);
in the formula (4), the amino acid sequence of the compound,Respectively represent a vertical displacement error, a pitch angle error and a yaw angle error,The target vehicle body posture value represents the target vertical displacement, the target pitch angle and the yaw angle respectively,Is the actual vertical displacement of the car body,Is the actual pitch angle of the car body,Is the yaw angle of the vehicle body.
Further, the vertical displacement error will beThe input PID controller unit processes the error through the PID controller which is optimized on line by the IPOA algorithm and outputs a control signalSuspension element acting on an adaptive suspension system, adjusting in real time the vertical support forces exerted on the 4 suspension elements-And the vertical displacement, pitch angle and yaw angle of the vehicle body are regulated, so that the stable posture of the vehicle body suspension system is ensured.
Further, a control signal is establishedThe dynamic mapping relation with the vertical supporting force is as follows:
(5);
In the formula (5), the amino acid sequence of the compound,-For vertical supporting force in 4 suspension elements, A is the pitch angle compensation coefficient of the suspension elements, the value is 0.1, B is the yaw angle compensation coefficient of the suspension elements, and the value is 0.08.
Furthermore, the attitude difference value is input into a PID controller unit, the difference value is processed through an online optimized PID control algorithm, and a control signal is output, and acts on a suspension element of the self-adaptive suspension system to adjust the rigidity and damping characteristics of the suspension system in real time.
Step 2, improving a planetary optimization algorithm, wherein the specific improvement strategy is as follows:
Im1, introducing a mass calculation equation in a topological insulator dimension sensitivity mechanism improvement algorithm, wherein the improved mass calculation equation is as follows:
(6);
In the formula (6), the amino acid sequence of the compound,For the mass of the ith planet,For the mass of the j-th planet, i=1,.. nPop, j=1,.. nPop, nPop is the algorithm population number,The fitness of the ith or jth planet,For the latitude value of the search space,For the sensitivity activation coefficient of dimension d, the value is 0.8,For the parameter value of the ith planet in the d dimension,For the value of the current optimal solution in the d-th dimension,The attractive force parameter between the planet and the optimal solution is 2,Is extremely small constant and takes the value of 0.000001;
Im2, introducing a global search mathematical model of a space-time nonlinear mapping and topology control mechanism improvement algorithm, wherein the improved global search mathematical model is as follows:
(7);
in the formula (7), the amino acid sequence of the compound,For the updated individual position of the ith planet, t is the current iteration number,For the individual position of the current iteration of the ith planet,As the gravitational acceleration factor,AndTo take the value of the random number between 0 and 1,For the optimal solution position in the current iteration,For the attractive force parameters between the planet and the optimal solution,For the purpose of the diversity of the search parameters,As a function of the state of the planet,In order to adjust the parameters of the neighborhood influence, the value is 0.05,Is the neighborhood of the ith planet in the current iteration, wherein,The mathematical model of (a) is:
(8);
in the formula (8), t is the current iteration number,For the fitness of the ith planet in the kth generation,For the fitness of the optimal solution,For the individual position of the ith planet at the kth generation,Individual position at the kth generation for the optimal solution;
im3, introducing a local search mathematical model of a self-organizing criticality mechanism improvement algorithm, wherein the improved local search mathematical model is as follows:
(9);
In the formula (9), the amino acid sequence of the compound,For the updated individual position of the ith planet,For the individual position of the current iteration of the ith planet, t is the current iteration number,Critical locations evolved for the system's self-organization,Is a locally optimal solution in the planetary neighborhood,To linearly decrease the critical disturbance intensity from 0.5 to 0,The neighborhood guiding strength is 0.2;
step 3, parameter optimization is carried out on the PID controller with the posture adjusted by improving an IPOA algorithm, the IPOA algorithm carries out optimization by simulating the law of planetary orbit motion, and the proportional coefficient Kp, the integral coefficient Ki and the differential coefficient Kd in the PID controller are optimized, wherein the specific steps are as follows:
s1, initializing a population scale nPop of an Improved Planetary Optimization Algorithm (IPOA) to be 30, wherein the maximum iteration number MaxIter is 20, the dimension Dim of a search space is 3, the upper bound Ub of the search space is [20,20,20], and the lower bound is [0,0.000001,0];
S2, initializing individual positions of the improved planetary optimization algorithm, generating initial positions of planetary individuals, and initializing mathematical models of the individual positions to be:
(10);
In the formula (10), the amino acid sequence of the compound,For a randomly generated initial position of the planet,For random numbers with values between [0,1], ub and Lb are the same as the meaning;
S3, calculating fitness of each planet, updating individual positions and fitness of all the planets in a greedy selection mode, and selecting the individual position with the minimum fitness in the population as an optimal solutionAnd recordIs the best fitness in the current iteration;
S4, calculating and updating a moment parameter M, wherein a mathematical model of the moment parameter M is as follows:
(11);
In the formula (11), the amino acid sequence of the compound,AndFor the mass of the ith planet and the mass of the jth planet,The Cartesian distance between two planets, G is the gravitational parameter;
S5, calculating Cartesian distance between the planet and the optimal solution,The mathematical model of (a) is:
(12);
in the formula (12), each parameter has the same meaning as above;
s6, when>The algorithm enters a global exploration stage, simulates a planet construction motion track far away from the sun to carry out global search, updates individual positions, and adopts the improved mathematical model of global exploration as above,The mathematical model of (a) is:
(13);
In the formula (13), the amino acid sequence of the compound,The initial Cartesian distance between the planet and the optimal solution is the same as the other parameters;
s7, when<The algorithm enters a local development stage, a planet construction motion track close to the sun is simulated to perform local search, the individual position is updated, and an improved mathematical model for local exploration is the same as the above;
S8, checking whether the current iteration times t is larger than MaxIter, if so, outputting an optimal solution of the improved planetary optimization algorithm, decoding the optimal solution into a proportional coefficient Kp, an integral coefficient Ki and a differential coefficient Kd in the PID controller, if not, executing t=t+1, and returning to S3 to continue iterative optimization.
Further, comprehensively considering the control precision, response time and system energy consumption of the attitude adjustment of the self-adaptive suspension system, and selecting and improving an objective function of the planetary optimization algorithm as follows:
(14);
In the formula (13), J is an fitness value calculated by the objective function,Respectively represent a vertical displacement error, a pitch angle error and a yaw angle error,In order to adjust the magnitude of the overshoot,The integral of the square of the control signal is output by the PID controller, T is the total running time of the system, n is the accumulated times of errors,As the weight coefficient of the light-emitting diode,The value of the water-soluble fluorescent powder is 0.03,The value is 0.01, and the method is used for balancing the control precision and the energy consumption of the system.
And 4, carrying out real-time adjustment on the posture of the suspension system by adopting an optimized PID controller, wherein the PID controller adjusts the rigidity and damping parameters of the suspension system according to the posture change data acquired in real time to ensure that the posture of the vehicle meets the target requirement, and specifically comprises the following steps:
Step1, setting the running time of a self-adaptive suspension system attitude adjustment simulation model to be 20s, setting the sampling time to be 0.5s, setting the initial vertical displacement of the vehicle to be 0, setting the initial pitching angle to be 0 degrees, setting the initial yaw angle to be 0 degrees, setting the target vertical displacement to be 3, setting the target pitching angle to be 0 degrees and setting the target yaw angle to be 1 degree;
Step2, establishing an improved planetary optimization algorithm mathematical model, and writing a connection function through Matlab for transmitting data between the simulation model and the improved planetary optimization algorithm mathematical model;
Step3, connecting Kp, ki and Kd coefficients of the PID controller with planetary individual positions of the improved planetary optimization algorithm through a connecting functionIn association with the correlation of the two,=[Kp、Ki、Kd];
Step4, running an improved planetary optimization algorithm mathematical model, carrying out algorithm iterative optimization, outputting individual solutions in each iteration, and decoding values of each individual solution in different dimensions into Kp, ki and Kd coefficients of the PID controller;
Step5, outputting Kp, ki and Kd parameters to a control system simulation model, and running the control system simulation model;
Step6, judging whether iteration is terminated, if so, outputting an optimal solution of the improved planetary optimization algorithm, and decoding values of each latitude of the optimal solution into Kp, ki and Kd parameters, wherein the optimal Kp, ki and Kd parameters are 14.7367, 0.0642 and 16.2716.
In the implementation step, the change curve of the fitness value along with the iteration times in the optimization process of the common planetary optimization algorithm and the improved planetary optimization algorithm is compared and analyzed, as shown in fig. 3, the fitness value of the common planetary optimization algorithm is rapidly reduced in the previous 6-generation iteration process and then enters a convergence platform period, the fitness value is basically stabilized at about 0.035, the overall convergence speed is relatively high but the final convergence quality is limited, the fitness value cannot be further reduced, the improved planetary optimization algorithm is obviously reduced at about 6-generation, but the follow-up still can keep a moderate reduction trend, and is further optimized after 10-generation, and finally converges to about 0.018 in 20-generation, which is obviously superior to the common algorithm. In a comprehensive view, the improved planetary optimization algorithm not only has higher convergence speed in the early stage, but also keeps better optimization capacity in the middle and later stages, finally obtains a lower fitness value, and verifies the effectiveness and superiority of the improved method in the processes of improving optimizing precision and accelerating convergence.
In the implementation step, the vertical displacement curve of the common planetary optimization algorithm and the vertical displacement curve of the improved planetary optimization algorithm are compared and analyzed, as shown in fig. 4, the common planetary optimization algorithm rises rapidly in the initial stage, but has obvious overshoot phenomenon, the maximum displacement is close to 4cm, the convergence process is slower and can be basically stabilized near the target displacement value only after about 10 seconds, the transition process is not smooth enough, the improved planetary optimization algorithm also rises rapidly in the initial stage, but the maximum overshoot is obviously reduced, the oscillation amplitude and the frequency are obviously reduced, the whole displacement response curve is more stable, the displacement response curve converges near the target displacement value within about 6 seconds, and the stability is kept better. Comprehensive analysis shows that the improved planetary optimization algorithm is superior to the common algorithm in aspects of overshoot suppression, convergence speed, system stability and the like, and the effectiveness and superiority of the improved strategy are verified.
In the implementation step, the pitch angle response curves of the common planetary optimization algorithm and the improved planetary optimization algorithm are compared and analyzed, as shown in fig. 5, the common planetary optimization algorithm is changed rapidly in the initial stage, but obvious negative overshoot appears, the maximum pitch angle is close to-0.6 degrees, then positive overshoot appears, the oscillation amplitude is larger, the convergence process is slower, and finally the stability can be gradually achieved within about 10 seconds, the improved planetary optimization algorithm also has negative offset with a certain amplitude in the initial stage, but the overshoot amplitude is obviously reduced, the positive oscillation is also greatly reduced, the overall curve change is more gentle, the convergence is close to the target pitch angle of 0 degrees within 6 seconds, and the follow-up fluctuation is extremely small. In a comprehensive view, the improved planetary optimization algorithm is superior to the common algorithm in the aspects of overshoot suppression, oscillation control and convergence speed, and the dynamic response performance and stability of the system are effectively improved.
In this implementation step, the yaw angle response curves of the ordinary planetary optimization algorithm and the modified planetary optimization algorithm are compared and analyzed, as shown in fig. 6, the ordinary planetary optimization algorithm rises rapidly in the initial stage, but obviously overshoots occur, the maximum yaw angle exceeds the target value by about 0.4 degrees, then a certain amplitude of oscillation is generated, the overall convergence speed is slower, and the stability is basically near the target value after about 8 seconds, while the modified planetary optimization algorithm rises rapidly in the initial stage, but basically has no obvious overshooting, the response process is smooth, and converges rapidly to near the target yaw angle by about 1 degrees in about 2 seconds, the subsequent fluctuation is extremely small, and excellent dynamic performance and steady state characteristics are shown. In a comprehensive view, the improvement of the planetary optimization algorithm can effectively reduce the overshoot of the system, remarkably improve the convergence speed and enhance the stability of the system, and has obvious advantages compared with the common planetary optimization algorithm.
In summary, the invention provides a self-adaptive suspension system attitude adjustment method, which effectively solves the problems of slow response, large overshoot and insufficient steady-state precision in the suspension system attitude adjustment process of the traditional PID control method by improving the Kp, ki and Kd coefficients of a PID controller for optimizing the suspension system attitude adjustment by an epicyclic optimization algorithm (IPOA), and can realize efficient optimization of the PID control coefficient by introducing a topological insulator dimension sensitivity mechanism and a self-adaptive search adjustment strategy, remarkably improve the attitude control performance of a suspension system under different working conditions, and particularly maintain the quick response and high stability of the system in the face of external disturbance such as complex road conditions, vehicle load change and the like, thereby effectively improving the travelling comfort and safety of a vehicle.

Claims (6)

Translated fromChinese
1.一种自适应悬挂系统姿态调节方法,其特征在于,具体包括:1. A method for adjusting the attitude of an adaptive suspension system, characterized by comprising:步骤1、采集悬挂系统的车辆姿态数据,包括车身的垂向位移、俯仰角、横摆角数据,建立自适应悬挂系统姿态调节的动态模型;Step 1: Collect vehicle attitude data of the suspension system, including vertical displacement, pitch angle, and yaw angle data of the vehicle body, and establish a dynamic model for attitude adjustment of the adaptive suspension system;步骤2、改进行星优化算法,具体改进策略为:Step 2: Improve the planet optimization algorithm. The specific improvement strategy is:Im1、引入拓扑绝缘体维度敏感性机制改进行星优化算法中的质量计算方程;改进后的质量计算方程为:Im1. Introducing the topological insulator dimensional sensitivity mechanism to improve the mass calculation equation in the planet optimization algorithm; the improved mass calculation equation is:(6); (6);式(6)中,为第i颗行星的质量,为第j颗行星的质量,i=1,…,nPop,j=1,…,nPop,nPop为算法种群数量,为第i颗或第j颗行星的适应度,为搜索空间的纬度值,为第d维的敏感度激活系数,为第i颗行星在第d维的参数值,为当前最优解在第d维的值,为行星与最优解之间的吸引力参数,为极小常数,防止分母为零导致计算错误;In formula (6), is the mass of the i-th planet, is the mass of the jth planet, i=1,…,nPop, j=1,…,nPop, nPop is the number of algorithm populations, is the fitness of the i-th or j-th planet, is the latitude value of the search space, is the sensitivity activation coefficient of the d-th dimension, is the parameter value of the ith planet in the dth dimension, is the value of the current optimal solution in the dth dimension, is the attraction parameter between the planet and the optimal solution, It is a very small constant to prevent the denominator from being zero and causing calculation errors;Im2、引入时空非线性映射和拓扑控制机制,使行星个体能够在搜索空间中进行自适应全局搜索;具体为:引入时空非线性映射和拓扑控制机制改进行星优化算法的全局搜索数学模型,改进后的全局搜索数学模型为:Im2. Introducing spatiotemporal nonlinear mapping and topological control mechanisms enables planetary individuals to perform adaptive global searches in the search space. Specifically, introducing spatiotemporal nonlinear mapping and topological control mechanisms improves the global search mathematical model of the planetary optimization algorithm. The improved global search mathematical model is:(7); (7);式(7)中,为第i颗行星更新后的个体位置,t为当前迭代次数,为第i颗行星当前迭代的个体位置,为引力加速因子,为取值在[0,1]之间的随机数,为当前迭代中最优解位置,为行星与最优解之间的吸引力参数,为多样性搜索参数,为行星状态函数,为调整邻域影响的参数,为第i颗行星在当前迭代中的邻域,其中,的数学模型为:In formula (7), is the updated individual position of the i-th planet, t is the current iteration number, is the individual position of the current iteration of the i-th planet, is the gravitational acceleration factor, and is a random number between [0, 1]. is the optimal solution position in the current iteration, is the attraction parameter between the planet and the optimal solution, For the diversity search parameter, is the planetary state function, To adjust the parameters of neighborhood influence, is the neighborhood of the i-th planet in the current iteration, where The mathematical model is:(8); (8);式(8)中,t为当前迭代次数,为第i颗行星在第k代的适应度,为最优解的适应度,为第i颗行星在第k代的个体位置,为最优解在第k代的个体位置;In formula (8), t is the current iteration number, is the fitness of the i-th planet in the k-th generation, is the fitness of the optimal solution, is the individual position of the i-th planet in the k-th generation, is the individual position of the optimal solution in the kth generation;Im3、引入自组织临界性机制,使行星个体在动态认知反馈与局部扰动的引导下进行自组织性演化,并引导行星个体在复杂空间中进行局部搜索;Im3. Introducing a self-organized criticality mechanism to enable individual planets to self-organize and evolve under the guidance of dynamic cognitive feedback and local perturbations, and to guide individual planets to conduct local searches in complex spaces;步骤3、通过改进行星优化算法对姿态调节的PID控制器进行参数优化,改进的行星优化算法通过模拟行星轨道运动的规律进行参数寻优,从而优化PID控制器中的比例系数Kp、积分系数Ki和微分系数Kd;Step 3: Optimize the parameters of the PID controller for attitude control by improving the planetary optimization algorithm. The improved planetary optimization algorithm optimizes the parameters by simulating the laws of planetary orbital motion, thereby optimizing the proportional coefficient Kp, integral coefficient Ki and differential coefficient Kd in the PID controller.步骤4、采用优化后的PID控制器进行悬挂系统姿态的实时调节,PID控制器根据实时采集到的姿态变化数据,调节悬挂系统的刚度和阻尼参数。Step 4: Use the optimized PID controller to adjust the suspension system posture in real time. The PID controller adjusts the stiffness and damping parameters of the suspension system according to the posture change data collected in real time.2.根据权利要求1所述的一种自适应悬挂系统姿态调节方法,其特征在于,所述步骤2中,所述引入自组织临界性机制,具体为:引入自组织临界性机制改进行星优化算法的局部搜索数学模型,改进后的局部搜索数学模型为:2. The method for adjusting the attitude of an adaptive suspension system according to claim 1, wherein in step 2, the introducing of the self-organized criticality mechanism comprises: introducing the self-organized criticality mechanism to improve the local search mathematical model of the planetary optimization algorithm, wherein the improved local search mathematical model is:(9); (9);式(9)中,为第i颗行星更新后的个体位置,为第i颗行星当前迭代的个体位置,t为当前迭代次数,为系统自组织演化出的临界位置,为行星邻域内的局部最优解,为临界扰动强度,为邻域引导强度。In formula (9), is the updated individual position of the i-th planet, is the individual position of the current iteration of the i-th planet, t is the number of current iterations, is the critical position evolved by the self-organization of the system. is the local optimal solution in the neighborhood of the planet, is the critical disturbance intensity, is the neighborhood guidance strength.3.根据权利要求1所述的一种自适应悬挂系统姿态调节方法,其特征在于,所述步骤3,通过改进的行星优化算法对姿态调节的PID控制器进行参数寻优,具体步骤为:3. The method for adjusting the attitude of an adaptive suspension system according to claim 1, wherein the step 3 optimizes the parameters of the PID controller for attitude adjustment using an improved planetary optimization algorithm, specifically comprising the following steps:S1、初始化改进的行星优化算法的种群规模nPop,最大迭代次数MaxIter、搜索空间维度Dim,以及搜索空间上下界[Ub,Lb],Ub和Lb为Dim阶的单位向量;S1. Initialize the population size nPop, the maximum number of iterations MaxIter, the search space dimension Dim, and the upper and lower bounds of the search space [Ub, Lb] of the improved planetary optimization algorithm, where Ub and Lb are unit vectors of order Dim;S2、初始化改进行星优化算法的b、c,G,Rmin参数,初始化改进行星优化算法的个体位置,生成种群个体初始位置,初始化个体位置的数学模型为:S2. Initialize the b, c, G, and Rmin parameters of the improved planetary optimization algorithm, initialize the individual positions of the improved planetary optimization algorithm, generate the initial positions of the individuals in the population, and the mathematical model for initializing the individual positions is:(10); (10);式(10)中,为随机生成的行星初始位置,为取值在[0,1]之间的随机数,Ub和Lb的意义同上;In formula (10), is the initial position of the randomly generated planet, is a random number between [0, 1], and Ub and Lb have the same meaning as above;S3、计算每颗行星的适应度;通过贪婪选择的方式更新所有行星的个体位置和适应度,选择种群中适应度最小的个体位置为最优解,并记录的适应度为当前迭代中的最佳适应度,计算并更新引力矩参数M;S3. Calculate the fitness of each planet; update the individual positions and fitness of all planets through greedy selection, and select the individual position with the smallest fitness in the population as the optimal solution , and record The fitness is the best fitness in the current iteration , calculate and update the gravitational moment parameter M;S4、计算行星与最优解之间的笛卡尔距离,当>,算法进入全局探索阶段,改进后的全局探索的数学模型同上;S4. Calculate the Cartesian distance between the planet and the optimal solution ,when > , the algorithm enters the global exploration stage, and the improved mathematical model of global exploration is the same as above;S5、当<,算法进入局部开发阶段,改进后的局部探索的数学模型同上;S5. When < , the algorithm enters the local development stage, and the improved mathematical model of local exploration is the same as above;S6、检查当前迭代次数t是否大于MaxIter,若是,则输出改进行星优化算法的最优解,并将最优解解码成PID控制器中的比例系数Kp、积分系数Ki和微分系数Kd,若否,则执行t=t+1,并返回S3继续迭代寻优。S6. Check whether the current number of iterations t is greater than MaxIter. If so, output the optimal solution of the improved planetary optimization algorithm and decode the optimal solution into the proportional coefficient Kp, integral coefficient Ki and differential coefficient Kd in the PID controller. If not, execute t=t+1 and return to S3 to continue iterative optimization.4.根据权利要求3所述的一种自适应悬挂系统姿态调节方法,其特征在于,所述步骤3、S2中的数学模型为:4. The method for adjusting the attitude of an adaptive suspension system according to claim 3, wherein in steps 3 and S2 The mathematical model is:(11); (11);式(11)中,为行星与最优解间的初始笛卡尔距离,为搜索空间的纬度值,Ub和Lb为搜索空间上下界。In formula (11), is the initial Cartesian distance between the planet and the optimal solution, is the latitude value of the search space, and Ub and Lb are the upper and lower bounds of the search space.5.根据权利要求3所述的一种自适应悬挂系统姿态调节方法,其特征在于,所述步骤3、S3中M的数学模型为:5. The method for adjusting the attitude of an adaptive suspension system according to claim 3, wherein the mathematical model of M in step 3, S3 is:(12); (12);式(12)中,为第i颗行星的质量和第j颗行星的质量,为两颗行星之间的笛卡尔距离,G为引力参数。In formula (12), and is the mass of the i-th planet and the mass of the j-th planet, is the Cartesian distance between the two planets, and G is the gravitational parameter.6.根据权利要求3所述的一种自适应悬挂系统姿态调节方法,其特征在于,所述步骤3、S4中的数学模型为:6. The method for adjusting the attitude of an adaptive suspension system according to claim 3, wherein in steps 3 and S4 The mathematical model is:(13); (13);式(13)中,为第i颗行星当前迭代的个体位置,d取值为1,…,Dim,为搜索空间的纬度值,为算法最优解。In formula (13), is the individual position of the current iteration of the i-th planet, d takes the value of 1,…,Dim, is the latitude value of the search space, is the optimal solution of the algorithm.
CN202510644958.2A2025-05-202025-05-20Self-adaptive suspension system posture adjusting methodActiveCN120171234B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202510644958.2ACN120171234B (en)2025-05-202025-05-20Self-adaptive suspension system posture adjusting method

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202510644958.2ACN120171234B (en)2025-05-202025-05-20Self-adaptive suspension system posture adjusting method

Publications (2)

Publication NumberPublication Date
CN120171234A CN120171234A (en)2025-06-20
CN120171234Btrue CN120171234B (en)2025-08-29

Family

ID=96039555

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202510644958.2AActiveCN120171234B (en)2025-05-202025-05-20Self-adaptive suspension system posture adjusting method

Country Status (1)

CountryLink
CN (1)CN120171234B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CA3190318A1 (en)*2021-06-262022-12-29Dingxuan ZHAOInertial regulation method and control system of active suspensions based on terrain ahead of vehicle
CN117360144A (en)*2023-10-172024-01-09北京理工大学Suspension dynamic travel self-adaptive coordination control method and system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN117284035A (en)*2023-11-032023-12-26江苏大学Active control method and system for vehicle body posture compensation
CN117601609A (en)*2023-11-302024-02-27合肥工业大学 Active suspension control method and system for in-wheel motor vehicles under complex road conditions

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CA3190318A1 (en)*2021-06-262022-12-29Dingxuan ZHAOInertial regulation method and control system of active suspensions based on terrain ahead of vehicle
CN117360144A (en)*2023-10-172024-01-09北京理工大学Suspension dynamic travel self-adaptive coordination control method and system

Also Published As

Publication numberPublication date
CN120171234A (en)2025-06-20

Similar Documents

PublicationPublication DateTitle
US6212466B1 (en)Optimization control method for shock absorber
JP2005538886A (en) Fuzzy controller using a reduced number of sensors
Thoresson et al.Efficient optimisation of a vehicle suspension system, using a gradient-based approximation method, Part 1: Mathematical modelling
CN113591360B (en)Magneto-rheological damper structural parameter optimization method based on whole vehicle dynamics model
CN107220405B (en)Automobile chassis system integration multi-objective optimization method based on improved cell membrane optimization algorithm
CN115826402A (en)Active suspension control strategy generation method based on deep reinforcement learning algorithm
Zhao et al.PID control of vehicle active suspension based on particle Swarm optimization
CN119975527A (en) A coordinated control method of steering and suspension for autonomous driving vehicles based on reinforcement learning
CN118578827A (en) A magnetorheological vehicle suspension backstepping compensation control method, system and controller
CN120171234B (en)Self-adaptive suspension system posture adjusting method
Mahmoodabadi et al.Optimal fuzzy robust state feedback control for a five DOF active suspension system
CN117841591B (en)ISD suspension control method based on improved fuzzy neural network PID
Pereira et al.Vibration control enhancement in a full vehicle dynamic model by optimization of the controller’s gain parameters
Vu et al.Multi objective H∞ active anti-roll bar control for heavy vehicles
CN118418637A (en) Vehicle suspension control method, device, system, electronic device and storage medium thereof
CN118311903A (en) Two-wheeled robot balance control method, system, storage medium and computer equipment based on LQR controller
Dridi et al.Optimizing full vehicle active suspension model with advanced reinforcement learning controller
He et al.Improved LSTM Neural Network-Based PID Control for Automotive Active Suspension Systems
Gao et al.The Study of Optimization and Matching to Spring and Antiroll Bar Stiffness of Suspension for Multiresponse Target of Whole Vehicle under Sine‐Swept Steering Input
CN119369876B (en) Vehicle active suspension control method and application for flexible chassis extension connection
Zhuo et al.Research on LQR control of active suspension based on Improved Grey Wolf Optimization algorithm
CN120245659A (en) A vehicle roll stability control method based on deep reinforcement learning
Fan et al.Quadruped robot locomotion via soft actor-critic with muti-head critic and dynamic policy gradient
CN120517117A (en)Active suspension intelligent control method, device and medium based on extreme learning machine
CN120327179A (en) An intelligent control method for active suspension based on deep Q-network

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp