Research on Suspension System Based on Genetic Algorithm and Neural Network Control

Tang, Chuan Yin; Zhao, Guang-Yao; Li, Hua; Zhou, Shu Wen

doi:10.1109/icicta.2009.120

Cited by 10 publications

(4 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…And then, take performance index (16) with q 1 = 150, q 2 = 100, q 3 = 200, r 0 = 1. The control law OVC is carried out by (26)- (28) and starts to training the NNOVC, where a three-layer NN as described in Fig. 2 is adopted, the bipolar sigmoid function (47) is employed, and the update rule (53) is applied to train the NN with the schedule: number of hidden layer: 2; number of neurons of each layer: 10; hidden neuron activation functions: σ s (ϑ) = 1 − e −ϑ 1 + e −ϑ ; learning rate in the weight tuning law: P 3 = diag{10}and λ w = 0.02; network input:…”

Section: Simulation Resultsmentioning

confidence: 99%

Neural Network-Based Supervisory Control for Quarter-Car Suspension with Nonlinearity and Uncertainty

Lei

Guo

2015

Integrated Ferroelectrics

View full text Add to dashboard Cite

This paper addresses a neural network (NN)-based vibration control problem for aquarter-car suspension model with nonlinearity and uncertainty. First, combining the state system and the disturbance exosystem, an augmented system is constructed. Thus, an optimal regulator problem of the augmented system is built. After using Pontryagin minimum principle and the dynamic programming approach, an approximative optimal vibration control (OVC) is obtained, which is derived from a Riccati equation and two vector differential equations. A feedforward compensator is designed to suppress the vibration. And through defining the adjoint vectors, the nonlinearity and uncertainty are compensated and the physical realization problem of the disturbance compensation term can be solved. Therefore, supervised by the designed OVC and trained by the designed update rule, the Neural Network-based optimal vibration control (NNOVC) is constructed. Consequentially, the NNOVC enables making the system satisfy performance requirements easily. The related proof is given. Moreover, some numerical simulations are made. It is verified that the NNOVC can replace OVC to control the suspension effectively in face of road excitation and vehicle velocity changes. The effectiveness, flexibility, and the intelligence of NNOVC are illustrated.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Neural Network-Based Supervisory Control for Quarter-Car Suspension with Nonlinearity and Uncertainty

Lei

Guo

2015

Integrated Ferroelectrics

View full text Add to dashboard Cite

show abstract

“…The selection of weighting coefficient refers to the references, By combining the subjective weighting ratio coefficient of each evaluation index with the same scale ratio coefficient, The final weighting coefficient related to each evaluation indexes of ride comfort in LQG control is determined and select q1=1; q3=675.51; q5=36274 at last. The other specific simulation parameters refers to the references [2] ,which is shown in table The corresponding Simulink simulation picture is shown in figure 3: With the limit of space, in order to observe and compare, the simulation pictures of performance of passive suspension and the active suspension with the optimal K by the default of road excitation based on GA are listed in 4-9.…”

Section: Simulation Calculation and Results Analysismentioning

confidence: 99%

Vehicle Active Suspension with Four-Degrees of Freedom of Optimizing the Value of K Based on Genetic Algorithm

Zuo

Yan

Yang

2011

AMR

View full text Add to dashboard Cite

A vehicle active suspension model with 1 / 2 ，four-degrees of freedom is established and by combining genetic algorithm with optimal control theory，the author presents a new control method of active suspension that is to optimize the value of K controlled by LQG in default of road input based on genetic algorithm and makes a simulation in the environment of Matlab / Simulink. By simulation and analysis，the result indicates that，this method has an obvious effect on improving comprehensive performance of vehicles，such as ride comfort and operate stability and so on.

show abstract

“…Recently, automobile ASS has become an effective and popular way to deal with this big trade-off between the inherent conflicting suspension performances [2][3]. For vehicle ASS under normal working operation, a number of control algorithms, such as backstepping control [4][5], ∞ control [6][7], sliding mode control [8][9], neural network control [10][11], etc., have been proposed to achieve the improvement of vehicle dynamics performances.…”

Section: Introductionmentioning

confidence: 99%

Design of a Sliding Mode Observer-Based Fault Tolerant Controller for Automobile Active Suspensions With Parameter Uncertainties and Sensor Faults

2020

View full text Add to dashboard Cite

This paper presents a sliding mode observer-based fault tolerant controller for a class of active suspension with the parametric uncertainties and sensor faults. First, T-S fuzzy approach is employed to represent the inertial parameters uncertainties and sensor faults, and then an augmented vehicle dynamics model is established. To estimate both system state variables and sensor fault signal simultaneously, a sliding-mode observer is investigated, and the fault tolerant control law is further derived in terms of the estimation of state and state error within Lyapunov theory framework, moreover, the sufficient conditions for the existence of this controller is deduced and solved via a set of linear matrix inequalities. Finally, a complete comparative simulation case is provided to demonstrate the effectiveness and feasibility of the designed control method. INDEX TERMS active suspension system, sliding mode observer, fault tolerant control, T-S fuzzy approach. Nomenclature mc Vehicle body mass [kg] ϕ Pitch angle of vehicle body [rad] Iy Moment of inertia in vehicle body [kg• m² ] cf Damping coefficient of front suspension [N•s•m-1 ] muf Mass of the front suspension wheel [kg] cr Damping coefficient of rear suspension [N•s•m-1 ] mur Mass of the rear suspension wheel [kg] ktf Stiffness of the front tire [N•m-1 ] a Distance from COG to the front axle [m] ktr Stiffness of the rear tire [N•m-1 ] b Distance from COG to the rear axle [m] uf Front actuator force [F] zuf Vertical displacement of the front wheel [m] ur Rear actuator force [F] zur Vertical displacement of the rear wheel [m] ∆yf Front suspension dynamic travel [m] zrf Road excitations at the front wheel [m] ∆yr Rear suspension dynamic travel [m] zrr Road excitations at the rear wheel [m] kf Stiffness coefficient of the front suspension [N•m-1 ] zc Vertical displacement of vehicle body [m] kr Stiffness coefficient of the rear suspension [N•m-1 ]

show abstract

Research on Suspension System Based on Genetic Algorithm and Neural Network Control

Cited by 10 publications

References 1 publication

Neural Network-Based Supervisory Control for Quarter-Car Suspension with Nonlinearity and Uncertainty

Neural Network-Based Supervisory Control for Quarter-Car Suspension with Nonlinearity and Uncertainty

Vehicle Active Suspension with Four-Degrees of Freedom of Optimizing the Value of K Based on Genetic Algorithm

Design of a Sliding Mode Observer-Based Fault Tolerant Controller for Automobile Active Suspensions With Parameter Uncertainties and Sensor Faults

Contact Info

Product

Resources

About