Design Of Ball-beam System Controller Based On Adaptive Genetic PID

Ball-beam system is a common experimental equipment in the control laboratory,because of its simple structure,small occupancy space and easy observation.It also has characteristics of non-linearity and open-loop instability,it is a typical device that verify control theory and methods.The main work of this paper as follows:Firstly,the components of ball-beam system and the nonlinearity and open-loop instability of the system is introduced,The mathematical model of the mechanical system and motor servo of the ball-beam system is established by the mechanism method,which is expressed by the transfer function.Secondly,the stability and controllability of the ball-beam system are analyzed.It is concluded that the ball-beam system is unstable but controllable,which lays the foundation for the controller design.The state feedback controller is designed and added to the ball-beam system for simulation and experiment.The experimental results are quite different from the simulation results.Analyzing the actual influencing factors,the compensation link is added to the system,and the controller calibration experiment is performed again.Thirdly,the PID control principle and the traditional PID controller parameter tuning method are introduced.The theoretical basis of genetic algorithm is discussed,and the genetic algorithm is applied to the optimization of PID control parameters.It is applied to the ball-beam system,and the simulation and experimental research are carried out.Finally,the emphasis is on the convergence of genetic algorithms.Because the probability of crossover and mutation in genetic algorithm can not meet the individualized requirements of individuals,Limiting the scope of the search,leading to premature or local convergence of the genetic algorithms.The adaptive genetic algorithm improves the crossover and mutation probability so that the genetic operator can change the probability as required,which will greatly improve the global convergence of the algorithm.Add the algorithm to the PID control ball-beam system for simulation and experimentation The results prove that the control effect is more accurate than the conventional PID and genetic algorithm PID control.