Controllers Design For Ball And Beam System Based On Adaptive Genetic Algorithm

With the development of modern industrial process continuously, control science also faces more and more complicated problems, for this reason many advanced control theory and method have been widely used in practical engineering. As one of them, the genetic algorithm has already become research focus of engineering science and social science nowadays. This thesis mainly takes advantage of ball and beam system to test and verify the control effect about traditional control algorithm, the simple and adaptive genetic algorithm. Finally the results will be realized by simulation and experiments, separately.Firstly, some important theory and design method required have been elaborated. Full theory preparation has been done about the specific design and simulation verification of ball and beam controllers for the next step. Here, it accounts the definition, simple principle, mathematical theory and the simple implementation techniques and so on,, then the adaptive genetic algorithm is proposed and detailed, which shows better optimizing performance, based on simple genetic algorithm.Secondly, during this design process, at first based on the analysis for system control structure,I simplify and create approximate mathematical model, including mechanical part, angle conversion part and DC motor part, then gain transfer function and state space model of ball beam system, and analyze controllability of the system. Thirdly, the specific design of adaptive genetic PID controllers is carried out. On the basis of created mathematical model of ball and beam system, I take full use of related theory and methods of PID control and adaptive genetic algorithm, then the design and MATLAB process simulation are realized for conventional, simple genetic and adaptive genetic PID controllers successfully, and through analysising and comparing the results of simulation and experiments, the PID parameters optimization about adaptive genetic algorithm is applied in ball and beam system, which has the better control effect than the others.Finally, some investigation and research works have been done by adopting LQR control for problem of parameters turning and optimization, on the basis of LQR control and simulation for system, simple genetic and adaptive genetic algorithm is introduced to turn the parameters of LQR and go on MATLAB simulation.