Inverted Pendulum Fuzzy Neural Control Based On Ga And Bp Algorithm

With the rapid development of science and technology, the change of modernindustry's control system is more and more complicated. It is much closer that thescientific development of control combines with development of the technology ofthe computer in addition. Because the control system is more and more complicated,the tradition control based on accurate mathematics model can not already get theideal control result. Intellectual control is an advanced control system of the humanintelligence of a kind of simulation. It is the control based on knowledge, and controlexperience and knowledge are regarded as the knowledge model of the controlledobject.Fuzzy system, neural network and genetic algorithm are thought as threeimportant fields with most prospect developments of artificial intelligence fields inthe 21st century. They form the so-called intelligence computing or call softcomputing.The control of inverted pendulum system is a typical example of application ofthe control theory. It's often used to testify the efficiency of novel control strategy.Moreover, due to the similarity of control of the inverted pendulum system to thecontrol of rocket, aircraft's aviation, the control algorithm and conclusion of thispaper will be instructive to solve other engineering problems.First this paper analyses the theories of fuzzy control and neural network control.On this basis, a fuzzy-neural network control structure is put forward. On thefoundation that the pluses and minuses of BP algorithm have been analyzed, thispaper introduces the genetic algorithm and optimizes the fuzzy-neural network bythe way of combining BP algorithm with genetic algorithm. And put experiments tothe inverted pendulum of Googoltech Company. It is proved that the control methodin this paper has much better control result.