Research And Application Of A Fuzzy Control Method That Can Avoid Rule Explosion

The intelligent control system is the heated topic in the controlling field.In this field,the fuzzy control plays an import role.The traditional math method can’t describe the model about fuzziness and uncertainty,because its object is the accurate model.However,the appearance of fuzzy mathematics solves this problem well,and presents the theoretical basis for the fuzzy control field.Fuzzy control system depends on the experience of expert and a large number of experiments.Comparing with the traditional control,fuzzy control is applied to the controlled object which is difficult to give accurate mathematical modeling.Although the fuzzy control system has the very mature control theory foundation,and the mathematics foundation,but in the industrial control field,experiments still face the two problems.The first is the fuzzy control system must have two parts of the fuzzification and defuzzification,because most of the controlled object input and output values are accurate,and the fuzzy controller can only handle fuzzy values,so these two parts in the traditional fuzzy controller is not governable,thus increasing the difficulty of industrial operations and costs.The second is the problem of rule explosion.The explosion of rules is that there are too many fuzzy conditional statements and the fuzzy reasoning mechanism can not afford a large number of rules reasoning.At the same time the rules of the explosion is fuzzy control chip cost is the main reason.This article aims at the two existing problems mentioned above,gives a new design idea of fuzzy control system,proposes a granular function method to replace the traditional fuzzy control algorithm,and experiments on inverted pendulum and intelligent lighting of the fuzzy control system of the particle function verification.This paper put forward the granulation structure is applied to the fuzzy conditional statement,and the original fuzzy inference process of fuzzy control system is replaced by granular function.The mapping relationship is obtained by the fitting of the regular particle points to obtain an accurate response function that is obtained by human logic.Therefore,this control method is a new idea.For the newly proposed algorithm,the particle response function solves the rule explosion problems.The rule explosion problem caused by too many fuzzy rules in fuzzy control system can be solved well by the granular function.In this paper,fuzzy control of inverted pendulum simulation experiments and fuzzy control of intelligent lighting simulation experiments.In this paper,the fuzzy control of fuzzy control of inverted pendulum simulation experiment and fuzzy control of intelligent lighting simulation experiments are separately carried out.In the inverted pendulum simulation experiment,the fuzzy control of inverted pendulum system,5 order response function of the inverted pendulum system,2 order response function of the inverted pendulum system and 1 order response function of the inverted pendulum system are respectively compared with each other.The results show that inverted pendulum system based on granular function has the same control effect with the one based on traditional fuzzy control system.It effectively reduces the complexity,and also solves the rule explosion problem.In the experiment of intelligent lighting simulation,experimental contrast between the traditional fuzzy control system and the fuzzy function control system with granular function are respectively carried out.The results show that the controller of the intelligent lighting fuzzy control system improved by the fuzzy control algorithm of granular function can achieve the same control effect on demand as before,and effectively avoids the rule explosion problem.The innovations of this thesis are as follows: 1.The fuzzy control uses the accurate response function based on the fitting of the granule function to realize the control,eliminating the fuzzification and defuzzification process and effectively reducing the complexity of the control.2.The application of granular function in this experiment solves the rule explosion problem.3.Apply the new algorithm to inverted pendulum simulation experiment and intelligent lighting simulation experiment,which simplifies the control difficulty and cost of these two systems.