智能控制智能控制智能控制 (66).pdf

《智能控制智能控制智能控制 (66).pdf》由会员分享，可在线阅读，更多相关《智能控制智能控制智能控制 (66).pdf（7页珍藏版）》请在得力文库 - 分享文档赚钱的网站上搜索。

1、Control Algorithm&Parameters TuningFrameworkWith PID and fuzzy controllers,the process of nonliner response canbe controlled.The fuzzy reasoning compensates for the PID controller outputdirectly.Therefore,the control rules are simpler than those whichadjust three parameters of the PID controller.The

2、 cycle of compensation by fuzzy reasoning is the same or lessfrequent than that of the PID controller.CharacteristicsThe fuzzy rules for this method is of the following form:Ri:if X1is A1iand and Xnis Anithen Yi=biFuzzy Reasoning SectionGiven an input(x1,x2,.,xn)at time t,the degree of matching in t

3、he premise for the j-th rule is calculated as follows:1122iiiinnwA(x)A(x)A(x)Membership&Control RulesTT:Y(t)-Yr,where Y(t)and Yr are therealandsetpointofthetoptemperature,respectivelyTT:Rate of top temperature changeStep 0:We set the PID parameters to correspond to those of a systemin steady state b

4、y a conventional method.Step 1:We set the cycle of fuzzy reasoning equal to that of anoperators compensation for the PID output.Step 2:We tune the parameters for the consequence of fuzzy rules.Wecheck that the output of the fuzzy controller equals that of anexperienced operator.Parameter Tuning Step

5、 3:We start to control the process with the fuzzy-PID hybridcontrol system and we tune the parameters of membership functions ofthe premise of fuzzy rules,step-by-step.Step 4:We shorten the cycle of the fuzzy reasoning until it satisfies thecontrol ability which we need to.Step 5:We made the Ti parameter of the PID controller smaller untilthe control ability which we need is satisfied.Parameter Tuning(ctld)