PID模糊控制器初始化的研究

Study on initialization of PID type fuzzy controller

为了克服线性PID控制器的缺点,把线性PID控制器转化为PID型模糊控制器,如PI,PD,PI+D,PI+I以及PI+PD.对于不同的实际线性度量,大多数PI和PID模糊控制器事实上是非线性的PI(PID)控制器.对于每个输入变量至少定义三个梯形或三角型输入模糊集合,PI(PD)控制器是非线性的PI(PD)控制器,其P增益,I增益和D增益都会随着被控系统的输出发生变化.对于采用乘积T范数估计模糊规则的前提部分,而且只有Mandani的乘积T范数的模糊蕴含同Zadeh与算子(如取小)结合才能给出与实际相符的控制效果.采用增量式作为控制器的输出来实现PID控制器上的模糊模拟,并在此基础上,推导了PID型模糊控制器初始化算法公式.

In order to overcome the shortcomings of the linear PID controller, the linear PID controller is transformed into a PID type fuzzy controller, such as PI, PD, PI ＋ D,PI ＋ I and PI ＋ PD. For different actual linear metrics, most PI and PID fuzzy controllers are in fact non - linear PI （PID） controllers. The PI （PD） controller is a non - linear PI （PD） controller that defines at least three trapezoidal or triangular input fuzzy sets for each input variable. The P gain, I gain, and D gain will vary with the The output changes. For the premise of using the product J＇-norm to estimate the fuzzy rules, and only the fuzzy implica-tion of the product T＇-norm of Mandani and the Zadeh and the operator （ such as taking the small） , can give the control effect that is consistent with the actual. Adopting increment as the output of the controller to realize the fuzzy simulation on the PID controller, based on which the formula of PID algorithm initialization algorithm was deduced.