Calibration of fuzzy controllers

Calibration of fuzzy controllers

This example shows the m code for calibrate the plant G3(s) using the Social Spider Optimization (SSO) algorithm as it’s described in the article: “Evolutionary calibration of fractional fuzzy controllers”
The needed files are:
• iniglob_var.m: initialize the variables to use among all the functions.
• SSO.m, FeMove.m, MaMove.m, MaMove.m, Mating.m, Survive.m: these files contains the SSO algorithm and its operator functions.
• fitPIDme7: calls the simulation of the control scheme and calculates the cost function.
• fitPIDme2.m: calculates and plots the cost function at the end of the algorithm.
• pdifractionalfuzzycontrol3.slx: simulates the control scheme using Simulink.
• FUZZIFRAC.fis: contains the fuzzy rules to execute in the control scheme.
Steps to execute the calibration algorithm:
In Matlab being in the same folder where are located the files execute the function iniglob_var and after that execute SSO(spidn,itn) where spidn is the total population number, and itn is the total iteration to realize, at the end of the algorithm it will return the optimal parameters for the plant and wil plot the control response to the step response.
Note: It is necessary to add FOMCOM toolbox to the path for the algorithm to work.
The complete method has been published in:
Erik Cuevas, Alberto Luque, Daniel Zaldívar, Marco Pérez-Cisneros, Evolutionary calibration of fractional fuzzy controllers, Applied inteligence, In press.
https://link.springer.com/article/10.1007/s10489-017-0899-y