How to train ANFIS using a wide range of input/output data that comes from the conventional controller like PID?
And how to extract the data (the error and change of error) from pid controller and use them as input to the ANFIS controller?
PID controller can take one input at a time (like step input or ramp) and gives you output. I want to try different methods to give a range of inputs to the PID controller as a reference (like array input). so that the ANFIS would work well for the input, and can be trained like in step input (not for all step input of all values, just one input of value 1), for the step of different value like step input = 2, For the time being, I am learning how to use anfis for simple plant model that is (0.021/(0.000347s^2 +s)). For future use, I will use anfis as fuzzy controller to control the trajectory of the welding robot.
for instance, We have the position equation of the welding torch below with 100 data samples .
This equation indicates the position of the welding torch in x-direction
x=0.177 ΔL1-0.29 ΔL2+0.177 ΔL3+12.23
Via Python: https://github.com/twmeggs/anfis also look at Chapter ANFIS and Fuzzy Tuning of PID Controller for STATCOM to Enha...
Chapter Performance Comparison of PID and ANFIS Controller for Stabi...
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training