Self-tuning PID control using an adaptive network-based fuzzy inference system

This paper presents a self-tuning PID control algorithm using an adaptive network-based fuzzy inference structure (ANFIS). In particular, the design of a self-tuning PID learning-based optimum controller is introduced which can be applied to nonlinear as well as linear systems. A recursive adaptation scheme is employed for on-line implementation of the self-tuning PLD controller in which an exponential forgetting factor is used to weigh old data and both an error and control rate cost are used in the backwards pass. Results show that the method is a viable approach for tuning the parameters of a PID controller.