PID self-tuning control using a fuzzy adaptive mechanism

The authors present a novel proportional-integral-derivative (PID) self-tuning control using a fuzzy adaptive mechanism for regulating industrial processes. The essential idea is to parameterize a Ziegler-Nichols-like tuning formula by a single parameter α, then to use an online fuzzy inference mechanism to self-tune the parameter. A comparative simulation study on various processes, including processes with long dead-time and non-minimum-phase processes, shows that the performance of the scheme improves considerably, in terms of set-point and bad disturbance responses, over the PID controllers well-tuned using both the classical and the more recent refined Ziegler-Nichols formula