Auto-tuning of fuzzy logic controllers for self-regulating processes

摘要

A normalised fuzzy logic controller is derived for the control of a normalised self-regulating process and fuzzy logic controllers for the control of processes which are deviated from the norm can be obtained by denormalising the normalised fuzzy logic controller. For processes with unknown system dynamics, biased relay feedback tests have been applied and first-order processes with dead time models are fitted to the system output responses. Based on the fitted first-order plus dead time model, a fuzzy logic controller is derived from the normalised fuzzy logic controller based on the approximated time delay, time constant and gain of the fitted model obtained from the biased relay test. For systems having large time delay, fuzzy logic is also included to compensate for the delay which will otherwise make the control loop less stable. Simulation and experimental results are included to demonstrate the effectiveness of the proposed scheme.