A modified self-organizing controller for real-time process control applications

The paper proposes a systematic approach for an analytic formulation of a modified self-organizing fuzzy logic controller (SOFLC) that can generate control rules on-line and modifies them to changing process conditions. The paper also presents the real-time application of SOFLC for process control and demonstrates its ability to operate over a wide range of process variables (nonlinearity and parameter variations). A comparison in terms of steady-state error, setting time and transient response is made with a proportional plus integral (PI) controller, a simple FLC and a modified SOFLC. Experimental results show that modified SOFLC has a performance slightly superior to that of a PI and FLC controllers. (C) 1998 Elsevier Science B.V. All rights reserved.