The principal objective of our proposed research is to design, analyze and develop a neural network fuzzy adaptive controller model for complex processes capable of learning fuzzy control rules using process data and improving its control through on-line adaption. The learned improvement should be according to a performance objective function that provides evaluative feedback; this performance objective should be broadly defined to meet long-range goals over time. Fuzzy control has proven effective for complex, nonlinear, imprecisely-defined processes for which standard models and controls are either inefficient, impractical or cannot be derived. Hitherto, procedures for deriving fuzzy control, however, have been mostly as hoc heuristics. The learning ability of neural networks can be used to systematically derive fuzzy control and permit on- line adaption to optimize control. The operation of neural networks integrates very naturally with fuzzy logic. The neural networks will be designed and tested using simulation software and simulated data, followed by realistic industrial data to the extent possible. The statistical procedures of the learning process will be investigated and evaluated with standard statistical procedures (such as ANOVA, graphical analysis of residuals, etc.). The computational advantage of dynamic programming- like methods of optimal control will be used to permit on-line fuzzy adaptive control. Tests for the consistency, completeness and interaction of the control rules, as well as tests for stability, will be applied. Comparisons to other methods will be made so as to identify any advantages of the resulting model.
