Soft computing optimizer Intelligent Control System

The present invention relates to, for example, soft computing optimizer for designing a knowledge base (KB) for use in a control system for plant control, such as an internal combustion engine and suspension system for a motor vehicle. SC optimization apparatus includes a fuzzy inference engine based on the fuzzy neural network (FNN). SC optimizer selects the fuzzy inference system (FIS) configuration, and select the FIS configuration optimization method, selecting and generating a teaching signal. The user input and / or output variables, the type of fuzzy inference model (e.g. Mamdani, Kanno, etc. Tsukamoto), selects a fuzzy model including one or more of the pre-type membership function. Genetic Algorithm (GA) is used to optimize the input and output training signal and the language variable parameters. GA also fuzzy model, the optimal linguistic variables parameter, used to optimize the rule base by using the instruction signal. GA generates a nearly optimal FNN. Nearly optimal FNN can be improved by using a classical function based optimization process. FIS configuration found by GA is optimized by actual plant model fit function based on the response of the control plant. SC optimizer generally smaller than KB produced by the prior art, to produce a robust KB. .The