Fuzzy inversion and rule base reduction

This paper proposes a new design method based on linguistic model inversion and fuzzy rule reduction using singular value decomposition. Firstly, a piecewise linear fuzzy approximation of the controlled plant is identified by measurement. Secondly, the linear cells of the fuzzy model is inverted to achieve a controller. The inversion increases the fuzzy rule base. Thirdly, the redundant or small weighted information are removed from the fuzzy rule base. Experimental results of a transputer controlled single-degree-of-freedom motion control system are presented. The experimental system consists of a conventional DC servo gear motor with encoder feedback and variable inertia load coupled by a relatively rigid shaft.