A parameter adjustment method for analog circuits based on convex fuzzy decision using constraints of satisfactory level

The adjustment problem of circuit parameters in analog circuits is formalized as a fuzzy mathematical programming problem, and a novel method is proposed to derive the acceptable parameters that satisfy the designer's requests, each of which is specified by a membership function and the permissible minimum satisfactory level. The search direction to adjust the parameters is determined by using a rule base in which the expert knowledge to improve the performance of the circuit is stored. The search step is determined by the golden selection method. The effectiveness of this method is checked by some experiments for the mu A709 operational amplifier circuit. From these experiments, it is found out that (1) it is useful to evaluate how much the circuit can satisfy the designer's requests at the satisfactory level, and (2) this method can derive a solution with the maximum satisfactory level even if there are trade-off relations between satisfactions.<
>