A partitioning strategy of multiple model adaptive control for interval systems

As a practical design method, we present a multiple model adaptive control(MMAC) for linear interval systems with real coefficient. In conventional MMAC, the identification models are generally composed of a set of subsystems which is generated by dividing each uncertain parameter of the plant model equally. Hence it results in large number of models, so that the computation burden becomes very heavy. This paper aims to propose a new method that can reduce the number of such a multiple model. The key idea is based on the fact that the sensitivity of the time response to plant parameter changes extremely depends on how the closed loop dynamics is designed. After selecting a set of nominal parameters within given interval, a nominal controller is first designed and then the parameter perturbation effects are evaluated by using sensitivity function methods. From these, we can determine the numbers as well as the parameters to be partitioned, so that satisfy the performance specifications. The proposed method is examined by simulations on a two-phase servomotor example. Copyright (C) 2000 IFAC.