Behavior modeling of uncertain dynamic systems with nonlinearity

This paper presents a new approach to a behavior modeling of a system with nonlinearity and uncertainty. Based on behavior modeling, a procedure for behavior identification and classification is developed. Consequently, a fuzzy logic controller (FLC) can be designed according to definite types of behaviors. The key advantage of this approach is that an analytical model of the dynamics is not a necessity. In order to do this, we first choose a set of second order systems as the original “pattern behavior”; then we use optimized FLCs to control the original behavior; after that, we apply these FLCs to control nonlinear systems. In process operation, the behavior of a system with uncertain dynamics are identified and classified, and their corresponding optimal fuzzy logic controllers are determined by mapping the behavior into the original “pattern behavior”. To demonstrate the effectiveness of the proposed approach, we use it to control the PUMA 562 robot system