Most possible qualitative behaviors of continuous systems

In this paper the approach of producing most possible qualitative behaviors is proposed. It contains fuzzy qualitative modeling, generating qualitative behaviors by simulation and searching the behavioral tree, to finally obtain one or several most preferential behaviors as the description of the system. The emphasis is put on the efficient search method with branch-bound- backtracking, evaluating and selecting fuzzy qualitative behaviors, which can be used in process monitoring and fault diagnosis, etc.