Due to its great flexibility in coping with ill-defined plants, fuzzy logic theory has found numerous successful applications in industrial engineering, e.g., pattern recognition 1, automatic control 2, and fault diagnosis 3. Generally, a fuzzy logic-based system with embedded linguistic knowledge maps an input (feature) vector to a scalar (conclusion) output 2. Most fuzzy systems deployed in practice are static. In other words, they lack the necessary internal dynamics, and can, thus, only implement nonlinear but non-dynamical input-output mappings. This disadvantage greatly hinders their wider employment in such areas as dynamical system modeling, prediction, signal filtering, and control. Inspired by the idea of utilizing linguistic information, we have proposed two kinds of dynamical fuzzy systems, Linguistic Information Feed-Back-based Dynamical Fuzzy System (LIFBDFS) 4, and Linguistic Information Feed-Forward based Dynamical Fuzzy System (LIFFDFS) 5. Instead of crisp system output, conclusion fuzzy membership function is fed back and forward locally with adjustable parameters in the LIFBDFS and LIFFDFS, respectively.
Uploading links
Rapid links
No comments:
Post a Comment