Fuzzy modelling through logic optimization

This study concerns a new approach to fuzzy model identification. Primarily focusing on the core of the model, we propose a two-phase design process realizing adaptive logic processing in the form of structural and parametric optimization. In recognizing the fundamental link between binary and fuzzy logic, effective structural learning is achieved through established methods in logic minimization. This underlying structure is then augmented with fuzzy neural networks in order to learn the finer details of the target system´s behaviour. The combination of a logic-driven architecture with this novel hybrid-learning scheme helps to develop transparent and accurate models while maintaining excellent computational efficiency.