Multimodel System Identification Based on New Fuzzy Partitioning Similarity Measure

The problem of identifying unstructured nonlinear systems is generally addressed on the basis multi-model representations involving several linear local models. In present work, models are combined to get a global representation using incremental fuzzy clustering. main contribution novel vector similarity measure defined in System Working Space (SWS) that combines angular deviation and usual Euclidean distance. Such combination makes new metric highly discriminating leading better partitioning operating space providing, thereby, higher accuracy model. developed method first evaluated by performing model (LLM) based identification academic benchmark multivariable system. Then, performances experimental tropospheric ozone data. These evaluations illustrate supremacy over standard Euclidian-distance approach.