Evolving Fuzzy Models: Incremental Learning,.

Nowadays data-driven models become more and more an essential part in industrial systems for application tasks such as system identification and analysis, prediction, control or fault detection. Data driven models are mathematical models which are completely identified from data, which can be available in form of offline data sets, most commonly stored in data matrices, or in form of online measurements. Data-driven models possess the nice property that they can be built up generically in the sense that no underlying physical, chemical etc. laws about the system variables have to be known. Whenever measurements are recorded online with a certain frequency, usually the models should be kept up-to-date, especially when new system states occur during online production processes. This requires an adaptation of model parameters and an evolution of model structures with incremental learning steps, as a complete rebuilding from time to time with all recorded measurements would not terminate in real-time. The book addresses the on-line evolution of fuzzy models and underlines its necessity by concrete application examples from on-line quality control systems.

Edwin Lughofer, Dr. Techn.: Post-Doc at the Department for Knowledge-Based Mathematical Systems, Best Paper Award September 2006 for 'Process Safety Enhancements for Data-Driven Evolving Fuzzy Models' at the internat. GEFS conference 2006, recieved Royal Society Grant, publ. more than 20 papers, participated in EU-Projects AMPA and DynaVis.