Dynamic Online Fuzzy Modelling of Nonlinear Systems
Modelling of complex non-linear systems using the process data is a challenging task. A lot of nonlinear techniques are used in systems modeling, but fuzzy logic based techniques proved to be very effective and efficient especially when the dynamics of the plant or system are complex and totally unknown. This work presents the fuzzy modelling of a cooling coil system using the input-output process data. The modelling is carried out in three steps. First, offline identification, in which, all the measured input-output data is processed at once to model the system. Second, the online identification, in which, the measured input-output data is fed sample by sample. Third, online dynamic modeling, in which, the estimated output is fed back to the model along with the measured input-output data.
Fuzzy modelling with self-learning evolving structure is also very efficient modelling technique and is of prime focus nowadays. This work also contains the modelling of cooling coil system using Evolving Fuzzy Models (EFMs).
Simulation results have been presented, respectively, which demonstrate the efficiency of evolving fuzzy models, and online dynamic modelling