Close Loop Identification for Combustion System by Recurrent Adaptive Neuro-Fuzzy Inference System and Network with Exogenous Inputs

Boiler-turbine is a multi-variable and complicated system in steam power plants including combustion, temperature and drum water level. Selecting control loops as a unique loop in order to identify and control the boiler as a whole unit is a difficult and complicated task, because of nonlinear time variant dynamic characteristics of the boiler. It is necessary to identify each control group in order to accomplish a realistic and effective model, appropriate for designing an efficient controller. Both the effective and efficient performance of the identified model during the load change is of major importance. Here, not all parts of the system should be considered as a unit part, if determining and effective and realistic model is sought. The combustion loop of the 320 MW steam power plant of Islam Abad, Isfahan is the subject. Due to the sensitivity and complexity of the system, with respect to its nonlinear and closed loop characteristics, the identification of the system is conducted through intelligent procedures like recurrent adaptive neuro-fuzzy inference system (RANFIS) and nonlinear autoregressive model with exogenous input (NARX). The comparisons of the findings with actual data collected from the plant are presented and the accuracy of the procedures is determined.