Neural Cascaded with Fuzzy Scheme for Combustion Control of Utility Boiler

A novel design for combustion control of utility boiler using neural cascaded with fuzzy controller is proposed. The main objective of the combustion controller in a thermal power plant is to regulate fuel and air in proper ratio to maintain the desired steam pressure at the turbine inlet, irrespective of the changes in steam demand. The existing control schemes have difficulty to cope up with inherent time delay, nonlinearity due to uncertainty of the combustion process and frequent changes load calorific value of the fuel etc. This paper presents the design of neuro controllers to regulate fuel, cascaded with fuzzy controller to control air for combustion process. An experimental setup is fabricated in the laboratory for fuel and air control and real time simulation studies were carried out using PID, and neural cascaded with fuzzy schemes. The performances of proposed schemes are evaluated by simulation and the results are compared with conventional controllers using real time data obtained from the thermal power plant. The advantages of the proposed scheme over the existing controllers are highlighted.