An application of neuro-like fuzzy method to gas turbine NO/sub x/ concentration prediction problem

Applies simplified fuzzy reasoning models to the gas turbine NO/sub x/ concentration prediction problem. First, a simplified fuzzy reasoning model and its learning algorithm are reviewed. The original simplified fuzzy reasoning model, where symmetrical triangular membership functions are adopted, is improved. The learning algorithms of several simplified fuzzy reasoning models are applied to a, data set obtained from a commercial combined cycle power plant. Using the other data sets, the identified models are compared with each other in order to check the generality.