Combustion Optimization Based on RBF Neural Network and Multi-Objective Genetic Algorithms

摘要

Coal-fired boiler operation is confronted with two requirements to reduce its operation cost and to lower its emission. In this paper, a model for boiler efficiency and a model for NO_x emission are set up respectively by RBF neural network. In order to obtain more accurate models without trying repeatedly, GA is introduced to optimize the parameter of RBF network. Then Non-Dominated Sorting Genetic Algoritth-m-II is employed to perform a search to determine the optimum solution of boiler operation after we obtain boiler combustion model. Eexperimental results prove that the method proposed in this paper can improve boiler efficiency and reduce NO_x emission obviously. Through analysis, we can see this method is better than the traditional method which uses weights to combine boiler efficiency and NO_x emission in one objective function.