Boilers are important components in Power systems. Boiler is quite complex and huge in size. The process happened in it is of multi-variables and the parameters are heavily coupled together. So its control and optimization are targeted as one of the most key Problems by researchers.In this paper, RBF Neural Network was selected as a tool for modeling according to the characteristics of boiler systems. A new design of RBF neural network is proposed. A dynamic K-means method based on Genetic algorithm is used to optimize the centers of the hidden units of RBF network. Genetic algorithm is used to train the weights of RBF network. So it enhanced the generalizing ability of RBF Neural Networks and thus improving its applicability.Taking boiler combustion process as an example, a RBF neural network is built up with the data on the spot, using the ameliorated algorithm. Genetic algorithms with real number encoding was used in optimization of the process based on the model gained, searching for the optimal in put parameters.
|