The Improvement and Application of Structure Optimization Algorithm Based on Flexible Neural Trees
Shouning Qu, Aifang Fu, Liu Zhao-lian, Chang Xiao-li · 2009
In this paper, the improved structural optimization algorithm of the flexible neural tree model is employed to select the parameters for the industrial production. With the highest accuracy and the shortest time to find the important parameters affecting the production so as to provide a theoretical support for the control of fluid industry production. In the period of learning of the flexible neural tree model, the evolution generation of algorithm is not a fixed value and the mean error rate is utilized to control the evolution generation. The flexible neural tree modelpsilas structure and parameters are optimized by the probabilistic incremental program evolution and the simulation annealing, respectively. The process of the decomposing furnace, which is one of the most important processes of the cement productions, is the object of this article. And it has been demonstrated that the given method is very effective.