Prediction model of microwave calcining of ammonium diuranate using incremental improved back-propagation neural network

Yingwei LI 1,2),Bingguo LIU 1,2),Jinhui PENG 1,2)*,Wei LI 2),Daifu HUANG 3) and Libo ZHANG 1,2) 1) Faculty of Metallurgical and Energy Engineering,Kunming University of Science and Technology,Kunming 650093,China 2) Key Laboratory of Unconventional Metallurgy,Ministry of Education,Kunming University of Science and Technology,Kunming 650093,China 3) No.272 Nuclear Industry Factory,China National Nuclear Corporation,Hengyang 421002,China

摘 要：The incremental improved Back-Propagation （BP） neural network prediction model using the Levenberg-Marquardt algorithm based on optimizing theory is put forward,which can solve the problems existing in the process of calcinations for ammonium diuranate （ADU） by microwave heating,such as long testing cycle,high testing quantity,difficulty of optimization for process parameters.Many training data probably were offered by the way of increment batch and the limitation of the system memory could make the training data infeasible when the sample scale was large.The prediction model of the nonlinear system is built,which can effectively predict the experiment of microwave calcining of ADU,and the incremental improved BP neural network is very useful in overcoming the local minimum problem,finding the global optimal solution and accelerating the convergence speed.

关键词：