Research of BP neural network based on improved particle swarm optimization algorithm

被引：3
作者：
School of Mechanical and Information Engineering, China University of Mining and Technology, Beijing, China [1 ]
不详 [2 ]
不详 [3 ]
机构：
[1] School of Mechanical and Information Engineering, China University of Mining and Technology, Beijing
[2] Software and Communication Engineering institute, Jiangxi University of Finance and Economics, Nanchang
[3] Information Engineering institute, Capital Normal University, Beijing
来源：
J. Netw. | 2013年 / 4卷 / 947-954期
关键词：
BPNN; Connection weight; Crossover operator; Metallogenic prediction; Mutation operator; PSO;
D O I：
10.4304/jnw.8.4.947-954
中图分类号：
学科分类号：
摘要：
The paper proposes an approach to optimize the connection weights and network structure of BP neural network (BPNN) which based on improved particle swarm optimization (PSO) algorithm. For each network structure, the algorithm generates a series of particles which consist of connection weights and threshold values, and selects the best network structure according to the improved PSO algorithm. Because the PSO algorithm is easy to fall into local optimums, the algorithm introduces crossover operator and mutation operator to heighten the ability of jumping the local optimums. Compared with the basic BP algorithm, the results show that performances of the improved PSO-BP algorithm are superior to it, and the paper applies this BPNN model to metallogenic prediction and give the detailed steps. © 2013 ACADEMY PUBLISHER.
引用
收藏
页码：947 / 954
页数：7
相关论文
共 27 条
  • [1] Suli Y.U., Qi X., The project evaluation of Private Equity on the basis of BP neural network, Journal of Tsinghua University (Science & Technology), 51, 12, pp. 1917-1920, (2011)
  • [2] Han L., Rui L., Huili Z., Comprehensive Evaluation Model of Soil Nutrient Based on BP Neural Network, The Chinese Society of Agricultural Machinery, 42, 7, pp. 109-115, (2011)
  • [3] Xiao-Ge Y., Jin H., Long-Qing S., Jiu-Chuan W., Lu Z., Shu-Cai L., Forecast of destroyed floor depth based on BP neural networks, Journal of China Coal Sciety, 34, 6, pp. 713-718, (2009)
  • [4] Jiang H., Ruan J., Investment risks assessment on high-tech projects based on analytic hierarchy process and BP neural network, Journal of Networks, 5, 4, pp. 393-402, (2010)
  • [5] Wei B., Zhao Z., Peng X., Spatial information based medical image registration using mutual information, Journal of Multimedia, 6, 3, pp. 236-243, (2011)
  • [6] Jia-Pei Y., Qiang L., Zheng L., Xiao-Lin Y., Research of improved BP algorithm based on self-adaptive learning rate, Computer Engineering and Applications, 45, 11, pp. 56-59, (2009)
  • [7] An G., Min Z., Research on self-adaptive learning rate of BP neural network, Science Technology and Engineer, 6, 1, pp. 64-66, (2006)
  • [8] Hui-Sheng Z., Wei W., A back-propagation algorithm with adaptive momentum factor, Journal of Dalian Maritime University, 34, 4, pp. 45-47, (2008)
  • [9] Wei Y., Qiandong N., Junji W., Research of relation between the initial weight and convergence of BP neural network, Proceedings of EPSA, 14, 1, pp. 20-22, (2002)
  • [10] Hao P., Qinglan H., A BP neural networks learning algorithm research based on particle swarm optimizer, Computer Engineering and Application, 16, 14, pp. 41-43, (2006)
123