引用本文: |
-
麦雄发,李玲,胡宝清.优化BP神经网络的快速细菌觅食算法[J].广西科学院学报,2011,27(3):221-223. [点击复制]
- MAI Xiong-fa,LI Ling,HU Bao-qing.BP Neural Network based on Fast Bacterial Foraging Algorithm[J].Journal of Guangxi Academy of Sciences,2011,27(3):221-223. [点击复制]
|
|
摘要: |
为了提高BP神经网络的全局收敛能力和预测精度,提出了混合PSO的快速细菌觅食算法优化BP神经网络(FBFABP)的方法,并以石漠化危险度预警为例进行验证。结果表明,通过使用粒子移动和简化细菌趋化操作,提高了算法的收敛速度和搜索全局最优值的能力。相对于其它神经网络训练算法,该方法具有较好的预测精度和泛化能力,具有一定的优势。 |
关键词: 细菌觅食算法 粒子群优化 BP神经网络 石漠化 |
DOI: |
投稿时间:2010-11-30 |
基金项目: |
|
BP Neural Network based on Fast Bacterial Foraging Algorithm |
MAI Xiong-fa1, LI Ling2, HU Bao-qing3
|
(1.School of Mathematical Sciences, Guangxi Teachers Education University, Nanning, Guangxi, 530001, China;2.School of Continuing Education, Guangxi Teachers Education University, Nanning, Guangxi, 530001, China;3.Faculty of Resource and Environmental Sciences, Guangxi Teachers Education University, Nanning, Guangxi, 530001, China) |
Abstract: |
In order to improve the global convergence speed and accuracy of forecasting,a new BP neural network based on fast bacterial foraging algorithm(FBFABP) combined with particle swarm optimization is presented.By means of the particle flying and simplifying the bacterial chemotactic action,the convergence speed and the capacity of searching global extremum of this algorithm are enhanced.Taking karst rocky desertification as example,the empirical results reveal that FBFABP some superiorities in forecasting ability and predicting accuracy comparing with the standard BP,BP with momentum factor,BP with Levenberg-Marquardt train method and PSOBP. |
Key words: bacterial foraging algorithm particle swarm optimization BP neural network karst rocky |