| BP算法的改进及其在库存数量预测中的应用 |
| |
| 引用本文： | 王红霞,僧德文.BP算法的改进及其在库存数量预测中的应用[J].浙江水利水电专科学校学报,2006,18(4):34-37. |
| |
| 作者姓名： | 王红霞 僧德文 |
| |
| 作者单位： | 浙江水利水电专科学校,浙江,杭州,310018 |
| |
| 基金项目： | 浙江省教育厅基金资助项目(20060001、21205)，浙江省水利厅基金资助项目(RC0605)，浙江省高校青年教师基金资助项目(21223) |
| |
| 摘 要： | 针对BP算法收敛速度较慢、局部极值等缺点,提出了一种改进的BP算法.根据训练误差改变转移函数的惩罚因子并且对学习步长作自动调节.实践结果表明,改进后的BP算法可大大提高算法的函数拟合度和收敛度,减少与实际值间的误差.
|
| 关 键 词： | 前馈神经网络 BP算法 学习步长 惩罚因子 |
| 文章编号： | 1008-536X（2006）04-0034-04 |
| 收稿时间： | 2006-09-20 |
| 修稿时间： | 2006-09-20 |
BP Algorithm and Its Application on Store Amount Prediction |
| |
| WANG Hong-xia,SENG De-wen.BP Algorithm and Its Application on Store Amount Prediction[J].Journal of Zhejiang Water Conservancy and Hydropower College,2006,18(4):34-37. |
| |
| Authors: | WANG Hong-xia SENG De-wen |
| |
| Affiliation: | Zhejiang Water Conservancy and Hydropower College, Hangzhou 310018, China |
| |
| Abstract: | BP Algorithm has some disadvantages,such as local minima,weights sensitivity of initial value,total dependence on gradient information,etc.An improved algorithm is proposed,which can train a neural network by changing the gradient of Sigmoid function according to the errors in training and adjusting neural network's learning rate automatically.The simulation results indicate that the algorithm proposed can improve the curve fitting and convergence,and reduce the discrepancy between the fact data and the measure data. |
| |
| Keywords: | feed forward neural networks BP algorithm learning rate punishment factor |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
