摘要
针对BP网络存在的缺点 ,有多种改进方法 .本文在文献 [3]的基础上 ,从算法和网络结构设计方面又进行了综合改进 ,不仅加快了网络的收敛速度 ,而且优化了网络的拓扑结构 ,从而增强了BP神经网络的适应能力 .将新改进的BP网络用于成都市消费预测 ,取得令人满意的效果 .
There are many kinds of improvement methods for the shortcoming of BP neural network. The synthetical improvement of algorithm and network structure is carried out on the foundation of document[3]. It not only improves the convergent rate but also improves the network structure, and increases the adaptive ability of BP neural network. Application to forecast the consumption of Chengdu City has acquired good impact.
出处
《吉林化工学院学报》
CAS
2000年第4期48-51,共4页
Journal of Jilin Institute of Chemical Technology
