摘要
在分析有关铁路客运量预测方法的基础上,针对BP神经网络模型存在的不足,提出基于粒子群优化算法(PSO)优化BP神经网络的参数,即改进的PSO方法(IPSO)。以我国1990—2007年的铁路客运量为研究对象,确定输入样本和输出样本,以及训练集和测试集,建立基于IPSO的BP神经网络优化模型预测铁路客运量。预测结果表明,IPSO-BP网络的算法训练时间短,收敛速度快,预测精度高。
Based on analyzing the forecast method of railway passenger traffic volume, targeting with the deficiency of BP neural network model, this paper puts forward the improved PSO method (IPSO), which means optimizing parameter of BP neural network based on PSO method. Take railway passenger traffic volume in 1990--2007 as research object, the input and output example as well as training set and testing set were considered, and forecast model of railway passenger traffic volume could be forecasted by establishing BP neural network optimization model based on IPSO. The forecast result shows IPSO-BP network method has the advantages of short training time, fast convergence speed and high forecast precision.
出处
《铁道运输与经济》
北大核心
2011年第9期78-82,共5页
Railway Transport and Economy
基金
甘肃省自然科学基金(1014RJZA042)
关键词
铁路
客运量
BP神经网络
预测
Railway
Passenger Traffic Volume
BP Neural Network
Forecast