摘要
为提高径流预报精度,采用单相关系数法挑选预报因子,建立了基于遗传算法的参数投影寻踪回归径流预报模型,利用该模型对雅砻江二滩水电站月平均流量进行了预报。结果表明,与BP神经网络模型预报结果相比,投影寻踪回归模型具有更好的预报结果和更高的预报精度。
In order to improve the runoff forecasting accuracy, single correlation coefficient method is used to select forecast factors and the runoff forecasting model of parametric projection pursuit regression based on genetic algorithm is established. Then the model is applied to average monthly runoff forecasting of Ertan hydropower station in the Yalong River. Compared with the BP neural network, the results show that the proiection pursuit regression model has better prediction effects and higher forecast accuracy.
出处
《水电能源科学》
北大核心
2013年第9期20-23,共4页
Water Resources and Power
关键词
径流预报
投影寻踪回归
遗传算法
BP神经网络
因子挑选
runoff forecast
projection pursuit regression
genetic algorithm
back propagation neural network
fac-tor selection
