摘要
针对二滩水电站的实际径流特性和水电站发电调度的要求,应用季节性自回归模型和人工神经网络模型对二滩水电站的月径流、汛期分段和年径流预报进行研究。结果表明,这两种模型对二滩水电站的月径流预报、汛期定性预报均达到了一定精度,可为二滩水电站优化调度的径流输入提供参考依据,尤其是AR(P)模型的非汛期月径流预测和BP模型年径流预测结果可在实际运行中使用。
This paper aims at the characteristics of the real runoff and the requirement of power generation dispatching of hydropower station, the seasonal autoregressive model and artificial neural network model are used to study the forecast of the monthly runoff, division of the flood season and annual runoff of Ertan hydropower station. The results show that both models possess a good precision for the prediction of the monthly runoff and the qualitative runoff forecast of the flood season in Ertan hydropower station, which provides the reference basis for input stream in optimal regulation of Ertan hydropower station, especially the prediction results of the monthly runoff in non-flood season based on AR(P) model and annual runoff based on BP model can be applied in practical operation.
出处
《水电能源科学》
北大核心
2009年第1期5-9,共5页
Water Resources and Power
基金
国家自然科学基金委员会、二滩水电开发公司雅砻江水电联合研究基金资助项目(50579095)
关键词
二滩水电站
季节性自回归模型
人工神经网络
中长期径流预报
Ertan Hydropower Station
seasonal autoregressive model
artificial neural network
medium and long term runoff forecasting