摘要
对于挑流泄洪雾化范围的BP神经网络模型,选用李家峡、二滩、漫湾和东江水电站的原型观测数据或泄流雾化计算数据作为训练样本进行学习训练,随后应用此模型对江垭大坝泄洪雾化范围进行了预测计算。通过与相应的原型观测资料比对分析,验证了模型的适用性、精确性,并指出了其应用的局限性。
Based on the BP neural network model , using the observation data or the corresponding computing data from Lijiaxia, Ertan, Manwan and Dongjiang Hydropower Station as the training samples, then perform the forecast of the range of atomized flow of Jiangya Dam. Through the contrast with the observation result, the BP neural network model is proved to be adaptable and accurate, on the other hand the limitation is pointed out.
出处
《水利科技与经济》
2009年第6期521-522,共2页
Water Conservancy Science and Technology and Economy
关键词
挑流泄洪
雾化
BP神经网络
预测
flip bucket
atomization
BP neural network
forecast
