摘要
基于BP人工神经网络算法的基本原理,采用水库水位与出力双决策控制,建立了溪洛渡、向家坝两库联合调度函数的BP人工神经网络模型。模拟调度结果表明,该模型能更好地映射调度函数中各变量之间的非线性关系,优化运行轨迹的拟合效果明显提高。
Based on BP artificial neural network theory, double decision control of reservoir water level and power is adopted to establish joint operation function model of Xiluodu and Xiangjiaba reservoirs. Simulated operation results show that the BP neural network model can transform the complicated nonlinear relationship between input and output variables of optimal operating rules for cascade reservoirs; the fitting effect of optimal operation track is obviously improved.
出处
《水电能源科学》
北大核心
2012年第12期48-51,共4页
Water Resources and Power
基金
国家科技支撑计划基金资助项目(2008BAB29B09)
国家自然科学基金资助重点项目(50539140)
关键词
调度函数
BP人工神经网络模型
多元线性回归
溪洛渡
向家坝
dispatching function
BP neural network
multiple linear regression
Xiluodu reservoir
Xiangjiaba reser- voir