径向基神经网络在水轮发电机组耗流量计算中的应用

APPLICATION OF RADIAL BASIS FUNCTION NEURAL NETWORK TO THE CALCULATION OF FLOW CONSUMPTION OF WATER-TURBINE GENETATOR UNITS

传统的水电站厂内经济运行的耗流量计算方法需要通过手工绘制机组效率特性曲线 ,适应性差 ,不利于在线操作。文中提出跳过绘制机组效率特性曲线 ,直接建立水头、出力与耗流量的径向基神经网络模型 ,以计算任意水头下的耗流量曲线的模型 ,并应用到某水电站。该模型具有快速、准确的特点 ,可在线训练、在线应用 。

A method based on radial basis function (RBF) neural network for calculating flow consumption of water turbine generator sets is proposed. First, the RBF neural network is trained with test data of a water turbine generator unit at a limited head, and then the trained RBF neural network is used to calculate flow consumption of the unit at any head quickly and accurately. By the application of this method to the inner plant economical operation of the hydropower station, the flow consumption can be introduced to the optimization of inner plant economical operation directly, and thus, the complicated efficiency calculation of water turbine generator unit is avoided. The analysis shows that the method has good adaptability and bright future in online optimization.