论文部分内容阅读
准确预测小水电富集地区小水电的发电能力是保证电网安稳运行、实现大小水电协调的重要措施。不同于大中型水电,小水电大多位于偏远山区,信息采集困难,管理薄弱,可用于发电能力预测的资料较少,难以利用和借鉴现有的大中型水电发电能力预测方法。文中结合小水电的实际情况,以地区小水电整体为对象,提出了耦合偏互信息的小水电发电能力预测方法。该方法以BP神经网络预测模型为手段,采用偏互信息方法筛选显著影响小水电发电能力的预报因子,并结合气象预测系统(CFS)的气象预报信息作为输入,实现贫资料地区小水电发电能力预测。最后,以云南小水电富集的德宏和大理为实例研究验证了所述方法的有效性。
Accurately forecasting the generating capacity of SHP in the area rich in small hydropower is an important measure to ensure the stable operation of the power grid and to achieve the coordination of large and small hydropower stations. Unlike large and medium-sized hydropower plants, small hydropower plants are mostly located in remote mountainous areas, where information collection is difficult and management is weak. There are few data available for power generation capacity forecasting and it is difficult to utilize and draw lessons from existing large and medium hydropower generation capacity forecasting methods. According to the actual situation of SHP, aiming at the overall situation of SHP, a prediction method of SHP capacity is proposed. This method uses the BP neural network prediction model as a measure to select the forecast factors that significantly affect the power generation capability of SHP by using the mutual information method. Combined with the meteorological forecast information of the meteorological forecast system (CFS), the method can achieve the small hydropower generating capacity prediction. Finally, the case studies of Dehong and Dali enriched with SHP show that the method is effective.