梯级水库优化调度中入库径流预报效益研究

Benefit research of flow prediction in optimal operation of cascaded reservoir

建立基于入库径流预报的长短期耦合梯级水库电力优化调度模型,并以雅砻江锦屏一二级水电站为例,探讨了该电站入库径流量的预报效益。结果表明,利用所建模型准确地获取锦屏梯级电站发电效益、预见期、预报精度三者间的量化关系,即预见期越长精度越高,梯级电站平均年发电量越高,据此提出锦屏梯级水库入库径流预报的要求是预见期不应低于5d,预报精度平均相对绝对误差应小于0.15。

An optimal operation model of hydropower stations in the cascaded reservoir was developed based on the coupling of long-term and short-term prediction models of flow into the reservoir.The model was applied to the first-and second-grade hydropower stations in the cascaded reservoir of Jinping,and the prediction benefit of flow into the reservoir was discussed.The results showed that the model can provide the quantitative relationship between power benefits,prediction time,and prediction accuracy of the hydropower stations in the cascaded reservoir of Jinping.In general,longer prediction time leads to higher prediction accuracy and greater power outputs.It is suggested that the prediction time of flow into the cascaded reservoir of Jinping should not be less than 5 days and the mean relative absolute error of prediction should be less than 0.15.