梯级水电站群超短期入库流量滞时时间序列预报方法

超短期预报调度对于实现短期调度和实时调度无缝衔接、降低调度风险、提高水电调度水平意义重大。而实际超短期预报调度面临着有效预报信息短缺、实时采集信息错报漏报、支流小水电调节、模型通用性差等一系列问题,为此,提出了一种梯级水电站群超短期滞时时间序列预报方法,首先分析了上游电站历史出库和下游电站历史入库流量,利用相关系数找出最强滞时流量匹配关系,从而将上游电站出库流量从下游电站入库中分解出来,还原出有效的区间流量,再通过区间流量逐日化进行时间序列建模,最后与上游电站滞时出库合成为下游电站预报入库流量。以云南澜沧江干流梯级电站为例,取得了良好的预测效果,从而验证了该方法的合理性、准确性与通用性。

Ultra-short Term Inflow Delay Time Series Forecasting Method for Cascaded Hydropower Stations

The ultra-short forecasting operation is great significance for realizing the seamless connection of short-term scheduling and real-term scheduling, reducing dispatch risk and improving hydropower operation level. Currently, the ultra-short forecasting operation faces the series problems, such as shortage of valid forecast data, mistaken or emission data collection, re-dispatch of small hydropower, and the poor generality of forecasting model, et al. Therefore, this paper proposes an ultra-short term delay time series forecasting method for cascaded hydropower stations. First, the upstream history outflow and downstream flow of each hydropower plant were analyzed to solving the rational delay time and flow relationship. Second, the upstream outflow was decomposed from the downstream inflow to obtain the effective inter-flow. Finally, a time series model of forecasting downstream flow was built by the daily interval flow. Taking Langcangjiang as an example, this model has achieved good effects, and the rationality, accuracy, and generality of the method has been verified.