Hydropower station runoff and associated source load power annual scene simulation and prediction method

Abstract

The invention discloses a hydropower station runoff and associated source load power annual scene simulation and prediction method. The method comprises the following steps: step 1, collecting historical daily runoff data of a plurality of schedulable hydropower stations in a certain area and historical daily power data of external associated energy and loads; 2, clustering the data in ten days by using a self-organizing mapping neural network to form a runoff volume typical ten-day scene and a source load power ten-day scene; 3, establishing a ten-day state transition probability model of the target hydropower station runoff, and establishing a multi-scene condition probability model between the target hydropower station runoff and the associated source load power and between the target hydropower station runoff and other schedulable hydropower station runoff; 4, randomly simulating a source load power and schedulable hydropower station runoff time sequence scene in the future year, and aggregating into a typical scene for simulating the annual time sequence of the source load power and the schedulable hydropower station runoff through a scene reduction method; and 5, predicting the occurrence probability of each typical scene.

Description

technical field [0001] The invention relates to the technical field of stochastic simulation and prediction of hydropower station runoff scenarios, in particular to a method for stochastic simulation and prediction of annual time-series scenarios of hydropower station runoff and external associated source-load power. Background technique [0002] As an important part of the power system, hydropower stations must arrange their operation correctly and rationally, which is crucial to the safe, stable and economical operation of the entire power system. The uncertainty of incoming water from hydropower stations is not only affected by random changes in weather, but also by the human scheduling behavior of upstream hydropower. In order to effectively reduce the peak-shaving water abandonment of hydropower and improve the utilization efficiency of hydropower generation, the fluctuation and randomness of runoff and source-load power are fully considered in the decision-making of hy...

AI Technical Summary

Problems solved by technology

For example, in the "Long-term runoff study using SARIMA and ARIMA models in the United States", the autoregressive moving average model estimated from historical data and the related derived time series model are used to maintain the autocorrelation characteristics of runoff in each period of the series, However, this method is only suitable for short-term runoff simulation, and it is difficult to adapt to the probability distribution characteristics of long-term runoff change simulation, which has certain limitations

Images