Medium and long-term runoff probability forecast method and system

Abstract

The invention provides a medium and long-term runoff probability forecast method and system. The method comprises the following steps: constructing a forecast factor set in a target region according to a forecast period, forecast factors and historical runoff data corresponding to the forecast factors; inputting the forecast factor set into a trained medium and long-term runoff forecast model to obtain a runoff probability forecast normal distribution result, wherein the trained medium and long-term runoff forecast model is obtained by training an improved support vector regression model based on a sample forecast factor set, and the improved support vector regression model is obtained by coupling a three-layer Bayesian inference framework and a support vector regression model; and performing data processing on the runoff probability forecast normal distribution result to obtain a runoff probability forecast result of the target region in the forecast period. According to the invention, the problem that a traditional data-driven model cannot generate medium and long term runoff probability forecasts is solved, and the uncertainty of future runoff is better reflected.

Description

technical field [0001] The invention relates to the technical field of medium and long-term runoff forecasting, in particular to a method and system for medium- and long-term runoff probability forecasting. Background technique [0002] Medium- and long-term runoff forecasts generally refer to runoff forecasts with a forecast period between 3 days and 1 year and beyond the confluence time of the basin, which can provide effective future runoff information and are useful in reservoir power generation planning, drought and flood control, and water resource planning management and comprehensive utilization. It plays an important role in the field of reservoir scheduling decision-making. Under the current climate change conditions, the amount of future water resources has great uncertainty, so medium and long-term runoff forecasts that can provide future information have received extensive attention. [0003] In the existing medium and long-term runoff forecast research, becaus...

AI Technical Summary

Problems solved by technology

[0004] However, in medium and long-term runoff forecasting based on data-driven models, it is often only possible to produce deterministic forecast results without considering the uncertainty of future runoff

But in fact, due to the continuous rise of greenhouse gas emissions such as carbon dioxide caused by human activities, the global climate has undergone significant changes in the past, and will show a more complex trend in the future. Future changes in water resources are more uncertain

In this case, reservoir scheduling based on deterministic runoff forecast results will bring greater risks to the practical application of runoff forecasting. Therefore, a method capable of generating probabilistic forecasts is needed to generate runoff probabilistic forecasts to reflect the future Uncertainty in runoff, thereby reducing the risk of reservoir dispatching based on runoff forecasts

Images