Medium and long term runoff forecasting method based on improved particle swarm optimization algorithm and support vector machine

Abstract

The invention discloses a medium and long term runoff forecasting method based on an improved particle swarm optimization algorithm and a support vector machine, which comprises the following steps: selecting historical data of a plurality of climate indexes and historical runoff volume of a drainage basin to be forecasted, and selecting forecasting factors from the historical data of the plurality of climate indexes and processing the forecasting factors, combining with the historical runoff volume of the drainage basin to be forecasted to construct a model data set, and dividing the model data set into a training set and a test set; performing stretching operation on the particles falling into local optimum to obtain an improved particle swarm optimization algorithm; obtaining an optimal parameter combination of the SVR based on an improved particle swarm optimization algorithm to establish an SVR forecasting model based on the improved particle swarm optimization algorithm, and training the forecasting model by using the training set; and comparing an output result obtained by inputting the test set into the forecasting model with real runoff data in the test set, and evaluating a forecasting effect of the forecasting model. The method has the advantages that the forecasting precision and generalization ability of the medium-and-long-term forecasting method are improved, and the problems of low forecasting precision and the like caused by easy falling into local optimum can be effectively avoided.

Description

technical field [0001] The invention relates to the technical field of hydrological forecasting, in particular to a medium and long-term runoff forecasting method based on an improved particle swarm optimization algorithm and a support vector machine. Background technique [0002] In recent years, machine learning algorithms represented by artificial neural networks and support vector machines have been gradually applied to medium and long-term hydrological forecasting. Among them, support vector regression (SVR) has shown excellent performance in small sample, nonlinear, high-dimensional regression prediction problems, using standard particle swarm optimization (PSO) to find the penalty coefficient C and insensitivity coefficient ε of SVR As well as the gamma parameter of the Gaussian radial basis kernel function, meteorological factors are used as predictors to identify the most relevant factors for watershed runoff, and the PSO-SVR model is used to learn the relationship ...

AI Technical Summary

Problems solved by technology

However, although the standard particle swarm optimization algorithm has a fast convergence speed, it is very easy to fall into the local optimal solution, and it is difficult to find a set of optimal parameters of SVR in the global scope.

Images