Study on Runoff Prediction of Representative Stations in Qiantang River Basin based on Particle Swarm Optimization and Support Vector Machine

Abstract

Qiantang River is the largest river in Zhejiang Province and the mother river of Zhejiang. The middle and upper reaches of the basin are the Jinqu basin of Zhejiang Province. On both sides of the estuary are Hangjiahu Plain and xiaoshaoning plain. It is a gathering place of population and economic factors in Zhejiang Province, with a population of 10.8 million. Frequent floods in the Qiantang River Basin have caused serious losses to people's lives, property and social economy on both sides of the river. The amount of water from Qiantang River is related to the harm of flood and waterlogging natural disasters, but also affects the utilization of water resources and the future social and economic development of Qiantang River area. This paper attempts to combine the support vector machine model with the global optimization search method PSO model. PSO algorithm is used to optimize the penalty factor C and kernel parameters of support vector machine to improve the accuracy of the model. The calculation shows that the calculation accuracy of this model is much higher than that of artificial neural network model. This model can be used to predict the runoff of representative stations in Qiantang River Basin.