基于PSO算法与BP神经网络模型的泥石流流速预测

打开文本图片集

关键词：PSO算法；BP神经网络；泥石流；平均流速；预测模型

中图分类号：P642.23 文献标志码：A 文章编号：1003-5168(2025)19-0116-06

DOI:10.19968/j.cnki.hnkj.1003-5168.2025.19.022

Prediction of Debris Flow Velocity Based on PSO Algorithm and BP NeuralNetwork Model

CHEN Tao (School of Civil Engineering, Chongqing Three Gorges University, Chongqing 404120, China)

Abstract:[Purposes] To clarify the relationship between the disaster-causing factors of debris flow and its average flow velocity,and to improve the prediction accuracy and computational effciency for debris flow.[Methods] The PSO algorithm was used to optimize the BP neural network model.Based on the analysis and processing of fundamental samples from the Jiangjia Gully debris flow,a network prediction model for the average flow velocity of debris flows was established.[Findings] The unstable thickness layer and gradient are the decisive influencing factors for the average flow velocity of debris flow,and there exists a complex coupling relationship among these influencing factors.By considering the coupling relationships among the influencing factors,the PSO-BP neural network model achieves a prediction accuracy for the average flow velocity of debris flow as high as 94.007% . [Conclusions] The values predicted by the PSO-BP neural network model are closer to the actual measured values and align with the actual situation of the study area.The PSO-BP neural network model can provide a scientific basis for predicting the average flow velocity of debris flows.

Keywords: PSO algorithm; BP neural network; debris flow; average flow velocity; prediction model

0 引言

泥石流是一种常见的地质灾害，往往在持续降雨或极端降雨时突然爆发。（剩余8204字）