粒子群优化BP神经网络的滑坡敏感性评价

滑坡敏感性评价是地质灾害预测预报的关键环节。针对BP神经网络易陷入局部最小值、收敛速度慢等问题,该文以三峡库区秭归县境内为研究区,采用粒子群优化(PSO)算法对BP神经网络的初始权值和阈值进行优化,构建PSO-BP神经网络滑坡敏感性预测模型,实现研究区滑坡敏感性评价。采用受试者工作特征曲线分析模型预测精度,得到PSO-BP神经网络预测精度为0.931,预测结果与实际滑坡总体空间分布具有良好的一致性,且预测能力优于BP神经网络。实验结果表明,PSO-BP神经网络耦合模型在实现滑坡敏感性评价上具有理想的预测精度和良好的适用性。

Landslide susceptibility assessment based on PSO-BP neural network

Landslide susceptibility assessment is of great importance in predicting and forecasting geological hazards.This paper presented the landslide susceptibility assessment on the Zigui County of the Three Gorges using a particle swarm-optimized BP neural network.The particle swarm optimization algorithm (PSO) was used to optimize the initial weight and threshold parameters of BP neural network.The analytical results were validated by comparing them with known landslides using a receiver operator characteristic curve,and the particle swarm-optimized BP neural network model has a higher accuracy than BP neural network model,with an area ratio of 0.931.The validation results displayed sufficient agreement between the predicted results and the spatial distribution of existing landslides,which showed that the proposed method has higher prediction accuracy and good suitability in landslide susceptibility assessment.