土石坝风险等级智能预测分析及模型优化研究  

Study on intelligent predictions and analysis of earth-rock dam risk levels as well as model optimization

作　　者：李炎隆[1] 张雨春 王婷[1] 殷乔刚 刘云贺[1] LI Yanlong;ZHANG Yuchun;WANG Ting;YIN Qiaogang;LIU Yunhe(State Key Laboratory of Eco-hydraulics in Northwest Arid Region,Xi'an University of Technology,Xi'an 710048,China)

机构地区：[1]西安理工大学省部共建西北旱区生态水利国家重点实验室,西安710048

出　　处：《水力发电学报》2024年第7期85-96,共12页Journal of Hydroelectric Engineering

基　　金：国家自然科学基金重点项目(52039008);国家杰出青年科学基金项目(52125904)。

摘　　要：大坝溃坝会造成大量的生命财产损失和巨大的环境破坏。精准快速确定土石坝风险等级,对于控制土石坝溃坝危害具有重要意义。本文采用K-最近邻(KNN)算法填补了数据库中大量缺失数据,引入遗传优化算法(GA)优化轻量级梯度提升机(LightGBM)超参数,建立了基于GA-LightGBM的土石坝风险等级快速预测模型。采用受试者工作特征曲线(ROC)、曲线下面积(AUC)值等其他评价指标对模型精度进行验证,并将其与传统机器学习模型进行了对比。研究表明,所提模型预测准确率为89.95%,准确度最高。模型的AUC值为0.977,说明模型在适用性和预测精度方面都优于传统预测模型。采用SHAP分析对该模型进行了全局影响因素分析及案例分析,结果表明,检查频次是导致土石坝风险最重要的影响因素之一。Dam failure often causes an enormous loss of life and property and huge environmental damage.Accurate and fast estimation of the risk level of earth-rock dams is of great significance for controlling their failure hazards.This paper develops a fast prediction model of the earth-rock dam risk grade based on GA-LightGBM,using the K-Nearest Neighbor(KNN)algorithm to fill a large amount of missing data in the database,and adopting a Genetic Algorithm(GA)to optimize the hyperparameters of Light Gradient Boosting Machine(LightGBM).The model accuracy is verified using the receiver operating characteristic(ROC)curves,the area under the curve(AUC),and other evaluation indexes;and it is compared with the traditional machine learning model.The results show that this new model has a high accuracy of 89.95%and its AUC value is 0.977,indicating it is better in terms of applicability and accuracy.Analysis of global influencing factors and case studies using Shapley Additive Explanations(SHAP)show the frequency of inspection is one of the most important factors leading to the risk of earth-rock dams.

关 键 词：风险等级 遗传算法 GA-LightGBM 快速预测模型 SHAP分析 

分 类 号：TV641[水利工程—水利水电工程]