基于智能算法的铁路边坡危岩稳定性预测研究

Research on the Stability Prediction of Dangerous Rocks on Railway Slopes Based on Intelligent Algorithms

研究目的:在铁路建设与运维中,危岩是边坡地质灾害防治的重点问题。为实现对铁路边坡危岩稳定性状态的精准、快速预测,提高铁路沿线防护准确性,提出基于PCA改进GASA-FCM模型(Genetic Algorithms Simulated Annealing Fuzzy C-Means, GASA-FCM)的铁路边坡危岩稳定性预测新方法,以黔桂线洛满-麻尾段为依托,对该模型的准确性和优越性进行验证。研究结论:(1)综合考虑边坡及危岩状况、岩体力学性能和水文地质条件三方面,构建包含13个二级指标的铁路边坡危岩稳定性预测指标体系;(2)与传统FCM模型和GA-FCM模型相比,本文提出的GASA-FCM预测模型收敛值稳定,均方误差更小;(3)该模型在工程实例洛满-麻尾段边坡危岩稳定性等级预测评价中,评价结果与工程实际勘查结果完全符合,验证了该模型的准确性和优越性;(4)本研究成果可为其他铁路工程沿线危岩稳定性等级预测评价提供借鉴,对加快铁路全线信息化管理具有重要意义。

Research purposes:In railway construction and operation and maintenance,dangerous rock is a key problem in the prevention and control of geological hazards on side slopes.In order to achieve accurate and rapid prediction of the stability state of dangerous rocks on railway slopes and improve the accuracy of protection along railway lines,a new method of predicting the stability of dangerous rocks on railway slopes based on PCA improved GASA-FCM model(Genetic Algorithms Simulated Annealing C-Means,GASA-FCM)is proposed.The accuracy and superiority of the model are verified based on the Luoman-Mawei section of the Qiangui line.Research conclusions:(1)Considering the slope and dangerous rock condition,physical and mechanical properties of rock body and hydrogeological conditions,a railway slope dangerous rock stability prediction index system containing 13 secondary indicators is constructed.(2)Compared with the traditional FCM model and GA-FCM model,the GASA-FCM prediction model proposed in this paper has stable convergence value and smaller mean square error.(3)The model is used in the prediction and evaluation of the stability level of dangerous rocks in the slope of the Luoman-Mawei section,and the results are fully consistent with the actual survey results,which verifies the accuracy and superiority of the model.(4)The research results can provide reference for the prediction and evaluation of the stability level of dangerous rocks along other railway projects,which is of great significance to accelerate the information management of the whole railway line.