中国安全科学学报 ›› 2019, Vol. 29 ›› Issue (9): 57-63.doi: 10.16265/j.cnki.issn1003-3033.2019.09.009

• 安全系统学 • 上一篇    下一篇

RAGA-PPC云模型在边坡稳定性评价中的应用

吴孟龙1, 叶义成1,2 教授, 胡南燕**1, 姚囝1, 江慧敏1, 李文1   

  1. 1 武汉科技大学 资源与环境工程学院,湖北 武汉 430081;
    2 湖北省工业安全工程技术研究中心,湖北 武汉430081
  • 收稿日期:2019-05-15 修回日期:2019-07-12 出版日期:2019-09-28 发布日期:2020-10-30
  • 通讯作者: ** 胡南燕(1991—),女,浙江金华人,博士,讲师,主要从事矿山安全技术方面的研究。E-mail:hunanyan@wust.edu.cn。
  • 作者简介:吴孟龙 (1995—),男,河南省开封人,硕士研究生,研究方向为矿山安全评价和安全预测。E-mail:1205548826@qq.com。
  • 基金资助:
    国家自然科学基金资助(51804224)。

Application of RAGA-PPC cloud model in slope stability evaluation

WU Menglong1, YE Yicheng1,2, HU Nanyan1, YAO Nan1, JIANG Huimin1, LI Wen1   

  1. 1 School of Resource and Environmental Engineering, Wuhan University of Science and Technology, Wuhan Hubei 430081, China;
    2 Industrial Safety Engineering Technology Research Center of Hubei Province, Wuhan Hubei 430081, China
  • Received:2019-05-15 Revised:2019-07-12 Online:2019-09-28 Published:2020-10-30

摘要: 为精确评价露天矿边坡稳定性,预防边坡事故发生,首先选取黏聚力、内摩擦角、边坡角、边坡高度、孔隙水压力比、天然容重等6个指标,建立边坡稳定性评价指标体系;其次基于20组样本数据,采用投影寻踪聚类(PPC)算法将多维数据投影在一维平面上,以基于实数编码的加速遗传算法(RAGA)优化投影指标,并计算各指标的权重值;然后采用正向云发生器得到边坡各稳定性等级的隶属度,并据此计算边坡稳定性等级的加权隶属度;最后构建定性与定量相结合的RAGA-PPC等级评价云模型,并用5组样本数据验证RAGA-PPC云模型的评价结果。结果表明:影响边坡稳定性的最主要因素是边坡高度;RAGA-PPC云模型评估的20组样本数据准确率为95.0%,检测5组样本数据准确率达到100%。

关键词: 露天矿山边坡, 加速遗传算法(RAGA), 稳定性评价, 云模型, 投影寻踪聚类(PPC)

Abstract: In order to accurately evaluate slope stability of open pits and prevent slope accidents, firstly, six indexes, including cohesion, slope angle, slope height, pore water pressure ratio and natural bulk density, were selected to establish a slope stability evaluation index system. Secondly, based on 20 sets of samples, PPC method was adopted to project multidimensional data on one-dimensional plane, and weight values of each index were calculated with RAGA optimized projection indexes. Then, membership degree of each stability level were obtained by using forward cloud generator, on the basis of which weighted membership degree of slope stability levels was calculated. Finally, a RAGA-PPC evaluation cloud model combining qualitative and quantitative analysis was constructed and verified through 5 groups of samples. The results show that the most important factor that affects slope stability is slope height. RAGA-PPC cloud model's accuracy rate for evaluating 20 sets of sample data is 95.0% and 100% for 5 sets of sample data.

Key words: open pit slope, real-coded accelerated genetic algorithm (RAGA), stability evaluation, cloud model, projection pursuit classification (PPC)

中图分类号: