采矿与安全工程学报 ›› 2015, Vol. 32 ›› Issue (6): 989-995.

• 论文 • 上一篇    下一篇

冻结黏土蠕变模型的模糊随机辨识与综合选优

  

  1. 1.安徽理工大学土木建筑学院,安徽淮南 232001;2.安徽大学资源与环境工程学院,安徽 合肥 230022;3.南通职业大学建筑工程学院,江苏南通 226001
  • 收稿日期:2015-03-11 出版日期:2015-11-15 发布日期:2015-12-04
  • 作者简介:姚亚锋(1978—),男,江苏省南通市人,博士研究生,讲师,从事智能计算在地下结构工程方面的研究。
  • 基金资助:

    国家自然科学基金项目(51374010,51474004);安徽高校省级自然科学研究重点项目(KJ2010A094,KJ2011A093)

Fuzzy random identification and comprehensive selection for frozen clay creep model

  • Received:2015-03-11 Online:2015-11-15 Published:2015-12-04

摘要: 针对淮南某矿区人工冻结黏土在-5,-10,-15 ℃下进行单轴抗压强度试验,得到温度对人工冻结黏土单轴抗压强度的影响规律;在-5,-10,-15 ℃下分别进行低中高3 种应力等级的单轴蠕变试验,获得温度、应力等级对蠕变的影响规律。通过最优解信息素模糊化系数对传统的蚁群算法进行改进,提高搜索效率的同时有效避免了局部最优;使用模糊蚁群算法对常用冻土蠕变模型的柔量参数进行随机反演;综合精度和复杂度,建立双指标综合选优目标函数,确定指标的模糊权重,最后评价出3种不同应力下的最优模型。

关键词: 冻结黏土, 蠕变模型, 模糊蚁群算法, 模糊随机辨识, 综合选优

Abstract: The uniaxial compressive strength test has been conducted on the artificial frozen clay in Huainan mine under the temperature of -5, -10 and -15 ℃. The result shows effect law of temperature for the uniaxial compressive strength of artificial frozen clay. The uniaxial creep test of high, mid and low stress levels has been carried out under the conditions of -5, -10 and -15 ℃ respectively, which leads to the effect law of temperature and stress level for creep. The traditional ant colony algorithm has been optimized into fuzzy ant colony algorithm according to optimal-solution pheromone of fuzzy coefficient, with the search efficiency increased and local optimum avoided. It is used in random inversion for the creep compliance of the common model, then the dual factors objective function for comprehensive selection is built by taking precision and complexity into consideration, finally the optimized model is obtained under three different stresses with appropriate fuzzy weight calculation method.

Key words: frozen clay, creep model, fuzzy ant colony algorithm, fuzzy random identification, comprehensive selection