Application of genetic algorithms in the optimization of thicknesses of final lining of caverns
LIU Hao1，FENG Jimeng2，3，WANG Zhiyong1，ZHANG Huijian2，3，QIU Wenge2，3
(1. CREEC(Chongqing) Survey，Design and Research Co.，Ltd.，Chongqing 401121，China；2. Key Laboratory of Transportation Tunnel Engineering，Ministry of Education，Southwest Jiaotong University，Chengdu，Sichuan 610031，China；3. School of Civil Engineering，Southwest Jiaotong University，Chengdu，Sichuan 610031，China)
Abstract：Due to the restriction of topography and the interaction between related engineering，the probability of occurrence of high density underground caverns is increasing. The relationship between each cavern and the overall stability of the caverns，the dominant failure mode and how to design the strength of supporting measures of each cavern reasonably to realize the equal strength theory are becoming increasingly urgent. The feasibility and reasonability of the application of the main current criterion to gain safety factor in caverns were analyzed based on the theory of finite element strength reduction and some large high-density underground caverns in Chongqing. Based on the least square method，the relationship between the safety factor and the final lining thickness of each cavern is expressed through the quadratic polynomial and the thickness of each final thickness is optimized by the genetic algorithm. The local failure of the secondary parts such as the surrounding rock between caverns influences little on the stability of each cavern or caverns. The dominant failure modes of parallel caverns can be classified into three categories determined by many factors such as the distance between them and the embedded depth. When the embedded depth is shallow and the distance is large，the plastic region near the vault of each cavern grows upward to the surface in the direction opposite to the center of the caverns and finally results in failure. When the embedded depth is deep and the distance is small，the plastic region near the vault of each cavern grows upward in the direction to the center of the caverns and finally results in failure. When the embedded depth and the distance are in moderate sizes，the above two failures contribute to the failure together. For the various dominant failure modes of caverns，the curves depicting the relationship between the displacement of key points and the strength reduction factor may be less obvious and the distribution and development process of plastic zone should be resorted to simultaneously to obtain the safety factor of cavern. The expression of implicit relationship of quadratic polynomial based on the least square method and the optimization of the thickness of final lining by the genetic algorithm achieved a certain fitting precision and ideal optimization results.
刘 浩1，冯冀蒙2，3，王志勇1，章慧健2，3，仇文革2，3. 遗传算法在洞群二次衬砌厚度优化中的应用[J]. 岩石力学与工程学报, 2016, 35(8): 1595-1601.
LIU Hao1，FENG Jimeng2，3，WANG Zhiyong1，ZHANG Huijian2，3，QIU Wenge2，3. Application of genetic algorithms in the optimization of thicknesses of final lining of caverns. , 2016, 35(8): 1595-1601.
江 权，冯夏庭，向天兵. 基于强度折减原理的地下洞室群整体安全系数计算方法探讨[J]. 岩土力学，2009，30(8)：2 483-2 488. (JIANG Quan，FENG Xiating，XIANG Tianbing. Discussion on method for calculating general safety factor of underground caverns based on strength reduction theory[J]. Rock and Soil Mechanics，2009，30(8)：2 483-2 488.(in Chinese))
<body>Your browse does not support frame!</body>
2016, Vol. 35 
