In mineral resources exploitation,design and construction of rock mass engineering and stability prediction of underground structures,it is very important to determine rock mass mechanical parameters exactly. In this paper,on the basis of the comprehensive analysis about back analysis methods for rock mass mechanical properties,a new intelligent displacement back analysis method was introduced. Firstly, the numerical ranges of rock mass back analysis parameters were selected by means of the practical experiences and GSI system,and the average values of vault settlement and horizontal convergence dis-placement were obtained by in-situ displacement measurement. Then,displacement back analysis was carried out by combining genetic algorithm ( GA) with FLAC3D numerical simulation software,and the more accurate rock mass mechanical parameters were determined. According to the above method,a displacement back analysis system combined GA with FLAC3D software was developed by using object oriented programming language VC++. This method has been applied in ore body,hanging wall and footwall rock masses of a lead-zinc mine to make back analysis on the rock mass mechanical parameters. The validity and prac-ticability of this method was proved by the comparative analysis with other back analysis methods.%在矿产资源开采、采矿设计及施工、地下结构物稳定性预测等岩土工程领域中,初步的也是重要的研究课题之一是确定岩体力学参数.经过综合分析多种岩体力学性质反演法,提出了一种智能位移反演方法.首先,在工程经验和GSI体系的基础上选定待反演的岩体力学参数的数值范围,通过巷道现场位移量测获得拱顶下沉和水平收敛位移.然后,运用遗传算法(GA)与FLAC3D数值模拟软件相结合,进行位移反演,确定比较正确的岩体力学参数.根据上述方法采用VC++高级程序语言开发了将遗传算法与FLAC3D软件相结合的位移反演系统,以某铅锌矿体和上下盘岩体为例,进行了矿岩体力学参数反演.进行与其他反演法的对比分析,证明了该方法的正确性和实用性.
展开▼
