Prediction of rock mass rating using fuzzy logic and multi-variable RMR regression model

来源 :International Journal of Mining Science and Technology | 被引量 : 0次 | 上传用户:yh124712
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
Rock mass rating system(RMR) is based on the six parameters which was defined by Bieniawski(1989)[1]. Experts frequently relate joint and discontinuities and ground water conditions in linguistic terms with rough calculation. As a result, there is a sharp transition between two modules which create doubts.So, in this paper the proposed weights technique was applied for linguistic criteria. Then by using the fuzzy inference system and the multi-variable regression analysis, the accurate RMR is predicted. Before the performing of regression analysis, sensitivity analysis was applied for each of Bieniawski parameters.In this process, the best function was selected among linear, logarithmic, exponential and inverse functions and finally it was applied in the regression analysis for construction of a predictive equation. From the constructed regression equation the relative importance of the input parameters can also be observed. It should be noted that joint condition was identified as the most important effective parameter upon RMR. Finally, fuzzy and regression models were validated with the test datasets and it was found that the fuzzy model predicts more accurately RMR than regression models. Rock mass rating system (RMR) is based on the six parameters which were defined by Bieniawski (1989) [1]. Experts frequently relate joint and discontinuities and ground water conditions in linguistic terms with rough calculation. As a result, there is a sharp transition between two modules which create doubts.So, in this paper the proposed weights technique was applied for linguistic criteria. Then by using the fuzzy inference system and the multi-variable regression analysis, the accurate RMR is predicted. , sensitivity analysis was applied for each of Bieniawski parameters. In this process, the best function was selected among linear, logarithmic, exponential and inverse functions and finally it was applied in the regression analysis for construction of a predictive equation. From the constructed regression equation equation the relative importance of the input parameters can also be observed. It should be noted that joint condition was identified as the Finally, fuzzy and regression models were validated with the test datasets and it was found that the fuzzy model predicts more accurately RMR than regression models.
其他文献
介绍了一种电荷载流子倍增寄存器(CCM)电极间隙的传输线模型,通过求解其波动方程得到了电荷载流子倍增寄存器电极间隙的电势分布及电场分布。模拟分析表明:电荷载流子倍增寄
In this paper, the characteristics of forces in active control systems connected to adjacent levels of a building are analyzed. The following characteristics ar
There were huge life and property losses during the Ms8.0 Wenchuan earthquake on May 12, 2008. Strain fluctuation curves were completely recorded at stress obse
The east-west location change of the East Asian westerly jet (EAWJ) at 200 hPa during Meiyu and the associated spatial distribution variation of precipitation i
流注的产生和传播是空气间隙击穿中一个重要的物理过程,研究低气压下流注的产生和传播过程有助于理解空气间隙的击穿放电,对高海拔地区输电线路及输变电设备的设计和运行有着重
In the North China Plain,fertilizer management and tillage practices have been changing rapidly during the last three decades; however,the influences of long-te
A new fluorescence quenching method has been developed for the determination of platinum in nonionic mieroemulsion medium.This method is based on the fact that
以喹啉为原料,在温和条件下制得喹啉类离子化合物1~9,并在室温下将其用于Knoevenagel缩合反应,以80%~95%的收率高效生成相应的烯腈a~f,离子化合物经过简单处理即可重复使用。 Q