Classification of surrounding rock is the cornerstone of tunnel design and construction.The traditional methods are mainly qualitative and manual and require extensive professional knowledge and engineering experience...Classification of surrounding rock is the cornerstone of tunnel design and construction.The traditional methods are mainly qualitative and manual and require extensive professional knowledge and engineering experience.To minimize the effect of the empirical judgment on the accuracy of surrounding rock classification,it is necessary to reduce human participation.An intelligent classification technique based on information technology and artificial intelligence could overcome these issues.In this regard,using 299 groups of drilling parameters collected automatically using intelligent drill jumbos in tunnels for the Zhengzhou-Wanzhou high-speed railway in China,an intelligent-classification surrounding-rock database is constructed in this study.Based on a machine learning algorithm,an intelligent classification model is then developed,which has an overall accuracy of 91.9%.Finally,using the core of the model,the intelligent classification system for the surrounding rock of drilled and blasted tunnels is integrated,and the system is carried by intelligent jumbos to perform automatic recording and transmission of drilling parameters and intelligent classification of the surrounding rock.This approach provides a foundation for the dynamic design and construction(both conventional and intelligent)of tunnels.展开更多
Based on the principle of Mahalanobis distance discriminant analysis (DDA) theory, a stability classification model for mine-lane surrounding rock was established, including six indexes of discriminant factors that re...Based on the principle of Mahalanobis distance discriminant analysis (DDA) theory, a stability classification model for mine-lane surrounding rock was established, including six indexes of discriminant factors that reflect the engineering quality of surrounding rock: lane depth below surface, span of lane, ratio of directly top layer thickness to coal thickness, uniaxial comprehensive strength of surrounding rock, development degree coefficient of surrounding rock joint and range of broken surrounding rock zone. A DDA model was obtained through training 15 practical measuring samples. The re-substitution method was introduced to verify the stability of DDA model and the ratio of mis-discrimination is zero. The DDA model was used to discriminate 3 new samples and the results are identical with actual rock kind. Compared with the artificial neural network method and support vector mechanic method, the results show that this model has high prediction accuracy and can be used in practical engineering.展开更多
In order to improve the discrimination precision of support vector machine(SVM) in classification of surrounding rock, a Genetic Algorithm(GA) was used to optimize SVM parameters in the solution space.The idea of exam...In order to improve the discrimination precision of support vector machine(SVM) in classification of surrounding rock, a Genetic Algorithm(GA) was used to optimize SVM parameters in the solution space.The idea of examination of model reliability was introduced to check the reliability of the SVM parameters,obtained by genetic algorithms.In the process of model reliability,a trend examination method is presented,which checks the reliability of the model via the influence trend of impact factors on the object of evaluation and their evaluation level.Trend examination methods are universal,showing new ideas in model reliability examination and can be used in any problems of examination of reliability of models,based on previous experience.We established a GA-SVM based reliability model of a classification the surrounding rock and applied it to a practical engineering situation.The result shows that the improved SVM has a high capability for generalization and prediction accuracy in classification of surrounding rock.展开更多
基金supported by the National Natural Science Foundation of China(NSFC)[Grant Nos.51578458,and 51878568]the China Railway Corporation Science and Technology Research and Development Program[Grant Nos.2017G007-H,2017G007-F,P2018G007,K2018G014,and K2018G014-01].
文摘Classification of surrounding rock is the cornerstone of tunnel design and construction.The traditional methods are mainly qualitative and manual and require extensive professional knowledge and engineering experience.To minimize the effect of the empirical judgment on the accuracy of surrounding rock classification,it is necessary to reduce human participation.An intelligent classification technique based on information technology and artificial intelligence could overcome these issues.In this regard,using 299 groups of drilling parameters collected automatically using intelligent drill jumbos in tunnels for the Zhengzhou-Wanzhou high-speed railway in China,an intelligent-classification surrounding-rock database is constructed in this study.Based on a machine learning algorithm,an intelligent classification model is then developed,which has an overall accuracy of 91.9%.Finally,using the core of the model,the intelligent classification system for the surrounding rock of drilled and blasted tunnels is integrated,and the system is carried by intelligent jumbos to perform automatic recording and transmission of drilling parameters and intelligent classification of the surrounding rock.This approach provides a foundation for the dynamic design and construction(both conventional and intelligent)of tunnels.
基金Project(50490274) supported by the National Natural Science Foundation of China
文摘Based on the principle of Mahalanobis distance discriminant analysis (DDA) theory, a stability classification model for mine-lane surrounding rock was established, including six indexes of discriminant factors that reflect the engineering quality of surrounding rock: lane depth below surface, span of lane, ratio of directly top layer thickness to coal thickness, uniaxial comprehensive strength of surrounding rock, development degree coefficient of surrounding rock joint and range of broken surrounding rock zone. A DDA model was obtained through training 15 practical measuring samples. The re-substitution method was introduced to verify the stability of DDA model and the ratio of mis-discrimination is zero. The DDA model was used to discriminate 3 new samples and the results are identical with actual rock kind. Compared with the artificial neural network method and support vector mechanic method, the results show that this model has high prediction accuracy and can be used in practical engineering.
基金supported by the Key Project of Ministry of Education (No.108158)the Natural Science Foundation of Shandong Province(No.Y2007F53)the Postdoctoral Science Foundation of China(No.2009 0461203).
文摘In order to improve the discrimination precision of support vector machine(SVM) in classification of surrounding rock, a Genetic Algorithm(GA) was used to optimize SVM parameters in the solution space.The idea of examination of model reliability was introduced to check the reliability of the SVM parameters,obtained by genetic algorithms.In the process of model reliability,a trend examination method is presented,which checks the reliability of the model via the influence trend of impact factors on the object of evaluation and their evaluation level.Trend examination methods are universal,showing new ideas in model reliability examination and can be used in any problems of examination of reliability of models,based on previous experience.We established a GA-SVM based reliability model of a classification the surrounding rock and applied it to a practical engineering situation.The result shows that the improved SVM has a high capability for generalization and prediction accuracy in classification of surrounding rock.
