| For a long time,coal has been an important support to promote the high-quality development of our economy and society and ensure national security.Due to the large demand for coal in China,mining of shallow coal mines has been basically completed up to now,and most coal mines have entered the deep mining stage.Deep coal and rock mass faces high gas,high ground stress,high temperature and high osmotic pressure.As a result,China’s coal mining is facing more severe challenges.In order to achieve safe and efficient coal mining,we must take the road of precision mining with fewer people,safety and intelligence.In the process of promoting the intelligent development of coal mining,3D geological modeling technology is indispensable.Based on this,in the context of the construction of intelligent mine,through the analysis of common geological exploration data,combining 3D geological modeling with machine learning,which is the core of artificial intelligence technology,an efficient,convenient and highly automated intelligent 3D geological modeling method is proposed.In addition,a systematic study was conducted on the process,applicability,accuracy and application value of the modeling method.The main research contents and achievements are as follows:(1)This paper analyzes the traditional 3D geological modeling methods at home and abroad,and probes into the research status of their theories,methods and techniques.Explicit modeling methods require a large amount of human-computer interaction and expert experience,and have high professional requirements for modelers,and the built models have sharp edges and corners and low smoothness.The commonly used implicit modeling methods have high requirements for data volume,and the acquisition of geological data is difficult,which is usually limited and sparse.The intelligent 3D geological modeling method can improve the basic algorithm and modeling process in view of the above deficiencies.(2)Explore the basic principles of machine learning,conduct in-depth analysis of four algorithms of support vector machine,nearest neighbor,naive Bayes and BP neural network,explore their calculation and prediction methods,and obtain four machine learning algorithms that can be applied to the prediction and classification of geological data.(3)Based on machine learning algorithm,the intelligent three-dimensional geological modeling method is proposed,and a set of systematic and comprehensive modeling process including data acquisition and processing,training optimization,model construction and reliability analysis is constructed.Each modeling step is described and analyzed in detail.Then,this modeling method is used to carry out experiments on common sedimentary strata,faults,folds and other complex geological structures as well as special fault structures.The results show that this method can meet common geological modeling requirements and has good modeling effect on complex multi-Z value geological structures.It can be applied to geological modeling work,and has the characteristics of fast modeling speed,high accuracy and easy updating.(4)The intelligent 3D geological modeling method is used to establish a 3D geological model for Gubei Coal mine,and the reliability of the method is verified by the actual geological exploration data.The results show that the method can build a 3D geological model conforming to the geological law and understanding,and the established model has a good agreement with the actual exploration data disclosure results.Based on the relevant research in this paper,the intelligent three-dimensional geological modeling method can solve some shortcomings of the current modeling methods,can carry out scientific and efficient integrated construction of the complex geological structure three-dimensional model,and be applied to the fine exploration of geological conditions,laying a solid foundation for the intelligent development of coal mining.Figure [63] Table [23] Reference [85]... |
PDF Full Text Request