| In the practical work of mud logging interpretation,because of many factors affecting the lithology of the formation profile and the low degree of data integration,there are a lot of manual correction phenomena in the lithology profile drawing of the mud logging comprehensive map.Therefore,the need for intelligent lithology identification of formation profile is increasingly urgent in mud logging interpretation.In this paper,the working principle and application requirements of logging interpretation are analyzed.To solve the above problems,a method of log profile location interpretation based on data mining is proposed.Firstly,the characteristics of source data are analyzed,the logical model and physical model of mud logging interpretation data are designed,and the ETL process and control mechanism from source data to target data are elaborated.Secondly,several data mining algorithms are studied and compared,and the neural network algorithm is selected to build the mining model of lithologic feature pattern based on the characteristics of mud logging interpretation data,and the optimal lithologic feature recognition pattern is mined.Finally,a rule-based core profile location method is designed,and the optimal lithologic feature recognition pattern obtained is applied to production data to quickly process the lithologic profile location and improve the efficiency of comprehensive logging mapping.On the basis of data mining-based logging profile location interpretation and processing method,a logging profile location interpretation and processing platform is developed.The rationality and practicability of the proposed method are verified by the actual data of a logging company.The efficiency of comprehensive logging mapping is improved,and the developed platform has good application value. |
PDF Full Text Request