Research On The Stratigraphic Correlation Method Of Logging Curve Based On Vortex Search Algorithm

Reasonable stratigraphic division and correlation is an important procedure from single well evaluation to multi well evaluation in reservoir description.It is also the basis for describing reservoir morphology and spatial distribution of reservoir parameters,which is of great importance for oil and gas field development.In reservoir description,the quality of the final results of oil field research by using multi well logging evaluation is largely determined by the work of inter well stratigraphic correlation.Through stratigraphic correlation,we can know the change of stratigraphic sequence,lithofacies and thickness,and study the horizontal and vertical change law of reservoir,which provides a strong basis for finding favorable oil-gas blocks and reasonable development.Artificial stratigraphic correlation depends on the experience and ability of logging personnel,which can not effectively avoid the influence of human factors.With the rapid development of artificial intelligence,automatic stratigraphic correlation has become inevitable.As a meta heuristic algorithm,vortex search algorithm has been widely used in various fields.Therefore,this paper uses the characteristics of logging curve to carry out the research of intelligent stratigraphic correlation method based on vortex search algorithm.As an important part of research on "13th Five-Year" National Science and Technology Major Project 19,Topic 5,Subproject 6 "Study on the formation model construction method of 3D logging attribute in horizontal well",the related technical theories and methods of the combination of stratigraphic correlation and intelligent algorithm is deeply studied on the basic of analyzing the research status and development trend of well logging software and stratigraphic correlation at home and abroad.Combined with the research objectives of major national science and technology projects,the research on stratigraphic correlation method of well logging curve based on vortex search algorithm is carried out based on the integrated software platform of CIFLog.The details are as follows:First of all,through the analysis and selection of the commonly used logging curve data for stratigraphic division,a stratigraphic division method based on Improved BP neural network is proposed,and a stratigraphic division model is established;Secondly,according to the result of stratigraphic division based on Improved BP neural network,the feature vector of stratigraphic correlation is extracted,the similarity between the feature vectors of two wells is measured by distance,the fitness function of stratigraphic correlation with constraints is constructed,and the function is optimized by vortex search algorithm.In order to solve the problem that vortex search algorithm is easy to fall into local optimization,the global searching ability of the vortex search algorithm is improved.Finally,based on CIFLog integrated processing and interpretation platform,and guided by logging data processing and comprehensive interpretation theory,combined with automatic stratigraphic correlation algorithm,the automatic stratigraphic correlation function module is designed and implemented in the integrated platform horizontal well 3D attribute system.Thewell logging data of oil field work area are processed,and good results are obtained,which has certain use value and promotion value.