Research On Intelligent Interpretation Method For Stratum Recognition Of Horizontal Well Geosteering

Geosteering is the technical guarantee for the safe and efficient development of horizontal wells.It mainly involves the data processing and analysis of comprehensive logging and LWD data while drilling.There are many difficulties in stratigraphic interpretation,such as large amount of data processing capacity and complicated calculation process.It requires high knowledge and experience of geosteering engineers,and it is difficult to form a one-to-many production development model.Oil companies in China urgently need geosteering software to improve the efficiency and accuracy of field analysis.Therefore,this paper researches the intelligent interpretation method of formation identification in horizontal well geosteering,and assists the geosteer to complete the guiding work.Based on the research of geosteering operation flow and geological renewal method of horizontal wells,this paper summarizes the key work of landing control,geological steering target analysis and drilling in horizontal reservoirs.The calculation model of the true thickness of the stratum is established,combined with the calculation of the apparent dip angle of the stratum,the contrast of the data in the stratigraphic contrast and the lower stratum prediction is enhanced.The vertical depth of target in target analysis is updated by the calculation of the slope and the determination of the length of the target entry segment.Horizontal drilling mainly relies on the analysis of the situation of drilling in non-target formation and the distance between bit and reservoir interface to ensure the trajectory of wellbore in reservoir and improve the drilling rate of high-quality reservoir.Many logging parameters can be collected in onshore drilling with comprehensive logging as the main method.In this paper,several parameters reflecting formation lithology sensitivity are optimized by direct optimization method.After collecting data from six horizontal wells for testing,drilling time,total amount and C1 are selected as calculation parameters when lithology identification is obtained.According to the principle of support vector classifier algorithm,an intelligent lithology recognition method for comprehensive logging formation is established.The validity of the method is verified by the actual data collected from two horizontal wells.The research shows that the method can improve the accuracy of lithology recognition based on comprehensive logging data classification.In the horizontal well construction with LWD,according to the principle of Case-Based Reasoning technology,a case base of adjacent well marker formation is established by collecting common LWD curve data(gamma logging while drilling,deep and shallow resistivity logging while drilling).The similarity between normal drilling data and adjacent well marker formation is calculated by Dynamic Time Warping algorithm,and an intelligent identification method of LWD markers formation is established.The results of four wells show that this method can accurately and effectively identify the marker layers in the deviation section.Aiming at the problem that the determination of the position of drilling bit from the horizontal section drilled to the reservoir interface lags behind the well trajectory,this paper researches the calculation model of drillability index.The drillability index of formation rock is determined by effective drilling pressure,bit torque,rotational speed and drilling speed.Through real-time monitoring of these parameters by comprehensive mud logging,an intelligent interpretation method of reservoir interface in horizontal section is established,and the sensitivity of this method to identification of formation interface is analyzed.According to the established model and calculation method,this paper develops an intelligent interpretation software for stratum recognition of horizontal well geosteering by using Visual Studio 2017 programming platform and SQL Server 2008 database management platform.The software is integrated into the software platform of drilling engineering integration of Southwest Petroleum University,which can realize the intelligent identification of stratum lithology,slope marker layer.The actual data of two horizontal wells in XinJiang well drilling field are collected,and the application of intelligent identification method and the stability of software operation are comprehensively analyzed.