Research On Preprocessing Method Of Nuclear Magnetic Resonance Logging Data

Nuclear magnetic resonance logging data preprocessing is crucial for subsequent inversion and application.This article investigates the commonly used nuclear magnetic resonance logging tools of Schlumberger Company,and studies a large amount of measurement information from various tools.Using commonly used porosity correction formulas,it clarifies the origin of the porosity scale factor curve in the instrument,and finally optimizes the correction formula.In addition,for echo signals with low signalto-noise ratio,this paper proposes a Kalman filter algorithm,and extends and optimizes the conventional Kalman filter.Finally,the effect is compared with other noise reduction algorithms.This article focuses on the following three aspects of research:First,study measurement information.Investigate three types of nuclear magnetic resonance logging tools commonly used by Schlumberger,including CMR＿PLUS(regular),CMR＿NG(multi-dimensional)and pro VISION(while drilling),the log files of the three instruments are decompiled through the compiled decompilation software to obtain the original measurement parameters and curves.Finally,a large number of parameters and curves of the three instruments are analyzed respectively.Secondly,study the porosity correction formula.Based on the measurement information of Schlumberger’s nuclear magnetic logging tools,the commonly used nuclear magnetic resonance porosity correction formulas are analyzed.Through the measurement information of the tools,each meaning of the correction formula is analyzed,including calibration amplitude,temperature,frequency correction,and other parameter items.And specify how to select the values of each parameter in the formula.Compare the results calculated by the correction formula with the porosity correction factor curve in the instrument measurement information of Schlumberger Company to investigate the applicability of the correction formula,and finally supplement and improve the correction formula.Third,improve the signal to noise ratio of the original echo.Aiming at the problem of low signal to noise ratio of the original echo signal measured by nuclear magnetic resonance logging tools,a noise reduction method based on Kalman filtering is proposed.The influence of several parameters of Kalman filter on its noise reduction effect is studied to make the algorithm more applicable.In addition,several algorithms based on Kalman filter expansion are introduced,and finally,other commonly used noise reduction methods are compared.It is proved that Kalman filtering can effectively improve the signal-to-noise ratio of nuclear magnetic resonance echo signals.