The Research And Application Of Logging Curves Similarity Algorithm Based On Feature Information

The quality of original perforating data is directly related to the final results ofperforating technology, also directly affect the precision of the logging data processing andinterpretation, so the acceptance work of perforation data is very important.The acceptance ofwell logging curve is of great value to application and research as the core part of perforationdata acceptance work can guarantee the consistency of perforating original dataeffectively,and it can provide accurate data for subsequent perforating depth calculating work.This paper focused on the research about realizing well logging curve automatedacceptance efficiently and accurately. The mainly research contents are as follows: Thelogging curve characteristics are analyzed, and the method of wavelet analysis and the basicprinciple of distance calculation by Fréchet distance curve are introduced. An improvedwavelet threshold denoising method is proposed by analysing well logging curve and thewavelet thresholding functions characteristics.The simulation experiment and logging curvesexample results show that, the improved wavelet thresholding method can preserve curvesdetails well while reducing noise efficiently. The relationship between Gauss wavelettransform and well logging curve feature points are introduced, then the preprocessed loggingcurves were made gaussian wavelet transform by using the properties of gaussian waveletbasis function to extract the curve features of extreme value point and inflexionpoint.Meanwhile, the method was applied to perforation original data automated acceptancesoftware, and it is of great significance in improving labor productivity and reducingproduction cost. Based on the advantages and disadvantages of discrete Fréchet distance, thepaper puts forward an improved discrete Fréchet distance method to check logging curvesimilarity. the curve space was shortened and the Frechet arrangement was optimize byintegrating logging curve trend to improve the efficiency of algorithm. And the accuracy ofthe proposed algorithm was proved through the analysis of experimental results.Combined with these research, which is applied in perforation original data automatedacceptance software, and the real logging data was usd to test,it further proved the feasibility,efficiency and practicability of the algorithm. It has important significance in the increase inlabor productivity, reduce production cost.