Research On Automatic Selection And Recognition Of Depth Correction Results Of Completion

Depth correction of completion is an important part of completion process in oilfield, the results are mainly used in the calculation of the design data for perforating construction, the result is accurate or not directly affect perforating construction work, affect the production of oil wells, affect the evaluation of a block, even to the development of an oil field has an important influence, so the accuracy of the results is very important.This paper focused on the research about the how to calculate the completion depth correction results efficiently and accurately. Firstly, using wavelet as well as environmental analysis method preprocessing of log data, to eliminate the influence of the environment and the noise on the curve data; then, using based on the global constraint curve segment polynomial matching method for effective extraction of logging curve trend; finally,according to the contrast curve of a dynamic pattern matching distance metrics to determine the similarity, in order to ensure the sample set of feature layer under the condition of accurate recall, to further improve the accuracy of the results of the correction, the main contents are as follows:1.Aiming at the influence of the non formation factor on the well completion depth correction of the completion depth correction in different logging environment, to the log curve carries on environmental correction and standardization of log data, to try to eliminate these non stratigraphic factors. And then according to the noise signal in the logging curve this problem, combined with the analysis of the existing logging data denoising method,proposed a new threshold denoising method of logging data based on wavelet packets, and applied to the comparison of log data, proving effectiveness of this method.2.Aiming at the research on automatic selection and recognition of depth completion results of correction is the curve of this layer accurately screen the central theme. Using based on the global constraint curve segment polynomial matching method for effective extraction of logging curve trend, this method first uses curve segmentation method based on feature points, describe the curve by section, then use these segmentation points as the restriction of the matching polynomial curve segment, and at the point of the sequence, to fit the constraints of polynomial fitting method, trend identification curve sequence features.3.Aiming at the similarity distance judgment of the curve sequence sample concentration log curve, using the distance measurement based on dynamic pattern matching(DPM), the comparison of the curve of the completion depth calibration curves of the sequence sampleswith the similar comparison, and combine the information of feature curves, amplitude and amplitude, extracting the characteristic layer sample set. Finally, combining with the actual,application of the previous research to the completion depth correction results automatic identification study.