The Application Of Similarity Query Based On DTW In Well-completion Depth Calculation

Oilfield Completion Depth Calculation is one of the vital links to ensure the perforating quality. Oil layer will be fully shot open and reach the design output when it be calculated accurately, or oil reservoir will not be fully exposed, and cause huge economic losses because of error perforation.Oilfield needs log twice the Open Hole Logging and Cased Hole Logging. Completion Depth comes from Open Hole Logging. However, affected by the difference type of cable, the material of cable, the tensile strength of cable, the underground environment and the frictional resistance of logging equipment, the logging curves of two logging are not entirely consistent in depth, but produced a certain degree of offset. In order to eliminate this error, a DTW-based method of Oilfield Completion Depth Calculation is put forward in this paper.Dynamic time warping (hereinafter referred to as DTW) is suitable such as voice recognition and other time-series of similarity comparison, which is initially applied to text data matching and visual pattern recognition field of study. Research shows that this algorithm based on nonlinear bending technology can obtain high recognition and matching accuracy, especially when there is a certain degree of deviation on the timeline of the time series. DTW technology can be very good at dealing with the local displacement. Using DTW to Calculate Depth Error Values can increases the accuracy of analysis, and the quality of Depth Error Values is more reliable.The mainly work is as followed.Based on Activity Analysis of optimized reservoir divided. First choose appropriate logging curves and calculate the activity of curves. Then calculate integrated activity curve by weighted. After dividing the reservoir, an optimized reservoir divided is used to make sure the results of reservoir divided more ccurate.Based on DTW of oilfield completion depth calculation. An improved efficiency DTW algorithm was used, which avoides some repetition computation in the old algorithm. The improvement of the computing efficiency is obvious.By use of Activity Analysis to locate the interface of oil layers on the standard curves, then take the interface of oil layers as the basic unit of moving. DTW is used to compare the similarity. This method reduces the computational volume, saves computer processing time and increases the accuracy of analysis, and the quality of Depth Error Values is more reliable.Based on relative amplitude of collar depth identification. According to the features, of collar signal, using the relative amplitude method to automatic track and identify collar depth, especially to accurately identify the collar depth which are effected magnetized collar. Application result shows this new technology is better than simple threshold, which ensures the perforation precision, safety, reliability and Reduces the labor intensity.