慢度-时间相关法与遗传算法结合提取阵列声波测井高分辨率声波时差

TO EXTRACT HIGH-RESOLUTION TIME DIFFERENCE OF ARRAY ACOUSTIC LOGGING WITH COMBINATION OF SLOWNESS TIME CORRELATION AND GENETIC ALGORITHM

随着油气勘探和开发程度的逐渐深入，迫切需要解决薄层划分和厚层细分及薄层评价问题，因此有必要进行测井曲线的高分辨率处理，以提高薄层评价精度。交叉偶极子阵列声波测井处理软件提取的声波时差的纵向分辨率不能满足现场的要求。在充分调研了国内外大量声波时差曲线高分辨率处理方法的基础上，提出了基于遗传算法提取阵列声波测井高分辨率时差的方法。该方法选择跨同一深度地层所有最小子阵列（二个接收器），然后用慢度一时间相关法与遗传算法结合提取每个子阵列的声波时差，最后取其算术平均值作为最终时差。用该方法处理了大庆油田2口阵列声波测井资料，并与高分辨率声波对比，具有较好的一致性，表明该方法可行、实用。

With deep exploration and development for oil and gas,it is urgent to solve the problems:dividing thin layer,subdividing thick layers and evaluating thin layers.Therefore,ii is necessary to process logs in order to improve the accuracy of thin layer evaluation.The axial resolution of acoustic time difference extracted with the processing software for crossed dipoles array acoustic logging can not satisfy field requirment.A new integrate processing technique is proposed based on full survey and study of many processing methods at home and abroad.In the method all the minimum subarrays(two receivers) at the same depth of formation are selected,and then acoustic time difference is extracted with combination of time correlation and genetic algorithm.Finally the arithmetic mean value is taken as ultimate time difference.It has been applied in 2 wells of array acoustic logging data processing in Daqing oilfield.Through comparison with high resolution acoustic logging it indicates that this method is feasible and practical.