【摘要】 声波测井是油气勘探中的重要技术之一，纵波时差可用于岩性划分、力学参数计算、优化钻完井方案等。在声波速度测井现场作业中，纵波时差曲线的实时提取普遍采用首波到时门槛法，该方法在薄互层地层与非常规储层的应用中，存在纵波时差提取精度低、纵向分辨率不足的问题，难以满足薄互层精细识别与评价的工程需求。本文采用多源相似叠加法构建具有相同跨度的子阵列，结合慢度—时间相关法（Slowness—Time Coherence,STC），实现不同分辨率纵波时差提取，在此基础上，针对岩性变化剧烈地层中固定时窗提取不稳定的问题，提出了一种动态时窗方法，使开窗位置能够随地层声学特征变化自适应调整，最终形成了高分辨率纵波时差实时提取方法，并将此方法集成至中海油田服务股份有限公司的增强型成像测井系统（Enhanced Logging Imaging System,ELIS）中，在山西、新疆和墨西哥等不同地质条件区块开展了现场应用验证。研究结果表明：(1)多源相似叠加与常规STC方法能够有效提高纵波时差提取的纵向分辨率，针对0.5 ft*薄层，厚度识别准确率可达91.3%；(2)动态时窗显著提高了岩性变化剧烈地层中纵波时差提取的稳定性，同时增强了时差提取结果与自然伽马曲线之间的相关性；(3)高分辨率纵波时差实时提取方法的现场应用验证了该方法实时提取的纵波时差曲线在岩性剧烈变化地层中的鲁棒性与普适性，以及所得纵波时差曲线的准确性和高分辨率特征。结论认为，本文提出的高分辨率纵波时差实时提取方法能够有效获取薄层声学响应特征，为薄互层及非常规储层的精细识别与实时评价提供了一种高效、可靠的技术手段。更多还原

【Abstract】 Acoustic logging is one of the key technologies in oil and gas exploration, where P-wave slowness can be applied for lithology classification, mechanical parameter calculation, and optimization of drilling and completion schemes. Currently, field operations in acoustic velocity logging primarily employ real-time extraction methods based on first-arrival time thresholds for compressional wave slowness curves. However, these methods exhibit low slowness accuracy and vertical resolution in thin interbedded formations and unconventional reservoirs, making it difficult to identify thin interbeds. This study constructs subarrays with identical spans using a multi-source similarity superposition method and combines it with the slowness—time coherence(STC) method to achieve compressional wave slowness extraction at varying resolutions. On this basis, a dynamic time-window method is proposed to overcome the instability of fixed time-window extraction in formations with strong lithological variations, allowing the window position to adaptively adjust according to formation acoustic characteristics. A high-resolution P-wave slowness real-time extraction method is thus developed and integrated into the enhanced logging imaging system(ELIS) of China Oilfield Services Limited, and field applications are carried out in blocks with different geological conditions, including Shanxi, Xinjiang, and Mexico. The results indicate that:(1) The multi-source similarity stacking combined with the conventional STC method can effectively improve the vertical resolution of P-wave slowness extraction; for 0.5 ft thin layers, the thickness identification accuracy of the high-resolution slowness extraction method reaches 91.3%.(2) The dynamic time window significantly enhances the stability of P-wave slowness extraction in formations with strong lithological variation, while simultaneously strengthening the correlation between the extracted slowness results and the natural gamma-ray(GR) curve.(3) Field application of the high-resolution P-wave slowness real-time extraction method verifies the robustness and general applicability of the proposed method for real-time slowness extraction in formations with intense lithological variations, as well as the accuracy and high-resolution characteristics of the resulting slowness curves. It is concluded that the high-resolution real-time P-wave slowness extraction method proposed in this paper can effectively capture the acoustic responses of thin layers, providing an efficient and reliable technical approach for fine identification and real-time evaluation of thinly interbedded formations and unconventional reservoirs.更多还原