基于最小二乘的时移地震高精度时间延迟校正方法研究

时间延迟的估算与校正是时移地震中的关键问题之一, 目前常用的拾取法、相关法、反演法等方法的精度仍然有待提高。针对此问题, 提出了一种基于最小二乘原理以及平滑约束和相位约束的时间延迟高精度校正方法。通过使参考地震道与校正之后的地震道之间振幅差异的平方和最小构建关于时间延迟的目标函数, 引入模型平滑项和相位约束项以改善反演效果, 使用高斯-牛顿法导出时间延迟的迭代反演方程。所提方法不仅可以实现时间延迟的高精度反演和校正, 而且计算效率高, 实际应用方便。建立理论模型合成时移地震记录, 测试了参数选取和储层厚度对校正结果的影响, 并与泰勒展开法、动态时间规整法等反演方法进行了对比, 结果表明, 基于最小二乘原理的时移地震高精度时间延迟校正方法能够实现时间延迟的高精度反演和校正。实际地震资料时间延迟的反演和校正结果表明, 该方法能够改善时移地震数据的重复性, 有利于进一步的数据解释, 证明了方法的可行性和有效性。

Time shifts refer to time-lapse differences in the two-way travel time of seismic waves in a reservoir and are often observed in time-lapse seismic data.Time shifts are primarily caused by seismic data acquisition, processing, and fluid variations resulting from oil and gas production in reservoirs.The estimation and correction of time shifts are key problems in time-lapse seismic monitoring and are related to the interpretation of time-lapse seismic data.Commonly used correction methods include picking, windowed correlation, and mismatch minimization.However, their accuracy needs to be improved.To address this problem, we proposed a method for high-precision time-shift correction with smoothing and phase constraints based on the principle of least squares.First, the objective function with respect to the time shifts was constructed by minimizing the sum of squares of the amplitude differences between the reference and corrected traces.Next, model smoothing and phase constraint terms were introduced into the objective function to improve the inversion results.Finally, the Gauss-Newton algorithm was used to derive an iterative equation for estimating the time shifts, which helped realize high-precision inversion and correction of time shifts.It also had high computational efficiency and was easy to implement in practical applications.A theoretical model was developed to generate synthetic time-lapse seismic records.A test was conducted to determine the influence of the parameter selection and reservoir thickness on the correction results.The method was also compared with other mismatch minimization methods, such as the Taylor series expansion and dynamic time warping methods.The test results have shown that this method could achieve high-precision estimation and correction of time shifts.Time-lapse field seismic data were used to perform the inversion and correction of time shifts to prove the feasibility and effectiveness of the method.The results have shown that this method could improve the repeatability of time-lapse seismic data and benefit further data interpretation.