论文部分内容阅读
频率-波数域单程波算子能高效地模拟地震波在复杂介质中的传播,但是在描述波的大角度传播和速度横向扰动变化较大介质中传播的问题时仍然存在一定误差。这类误差是由于对单平方根算子使用Taylor展开式的近似程度不足所造成。为了进一步提高泰勒展开式的精确性,本文提出一种利用粒子群智能算法优化级数展开系数的高阶广义屏算子对单平方根算子的展开级数进行优化处理。新的偏移算法能在保持单程波偏移算法高效的前提下进一步提高偏移算子在大角度的成像精度和对强横向速度变化介质的适应性。通过脉冲响应实验,验证了基于粒子群算法优化级数的高阶广义屏算子能够提高常规的高阶广义屏算子的成像精度和成像角度。根据对二维SEG/EAGE盐丘模型的成像处理,基于粒子群算法优化级数的高阶广义屏算子对盐丘下面的断层取得了更高质量的成像,说明粒子群优化级数的高阶广义屏算子比常规的高阶广义屏算子具有更好的横向速度适应性。为了检验本文所提算法对实际资料的处理能力,我们利用常规的偏移处理技术和本文所提算法对一条海上二维数据进行了偏移成像处理,对比分析成像剖面发现本文所提算法描述了更加清晰的层位信息和更高质量的偏移剖面。本文所提算法能有效提高高阶广义屏偏移在广角度成像的能力,具有一定实际应用价值。
The frequency-wavenumber one-way wave operator can simulate the propagation of seismic waves in complex media efficiently, but there are still some errors in describing the propagation of large-angle waves and velocity transverse perturbations in larger media. This type of error is due to a lack of approximation of the Taylor Square expansion of a single square root operator. In order to further improve the accuracy of the Taylor expansion, this paper presents a new method of optimizing the expansion number of single square root operators by using the PSO algorithm to optimize the higher-order generalized screen operator of series expansion coefficients. The new migration algorithm can further improve the imaging accuracy of the migration operator at high angles and the adaptability to the media with strong lateral velocity changes, while keeping the one-way wave migration algorithm efficient. Through the impulse response experiments, it is verified that the higher order generalized screen operator based on particle swarm optimization algorithm can improve the imaging accuracy and imaging angle of the conventional high-order generalized screen operator. According to the imaging processing of two-dimensional SEG / EAGE model, higher-order generalized panel operator based on particle swarm optimization algorithm achieves higher-quality imaging of the fault beneath the salt dome, which shows that the optimal number of particle swarm optimization The generalized panel operator has better lateral velocity adaptability than the conventional high-order generalized panel operator. In order to test the ability of the proposed algorithm to process actual data, we use a conventional offset processing technique and the proposed algorithm to off-set an ocean two-dimensional data. Comparing the imaging profile, we find that the proposed algorithm described Clearer horizon information and higher quality offset profiles. The proposed algorithm can effectively improve the ability of high-resolution generalized screen migration in wide-angle imaging and has certain practical value.