论文部分内容阅读
本文概述了复数道分析的目的在于求取表征地震波动力学特征的一些参数——瞬时振幅、瞬时相位、瞬时频率和视反射极性。这对解释油气藏和获取岩性信息有重要意义。 要获得较好的参数剖面,必须正确选择和改进某些计算公式,正确使用合理的处理方法。本文通过对合成记录的处理说明,对瞬时振幅做积分滤波处理可得到光滑可靠的瞬时振幅剖面,并可大大消除由随机干扰造成的假视极性。用反正弦公式代替常用的反正切公式求取瞬时相位,可获得与地震波周期一致的相位剖面,克服了相位畸变。利用经过积分滤波的瞬时振幅对瞬时频率做加权处理,可使瞬时频率更加稳定。 本文借用有限元素表示的射线轨迹法,在6912计算机上对一种常见的构造地质模型作了合成记录,用它计算了瞬时振幅、相位、频率和视极性剖面;分析了这些剖面的特征以及利用这些特征进行地震解释的可能性。 本文还对形成复数道的基本算法——离散希尔伯特变换的各种改进算法进行了比较,计算结果表明,在频率域中做希尔伯特变换较为有利。
This paper summarizes the purpose of complex tract analysis to find some parameters that characterize the seismic wave dynamics - instantaneous amplitude, instantaneous phase, instantaneous frequency, and apparent reflectivity. This is of great significance to the interpretation of oil and gas reservoirs and lithology. To get a better profile of the parameters, some formulas must be correctly selected and improved, and proper methods should be used. In this paper, through the processing of synthetic recording instructions, the instantaneous amplitude of the integrated filtering process can be smooth and reliable instantaneous amplitude profile, and can greatly eliminate the random interference caused by the pseudo-polar. By using the arcsine formula instead of the usual arc tangent formula to obtain the instantaneous phase, the phase profile consistent with the seismic wave cycle can be obtained and the phase distortion can be overcome. The instantaneous frequency can be more stable by weighting the instantaneous frequency using the instantaneous amplitude of the integrated filtering. In this paper, a finite element-based ray-trace method was used to synthesize a common structural geological model on a 6912 computer. The instantaneous amplitude, phase, frequency and apparent polar profile were calculated. The characteristics of the profiles and Use these features for earthquake interpretation possibilities. This paper also compares the various improved algorithms of discrete Hilbert transform, which form the basic principle of complex paths. The calculation results show that it is more advantageous to do Hilbert transform in the frequency domain.