【摘要】 动校正叠加、F-X预测去噪、KL变换三种方法都能在一定程度上提高地震数据的信噪比和分辨率,但在去噪的同时都会在不同程度上削弱或者破坏振幅随炮检距变化的关系。为此,本文提出了一种保持叠前AVO属性的噪声衰减方法,假设叠前地震数据有效信号在各个地震道上满足简化的佐普里兹三项式方程,不满足该方程的地震信号被认为是噪声而被衰减。简化的佐普里兹三项式方程严格遵循了振幅随炮检距变化的关系,因此去噪后数据信噪比有显著提高,同时保持了原叠前数据各地震道的相对振幅关系,具有较高的保真度。理论模型试算和实际地震数据去噪结果表明,文中方法在去噪过程中不会改变信号振幅随炮检距变化的关系,对于低信噪比资料的叠前AVO属性提取非常有利。更多还原

【Abstract】 All the 3 method,NMO stacking method,F-X prediction denoise method and K-L transform method could to a certain extent raise both the signal to noise ratio and the resolution for seismic data,however while they remove the noise they also could to a certain extent weaken or damage the relationship of amplitude verse offset.In order to solve the problem the authors of this paper proposed a denoise method in which pre-stack AVO attribute could be preserved.It was assumed in this paper that the effective signals of the pre-stack seismic data meet the simplified Zoeppritz trinomial equation over each of the seismic traces,then all the seismic signals which do not meet the equation were regarded as noise and were suppressed.The simplified Zoeppritz trinomial equation strictly follows the relationship of amplitude verse offset,as a result the signal to noise ratio after denoise was greatly raised,the relative amplitude relation with the pre-stack data was preserved,and the fidelity was greatly improved.The theoretical model trial calculation and field date processing results show that the new method proposed in this paper does not change the relationship of amplitude verse offset,therefore the advantage of the new method is especially favorable for the extraction of the pre-stack attributes for seismic data with low signal to noise ratio.更多还原