论文部分内容阅读
在声波测井及地震处理中,根据组合波形数据估算波速(或慢度)是基本而且非常重要的过程。已开发了一种能处理含有多种波型的组合波形数据的预测方法。这些波型可能在时间和频率上重叠,而且用常规技术不能使其分开,采用这种新技术,一个接收器的波形可由其它接收器的波形数据组合,应用时间域预测理论来模拟。其假设组合数据包含许多传播模式。本文阐述了一种使预测波形与实测波形之间匹配最佳化的简化过程,从而得到通过组合的波型的慢度估算。最重要的是,用包含所有可能数据组合的整个组合数据集,可直接在时间域中进行优化。这种方法有效地降低了噪声影响并增强了估算的稳健性。而且,考虑到所要分析的波形特性,用估算的慢度值形成一个过程,从而将组合数据分解成各个波形。本文用实例证明该技术从复合波形数据中提取波慢度的能力。
In sonic logging and seismic processing, estimating the wave speed (or slowness) from the combined waveform data is a fundamental and very important process. A prediction method that can process combined waveform data containing multiple waveforms has been developed. These waveforms may overlap in time and frequency, and can not be separated by conventional techniques. With this new technique, the waveform of one receiver can be modeled by the waveform data of other receivers, using the theory of time domain prediction. It assumes that the combined data contains many propagation modes. This article describes a simplified procedure for optimizing the match between a predicted waveform and a measured waveform, resulting in a slow estimate of the combined waveform. Most importantly, you can optimize directly in the time domain with an entire combined dataset that contains all possible combinations of data. This method effectively reduces the noise impact and enhances the robustness of the estimation. Also, taking into account the characteristics of the waveform to be analyzed, a process is formed using the estimated slowness value to decompose the combined data into individual waveforms. In this paper, an example is given to demonstrate the ability of this technique to extract wave slowness from composite waveform data.