基于改进的GM(1,1)模型的高分辨率声波测井数据滤波处理

Filtering of high-resolution acoustic log data based on modified GM(1,1) model

(长江大学地球科学学院,湖北荆州434023)

Shi Dong,School of Geosciences,Yangtze University,Jingzhou 434023,China

高分辨率声波测井数据中包含了大量与地层无关的干扰信号,这些干扰信号会导致测井数据与岩心分析得到的孔隙度的匹配性变差,从而降低了所构建的孔隙度模型精度。在分析GM(1,1)模型灰色滤波方法和步骤的基础上,对其进行了改进,即通过修正模型中的k值来提高灰色模型的预测精度，并利用改进的GM(1,1)模型对实际高分辨率声波数据进行了滤波处理。结果表明,采用改进的GM(1,1)模型对高分辨声波测井数据进行滤波处理能够有效降低随机干扰噪声,提高声波测井数据与岩心分析数据的相关性。

High-resolution acoustic log data involves abundant interference signals,which decreases the adaptation of porosities acquired from logging and core analysis,therefore the accuracy of constructed porosity model is declined.After analyzing the gray filtering methods and procedures of GM(1,1) model,we modified the k value to improve the forecasting accuracy of gray model.The modified GM(1,1) model was applied on actual high-resolution acoustic data.The result shows that the random interference was effectively decreased;the coherence of acoustic logging data and core analysis data was increased.