Identification of lithology using well logs : A case study

• 1、school of resources and geosciences,china university of mining and techonlogy,Xuzhou Jiangsu 221116
• 2、school of computer science and techonlogy,china university of mining and techonlogy,Xuzhou Jiangsu 221116

Abstract：In the exploration of mineral resources, the identification and division of lithology is a basic and important link. Log data contains rich geological information and can indirectly represent the lithology information of underground media. In practice, due to the influence of mud, well diameter, instrument and other measurement factors on log data, conventional lithology identification and classification has the defects of low efficiency and low accuracy. This paper takes well L44 of luling coal mine in anhui province as the research target, and identifies the lithology of the selected interval by using the characteristics of different lithology and different logging response. In order to improve the accuracy of identification, the wavelet multi-scale analysis method was used to extract the low-frequency and intermediate-frequency components of the signal. The low-frequency and intermediate-frequency components were used to reconstruct the curve to achieve the effect of eliminating noise. Then the low-frequency and intermediate-frequency components were used as the input, and the software k-means clustering analysis method was used to classify and identify lithology. Through identification, the target strata can be distinguished into sandstone, mudstone, sand-mud interbed and coal seam, and good results have been achieved.

Keywords： well logs wavelet multi-scale analysis SPSS K-means clustering analysis identify lithology