Improved numerical inverse Laplace transformation to improve the accuracy of type curve for analyzing well-testing data

Acta Geophysica

article

Kim, Song Chol¹  Song, Yong Il¹  Han, Chol Gwang¹ 

[1] Faculty of Mining Engineering, Kim Chaek University of Technology

关键词: Numerical inverse Laplace transformation;    Well-testing;    Type curve;    Convergence acceleration;    Stehfest;    Infinite series;   

DOI  :  10.1007/s11600-021-00585-7

学科分类：地球科学(综合)

来源: Polska Akademia Nauk * Instytut Geofizyki

【 摘 要 】

Analyzing well-testing data by the type-curve matching is a modern well-testing analysis method and is widely used in the petroleum and gas industry. By improving accuracy of type curve, we can get more accurate results from analyzing well-testing data, which provide a scientific base for development of oil, gas and water resources. By solving percolation equations, we can obtain type curves. The Laplace transformation methods are often used to solve them. In this paper, we improve the accuracy of type curve by improving the numerical inverse Laplace transformation (NILT) based on infinite series. We combine the NILT based on infinite series with Levin convergence acceleration and determine necessary parameters through numerical experiments to improve accuracy and speed. To verify this method, we compare the improved method with the Stehfest method using some functions such as trigonometric function. Type curves for analysis of well-testing data for the homogeneous reservoir with elastic outer boundary and a dual porosity reservoir are plotted and compared by using the improved numerical inversion and the Stehfest numerical inversion, respectively. These results show that type curves plotted by the improved method are less in vibration and fluctuation than ones plotted by the Stehfest method.

【 授权许可】

Unknown