Global parametric sensitivity analysis of numerical simulation for CO2 geological sequestration in saline aquifers: a case study of Yancheng Formation in Subei basin

Fei Zheng,Xiaoqing Shi,Jichun Wu,Liang Zhao,Yang Chen
DOI: https://doi.org/10.13278/j.cnki.jjuese.201401209
2014-01-01
Abstract:TOUGH2/ECO2N code was used to simulate the complex coupled processes taking place during and after CO2 injection in the deep saline sandstone aquifer of Lower Yancheng Formation in Subei basin. Morris, Sobol and EFAST methods were used to analyze the global sensitivity of parameters, i. e., kx, n, Slr, ��, C, 1/p0 and S, on three response variables: 1)the pressure located in the injection well; 2)the amount of gas-phase CO2 and 3)the spread distance of CO2 gas plume. Results from the Morris and Sobol methods show that n is the most sensitive parameter for the injection well pressure, kx is the most sensitive parameter both for the amount of gas-phase CO2 and the spread distance of CO2 gas plume. In summary, the sensitivity orders of these parameters are totally different for different corresponding variables. Comparative results from EFAST and Sobol show that the main and total effect obtained from the two quantitative sensitivity analysis methods are basically identical, however, EFAST is computationally more efficient than Sobol in numerical experiment.
What problem does this paper attempt to address?