ABSTRACT

Underground storage sites are critical reservoirs for storage and waste management, encompassing various substances, including groundwater, wastewater, carbon dioxide and hydrogen. Given the economic significance of these stored resources and the potential hazards the materials pose, robust monitoring of reservoir integrity is essential. Direct monitoring of such reservoirs often poses significant challenges due to the logistical and technical complexities involved. Consequently, indirect monitoring methods, such as those inferring changes based on surface deformation data, have become increasingly valuable. This study addresses the indirect monitoring of reservoir integrity using tiltmeter data by estimating and analysing reservoir pressure distribution. This inversion problem presents a unique set of challenges due to the ill-posed nature of the issue. To tackle this complexity, our research integrates the finite element method (FEM) with particle swarm optimisation (PSO) to estimate pressure distributions accurately and efficiently. The FEM model used is optimised by precomputing Green’s matrix, which encapsulates the geometric and physical properties of the reservoir, thereby enhancing computational efficiency and reducing time costs. This approach allows for the effective application of PSO, a robust optimisation method well-suited to addressing ill-posed problems characterised by fewer observations than unknown parameters. The system’s reliability was tested against complex reservoir models using synthetic data, achieving an error rate of less than 7% in the predicted pressure distributions. This demonstrates the efficacy of our method in providing reliable and timely insights into reservoir integrity, thereby enhancing the safety and efficiency of underground storage operations.