Reservoir Static Property Estimation Using Nearest-Neighbor Neural Network

Reservoir modeling is a critical process in the development of subsurface reservoirs, such as those found in oil and gas fields [1, 2]. Its primary objective is to characterize the spatial distribution of key reservoir properties, including porosity and permeability, which are essential for assessing reservoir reserves, evaluating properties, and determining overall potential [3, 4]. By integrating data from core samples, well logs, seismic surveys, and other sources, reservoir model offer a detailed representation of the spatial relationships between the essential reservoir properties. This modeling process is not only fundamental for understanding the current condition of the reservoir but also serves as the foundation for subsequent numerical simulations [5, 6] and the development of effective management strategies [7, 8]. Spatial interpolation is a widely used technique in reservoir modeling, involving the estimation of reservoir property distributions across a reservoir based on observations at discrete points [9].