... was an opportunity to address some of the challenges in working with model ensembles and 4D seismic data. The main practical objective for the study was to test if new hypothesis about the depth of top reservoir and alternative parametrization could improve ensemble coverage of observed production. The large amounts...

... Hardening is clearly seen on the cross sections for each timestep in both reservoirs, indicating the seismic data has sufficient bandwidth to resolve each event in this area. This hardening is associated with the displacement of oil by water due to the water injection and is limited to the areas...

... Despite the appealing features of tracer and 4D seismic data, they are still often qualitatively employed for reservoir monitoring and management in the petroleum industry, but are relatively less used in a quantitative way, due to various reasons arising from both theoretical understands...

... in getting the representation coefficients and the built-in capacity in estimating noise levels (in terms of standard deviations) of observed seismic data AI° through wavelet-based denoising ( Luo and Bhakta, 2017 ). On the other hand, wavelet basis functions are independent of the dataset under...

... because of irregular channel patterns. With the accessibility of 3D or 4D seismic data, there has been great interest in using machine learning models for seismic reservoir characterization such as facies classification and determining the nonlinear relationship between reservoir properties and seismic...

... Abstract This work presents the use of Dictionary Learning methods for a sparse representation of 4D seismic data in history matching. We consider a trade-off between the number of coefficients retained in the sparse data representation, the computational cost, and how well we can capture...

... reservoir monitoring (PRM) system was installed in 2014. By today seven PRM acquisitions have been completed. The data that is used in this study includes both 3D and 4D seismic data, production logging tool data (PLTs), pressure data and production history data. A cross-disciplinary history matching...

... data in use. In contrast, correlation-based localization not only works well to prevent ensemble collapse, and also renders good history matching results. machine learning Artificial Intelligence seismic data localization scheme Upstream Oil & Gas model variable poro field ensemble...

... production data (e.g., bottom hole pressures, oil production rates, water cuts etc.) into reservoir models. Production data are often frequent in time, but have relatively sparse coverage in space. Complementary to production data, 4D (or time-lapse) seismic data are usually denser in space, and thus provide...

... promising data assimilation techniques employed in the industry is the ensemble Kalman filter (EnKF) because of its ability to deal with non-linear models at reasonable computational cost. In this paper we study the use of crosswell seismic data as an alternative to 4D seismic surveys in areas where...