Browsing by Author "Li, Sheng"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Quantification of Fugitive Area Emissions Using Optical Remote Sensing Computed Tomography and Inverse Dispersion Modeling Techniques
    (2022-12-17) Li, Sheng; Du, Ke; Johansen, Craig; Kim, Seonghwan; Ramirez-Serrano, Alejandro; Hugenholtz, Christopher; Hashisho, Zaher
    Reducing methane (CH4) emissions from anthropogenic activities is critical to climate change mitigation efforts. But there is still considerable uncertainty over the amount of fugitive CH4 emissions due to large-scale area sources and heterogeneous emission distributions. To improve the accuracy, a new hybrid method with high spatial and temporal resolutions is developed by combining optical remote sensing (ORS), computed tomography (CT), and inverse-dispersion modeling techniques. A multi-path scanning system is also developed using the horizontal radial plume mapping (HRPM) path geometry. The method adapts a Lagrangian stochastic (LS) dispersion model into CT reconstruction to derive emission distribution. Forward and backward LS models are developed and comprehensively compared to increase the model accuracy by supporting time-varying inputs. Controlled-release experiments of CH4 show this approach improves the accuracy by 5%-20% compared to the method using constant inputs. To address the challenges posed by limited path number and uneven sampling of HRPM path configuration, it uses a smooth algorithm of low third derivative (LTD) coupled with high-resolution grid division to improve the performance. By adapting the interpolation technique into CT reconstruction, a new minimal curvature (MC) algorithm based on the variational interpolation technique is formulated, showing significant improvement to the non-negative least squares (NNLS) algorithm and requiring approximately 65% computation time of LTD. To reconstruct realistic plume shape, a new Gaussian dispersion transformation algorithm is developed by introducing dispersion-related information. Numerical studies and controlled-release experiments show approximately 16% better performance on nearness than the NNLS and LTD algorithms. The combined method was evaluated through two controlled-release experiments of CH4, showing relative errors of only 2% and 3% compared to 5% - 175% for the single-path method. Field measurements of landfill CH4 emissions were first conducted in a scale-up site. Hotspots are successfully identified. The relative emission rate is 12.5%. Then, a test in a landfill biowindow area shows that more hotspots are successfully identified than the flux chamber method. The average emission rate over the site is 0.84 mg.m-2.s-1. The outcome of this research would bring broad application of the ORS-CT and inverse-dispersion techniques to other gases and sources.
  • Thumbnail Image
    ItemOpen Access
    Study of Oil Flow Mechanisms in Shale Reservoirs
    (2020-07-13) Li, Sheng; Dong, Mingzhe; Chen, Shengnan; Hassanzadeh, Hassan
    Shale oil is rapidly emerging as a significant and new unconventional resource. There is potential for shale oil production to spread globally over the next couple of decades. Similar to the unconventional formations, the shale oil formations are characterized with low porosity and low permeability. Besides, they have their own characteristics due to the existence of the organic material (kerogen). Therefore, the oil stored in a shale formation includes free oil which is primarily contained in inorganic and organic pores and ad- and absorbed oil in kerogen. The later can represent a significant portion of the total oil content but has not been clearly characterized. Instead of treating the shale oil formations as other tight oil formations, we have to understand the fundamental physics governing the flow of oil contained in kerogen. The main objective of this research is to understand how to unlock the oil in different storage status in shale formations. This objective is achieved by a comprehensive modeling approach which couples two of the most important mechanisms in petroleum engineering, fluid flow and mass transfer. At first, important petrophysical properties of shale formations are determined by considering the shale rock as a superposed system with both inorganic and organic matrices. The permeability of the organic pores is hundreds to thousands of times less than it of the inorganic pores and the flow of the oil in the organic pore system is accompanied by the dissolution/mobilization process of dissolved oil in kerogen. After that, the recovery of shale oil under reduced pressure is evaluated conceptually, by coupling the axial flow for free oil and radial diffusion for dissolved oil. It shows that both free oil and dissolved oil contribute to the total shale oil production. Depending on the reservoir properties, the bottleneck of the transport of the dissolved oil can be either low permeability or small diffusion coefficient. Finally, the releasing and flow of dissolved oil is simulated in the large-scale using the non-equilibrium mass transfer model in the commercial simulator, in which the parameters are identified to reflect the actual diffusion process of dissolved oil in kerogen. The original oil in place and oil production has been significantly increased by incorporating the dissolved oil in the reservoir simulation. The CO2 huff-n-puff process can slightly improve the shale oil production, but optimization of the related parameters is still to be desired. This dissertation will greatly increase the understanding of the organic pore system in shale, and the contribution of the dissolved oil to shale oil production. The results will also provide a technical reference for accurate prediction of shale oil production.