A Novel Particle-Tracking Based Proxy for Capturing SAGD Production Features under Reservoir Heterogeneity
- Chang Gao (University of Alberta) | Zhiwei Ma (University of Alberta) | Juliana Y. Leung (University of Alberta)
- Document ID
- Society of Petroleum Engineers
- SPE Western Regional Meeting, 23-26 April, San Jose, California, USA
- Publication Date
- Document Type
- Conference Paper
- 2019. Society of Petroleum Engineers
- physics-based proxy, reservoir heterogeneity, simplified flow model, shale barriers, heavy oil
- 3 in the last 30 days
- 75 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 28.00|
Steam-assisted gravity drainage (SAGD) process is strongly impacted by the distributions of shale heterogeneities, which tend to impede the development of a steam chamber and potentially reduce oil production. Detailed compositional flow simulators are often employed to assess the impacts of reservoir heterogeneities on the steam chamber growth and to forecast production. To reduce the computational costs, machine learning techniques have been widely proposed in recent studies to develop various data-driven models. In all cases, a training data set consisting of many (>1000) synthetic simulation cases are required to achieve reasonable accuracy, especially in the case of 3D models. A suitable training data set should be large enough to sufficiently span the parameter space without exhaustively sampling cases with similar production characteristics. A novel physics-based proxy is proposed to approximate key SAGD production features in heterogeneous reservoirs.
A simplified flow model based on particle-tracking principles is developed to approximate how steam would travel in a three-dimensional heterogeneous reservoir. First, a large number of steam particles are launched, and the particles’ transition probabilities to nearby cells are calculated based on Darcy's law and energy balance: the directions of particles’ movement are governed by the intrinsic permeability, while the energy released by the steam particles upon condensing is used for heating the bitumen. Next, it is assumed that the steam particles would become immobile (instead of draining down) upon condensing. The temperature of nearby cells is updated after each time step. The process is repeated, and new steam particles are launched to represent a continuous injection. The locations traveled by the steam particles are tracked to quantify the chamber development as a function of time.
A set of 3D synthetic models using representative petrophysical properties and operating constraints extracted from Suncor's Firebag project is tested. The predicted steam chamber profiles match reasonably well to those obtained from detailed compositional simulations. As expected, shale barriers that are located in the near well region would have a more pronounced impact. Increasing the number and/or size of the shale barriers may delay the steam chamber development. However, the run time for this simplified model is much less than the conventional simulations.
This work provides a novel and fast particle-tracking based method to approximate the effects of complex shale heterogeneities on SAGD production and chamber development. It can be used to effectively screen a large number of shale heterogeneity realizations and to categorize them into different groups in accordance to their steam chamber development characteristics. It presents a significant potential to be integrated with many other data-driven approaches, such as cluster analysis, to visualize the (dis)similarity among a set of shale barrier configurations.
|File Size||1 MB||Number of Pages||17|
Zhang W., Youn S., & Doan Q. (2005). Understanding reservoir architectures and steam chamber growth at Christina Lake, Alberta, by using 4D seismic and crosswell seismic imaging. Paper presented at the SPR/PS-CIM/CHOA International Thermal Operations & Heavy Oil Symposium, Calgary, Alberta, Canada.
Zheng, J., Leung, J.Y., Sawatzky, R.P., and Alvarez, J.M. (2018a). An AI-based workflow for estimating shale barrier configurations from SAGD production histories. Neural Computing and Applications. https://doi.org/10.1007/s00521-018-3365-9.
Zheng, J., Leung, J. Y., Sawatzky, R. P., & Alvares, J. M. (2018b). A cluster-based approach for visualizing and quantifying the uncertainty in the impacts of uncertain shale barrier configurations on SAGD production. Paper presented at the SPE Canada Heavy Oil Technical Conference, Calgary, Canada.