Abstract
Developing an automated framework for real-time optimization (RTO) of the Steam Assisted Gravity Drainage (SAGD) process has significant potential because of the large number of parameters that must be monitored at a high frequency. However, the industry has not yet adopted a standard RTO framework for SAGD because of the intrinsic complexity of the process, the large number of parameters that must be monitored, harsh operating conditions, the lack of integration between various data acquisition systems, and the complex criteria required to optimize SAGD performance.
In this paper, a real-time monitoring workflow for SAGD is proposed that streams field data from multiple sources, including fiber optic distributed temperature sensing (DTS) directly into an engineering desktop application that has artificial intelligence (AI) and data mining capabilities. This system is used to derive advanced criteria to make decisions in a timely manner to improve the performance of the SAGD process.
It also demonstrates how subcool calculations can be effectively performed along the length of the horizontal well in real time and how the results are used to improve SAGD operation. Observations are compared "live" against simulated predictions from a multisegmented wellbore model that is fully coupled to a thermal/compositional reservoir simulator.
