Advanced Control and Optimization for the SAGD Process and Bitumen Upgrading
Date
2018-09-07
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Thermal recovery techniques, like Steam-Assisted Gravity Drainage (SAGD), are used to produce the majority of the crude bitumen, in Canada. However, suboptimal production techniques have led to the use of automatic control techniques for production, in recent times. Concurrently, while Proportional Integral Derivative and single variable Model Predictive Control (MPC) strategies have proven to be superior to manual control, they have resulted in comparable performance. Consequently, for improved performance, a novel Multi Input Multi Output (MIMO) MPC is presented in this thesis and compared with a Multi Input Single Output (MISO) MPC. The results indicated a 171% improvement in oil recovery for the novel MIMO MPC over the MISO MPC. This thesis also presents an optimization strategy for integrated design and schedule of a partial bitumen upgrader. The key consideration is in identifying that the design and operation problems are not mutually exclusive, but instead, synergistic in nature. Consequently, the research documented in this thesis, elucidates two formulations; maximizing profit and minimizing energy usage to highlight this concept. The results highlighted that both the design and scheduling decision variables change as per the medium term forecasts of the volatile commodity and energy pricing markets. Therefore, a design independent of the scheduling constraints or a schedule based on a fixed design may lead to suboptimal results in the design and/or schedule decision space(s). Finally, the last part of the thesis focuses on the optimal power management of a microgrid on a depleted SAGD facility, comprising of an Organic Rankine Cycle based turbine to convert the geothermal SAGD waste heat into electricity, a Gas Turbine, a Battery Storage System, the central grid, and the facility itself. Furthermore, this work also introduces a Kelly Criterion (KC) based microgrid scheduling technique, which is based on maximizing information gain and is independent of supply-demand relationships. Moreover, for this study, a wavelet network based forecasting technique is used to capture the electricity market volatility. The case study presented corroborated the hypothesis that the KC approach is independent of demand and supply forecasts, and is able to perform optimally in a highly volatile energy market.
Description
Keywords
Citation
Purkayastha, S. N. (2018). Advanced Control and Optimization for the SAGD Process and Bitumen Upgrading (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/32914