Stacking Approach for Soft Sensor Design in Steam Assisted Gravity Drainage Applications. Issue 21 (2018)

• Record Type: Journal Article
• Authors: Li, Chaoqun; Jan, Nabil Magbool; Huang, Biao
• Is Part Of: IFAC-PapersOnLine. Volume 51:Issue 21(2018)
• Page Start: 30
• Page End: 35
• Publication Date: 2018
• Subjects: soft sensor -- stacking model -- machine learning -- oil sands -- SAGD; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2018.09.388
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: This paper proposes a stacking soft sensor approach for the oil sands process, known as Steam Assisted Gravity Drainage (SAGD), which is an advanced underground bitumen extraction process. It involves developing multiple predictive models at the base level, followed by correcting the predictions using a higher level synthetic model. Multiple linear regression is utilized as the higher level model. Two case studies, Annual Reservoir Pressure soft sensor and Water Content soft sensor, are presented to demonstrate the feasibility and effectiveness of the proposed stacking approach for soft sensor design in SAGD process.