On gray-box modeling for virtual flow metering. (January 2022)

• Record Type: Journal Article
• Authors: Hotvedt, Mathilde; Grimstad, Bjarne; Ljungquist, Dag; Imsland, Lars
• Is Part Of: Control engineering practice. Volume 118(2022)
• Publication Date: 2022-01
• Subjects: Gray-box -- Virtual flow meter -- Multiphase flow -- Neural network; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2021.104974
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: A virtual flow meter (VFM) enables continuous prediction of flow rates in petroleum production systems. The predicted flow rates may aid the daily control and optimization of a petroleum asset. Gray-box modeling is an approach that combines mechanistic and data-driven modeling. The objective is to create a computationally feasible VFM for use in real-time applications, with high prediction accuracy and scientifically consistent behavior. This article investigates five different gray-box model types in an industrial case study using real, historical production data from 10 petroleum wells, spanning at most four years of production. The results are diverse with an oil flow rate prediction error in the range of 1.8%–40.6%. Further, the study casts light upon the nontrivial task of balancing learning from both physics and data. Therefore, providing general recommendations towards the suitability of different hybrid models is challenging. Nevertheless, the results are promising and indicate that gray-box VFMs can reduce the prediction error of a mechanistic VFM while remaining scientifically consistent. The findings motivate further experimentation with gray-box VFM models and suggest several future research directions to improve upon the performance and scientific consistency. Highlights: Five gray-box virtual flow meter models are developed. A case study on real production data from 10 wells is conducted. Gray-box modeling may improve the predictive performance over mechanistic models. Gray-box modeling may improve the scientific consistency over data-driven models. The results are diverse with prediction errors in the range 1.8%–40.6%.