Journal
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
Volume 69, Issue 5, Pages 1948-1961Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIM.2019.2923864
Keywords
Decision tree; dual-modality electrical tomography; fuzzy logic; gas-oil-water flow; multi-dimensional data fusion; multiphase flow visualization
Funding
- Engineering and Physical Sciences Research Council [EP/H023054/1]
- European Metrology Research Programme (ENG58MultiFlowMet) project Multiphase flow metrology in the Oil and Gas production
- European Metrology Programme for Innovation and Research Project - European Commission [16ENG07]
- European Metrology Programme for Innovation and Research Project - Euramet [16ENG07]
- Europea Union
- EPSRC [EP/H023054/1] Funding Source: UKRI
Ask authors/readers for more resources
This paper proposes a novel approach, whereby fuzzy logic and decision tree are utilized to overcome the challenges in analyzing images of gas-oil-water pipeline flow obtained using electrical resistance and capacitance dual-modality tomography. The first approach generates two axially stacked concentration images from two stacks of the cross-sectional concentration tomograms reconstructed from different modalities, respectively, and then registers two generated images in temporal and spatial terms. Afterward, a fuzzy logic method is applied to perform a pixel-level fusion to integrate the registered images based on the characteristics of electrical tomograms for multiphase pipeline flow. Later, a decision tree is utilized to derive the local concentration of each individual phase according to the fusion results. Using the data from real industrial cases, both feasibility and robustness of the proposed approach are demonstrated. In addition, the proposed approach also overcomes the limitations of conventional threshold-based methods on the request of a priori knowledge for the qualitative and quantitative analyses of gas-oil-water pipeline flow.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available