Viscosity estimation of Athabasca bitumen in solvent injection process using genetic programming strategy
| Autor: | Amir H. Mohammadi, Alireza Baghban, Hojatollah Ebadi, Alireza Rostami |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Petroleum engineering
Renewable Energy Sustainability and the Environment 020209 energy Process (computing) Energy Engineering and Power Technology Genetic programming 02 engineering and technology Pipeline transport Variable (computer science) Fuel Technology 020401 chemical engineering Nuclear Energy and Engineering Asphalt Viscosity (programming) 0202 electrical engineering electronic engineering information engineering Fluid dynamics Sensitivity (control systems) 0204 chemical engineering Mathematics |
| Zdroj: | Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 40:922-928 |
| ISSN: | 1556-7230 1556-7036 |
| DOI: | 10.1080/15567036.2018.1465490 |
| Popis: | A large portion of the total oil-in-place around the globe is composed of heavy/ultra-heavy oils and bitumen. The main challenge for producing bitumen is its large viscosity inhibiting the easy fluid flow through the pipeline transportation and wellbore productions. In this study, genetic programming (GP) as a powerful strategy was utilized to develop a symbolic formula for accurate prediction of bitumen/n-tetradecane mixture viscosity. Several statistical and graphical tools were applied to demonstrate the supremacy of the GP model in comparison with the formerly published correlations. As a result, the GP model is the best acting model with the lowest average absolute relative deviation percent (AARD%) of 9.43. Based on the sensitivity analysis, it is proved that solvent concentration is the most affecting variable on output estimation. Lastly, it can be concluded that the proposed GP equation can be a good nominee in research studies dealing with bitumen recovery. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|