Date of Graduation
2016
Document Type
Thesis
Degree Type
MS
College
Statler College of Engineering and Mineral Resources
Department
Mechanical and Aerospace Engineering
Committee Chair
Ebrahim Fathi
Committee Co-Chair
Ebrahim Fathi
Committee Member
Shahab D Mohaghegh
Committee Member
Mehrdad Shahnam
Committee Member
Ali Takbiri Borujeni
Abstract
Using fossil fuel which has been grown dramatically during the recent century, causes an increase in greenhouse gas emission. The global warming issue pushes the engineers toward the cleaner type of energy like Hydrogen. Coal gasification is one of the cheapest methods to obtain Hydrogen. Coal gasification is a special case of more general problem called fluidized bed. In order to design and optimize a gasification process, a deep understanding of multiphase flow in a gasifier is needed. MFiX is a commercial multi-phase flow simulator which has been used to simulate the gas and solid transport and reaction in the gasifier using Computational Fluid Dynamics (CFD). Although simulating multiphase flow using commercial CFD software has a lot of flexibilities, it is really time-consuming and some other way could be implemented to reduce the run time. The effort of this project is to develop an alternate method to perform the same analysis but with much lower computational cost. A data-driven approach is used to build a smart proxy by employing the knowledge of Artificial Intelligence (AI) and Data Mining (DM).;In this project, a smart proxy will be developed to study and analyze the fluidized bed problem. This smart proxy is then will be used as a replicate of the CFD solver, with a good accuracy and faster speed. This proxy needs an incredible less amount of time in comparison to the CFD solver with a reasonable error (less than 10%). MATLAB neural network toolbox is used for training.;The goal of this project is to prove the concept of using AI&DM; for computational fluid dynamics especially predicting multiphase flow. Multiphase flow has a wide range of application in petroleum industry such as multi-phase flow in the wellbore, surface lines, and hydraulic fracturing such as proppant transport in the hydraulic fracture. This project opens a new way to accelerate the fluid dynamics analysis and reduce its costs.
Recommended Citation
Ansari, Amir, "Developing a Smart Proxy for Fluidized Bed Using Machine Learning" (2016). Graduate Theses, Dissertations, and Problem Reports. 5113.
https://researchrepository.wvu.edu/etd/5113