Fluid Velocity Prediction Inside Bubble Column Reactor Using ANFIS Algorithm Based on CFD Input Data
- PDF / 9,735,979 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 34 Downloads / 153 Views
RESEARCH ARTICLE-CHEMICAL ENGINEERING
Fluid Velocity Prediction Inside Bubble Column Reactor Using ANFIS Algorithm Based on CFD Input Data Quyen Nguyen1 · Iman Behroyan2 · Mashallah Rezakazemi3 · Saeed Shirazian4,5 Received: 3 December 2019 / Accepted: 29 April 2020 © King Fahd University of Petroleum & Minerals 2020
Abstract Since machine learning and smart methods can be used to study hydrodynamics in the bubble column reactor, it is possible to create highly intelligent bubble column reactors that have not been previously simulated and optimized them with computational fluid dynamics (CFD) methods. The previous studies considered the position of each node (in three directions) inside the bubble column reactor as the input in the artificial intelligence model. Machine learning methods have been used for processing big data related to the bubble column reactor. These big data are associated with the points inside the bubble column reactor, which represent the gas volume fraction and the fluid velocity in the x-direction. In this study, adaptive-network-based fuzzy inference system (ANFIS) was used to find out the relationship between the outputs of the bubble column reactor. The present study also intends to investigate the relationship between two outputs, namely the amount of gas in the bubble column reactor and the velocity of the fluid in the x-direction. Various parameters were investigated in this system, including the number of rules, the type of membership function, and the amount of input data. The mentioned parameters were regularly changed to find out the state where the system can achieve its intelligence. In this study, the best parameter that helped the system was the amount of data in the training process. The results showed that the lower the amount of data used in training, the better the prediction. Keywords Bubble column reactor · Numerical method · Soft computing · CFD · ANFIS · Multiphase flow List of symbols CD Coefficient of drag force for dispersed phase (–) CTD Turbulent dispersion coefficient for dispersed phase (–) C𝜀1 Turbulent dissipation energy equation (–) C𝜀2 Turbulent dissipation energy equation (–) C𝜇 Constant in turbulence modeling of dispersed phase (–) * Saeed Shirazian [email protected] 1
Institute of Research and Development, Duy Tan University, Danang 550000, Vietnam
2
Mechanical and Energy Engineering Department, Shahid Beheshti University, Tehran, Iran
3
Faculty of Chemical and Materials Engineering, Shahrood University of Technology, Shahrood, Iran
4
Department for Management of Science and Technology Development, Ton Duc Thang University, Ho Chi Minh City, Vietnam
5
Faculty of Applied Sciences, Ton Duc Thang University, Ho Chi Minh City, Vietnam
C𝜇,BI Constant in bubble-induced turbulence modeling of dispersed phase (–) dB Dispersed phase size (m) D Size of reactor (m) DS Ring sparger size (m) g Gravitational force in modeling (m s−2) H Height of reactor in modeling (m) k Turbulent kinetic energy for modeling of dispersed p
Data Loading...
