Automated Process and Geometry Design Optimization of a Coal Combustion Reactor | www.esteco.com

Automated Process and Geometry Design Optimization of a Coal Combustion Reactor

Author(s): 
Sara Salahi (Graduate School New Brunswick Rutgers, The State University of New Jersey)

At the current situation of the world with the rising demand of energy consumption, increasing energy prices and environmental concerns, design of efficient combustion and gasification processes becomes an important engineering problem. Moreover, design of an efficient and environmental friendly system, with all the complexities and conflicting objectives is a challenging process which is not solvable by conventional design and optimization methods. Computational Fluid Dynamics (CFD) has been a powerful design tool in the past decade in the combustion and gasification fields. However, this method alone cannot be a suitable solution for engineering multi-objective design optimization problems because of its random and inefficient procedure which does not necessarily cover the entire design space. The purpose of the current study is to integrate CFD simulation of coal combustion with multi-objective optimizations in an automated fashion.

 

In order to achieve this goal, a coal combustion reactor with swirl flow has been considered. Coal particles are mixed with air and are injected into the reactor via four tangential inlets to create swirl flow in order to achieve better turbulence and better combustion. Simplified single phase and multi-phase reactions have been considered to simulate combustion process. All the steps of geometry creation, grid generation and CFD simulation have been integrated automatically using macro files to run in batch mode in an optimization platform, i.e., modeFRONTIER.

Three sets of multi-objective optimization problems have been solved with two, four and six input variables respectively. Each multi-objective optimization problem consists of 117 individual single objective problems solved by the SIMPLEX method. The ∈ constraint method has been implemented for multi-objective optimization. Two conflicting objectives have been selected for all optimization problems: minimizing NO mass fraction and CH4 mass fraction. It is noticed that the AF ratio has considerable influence on NO and CH4 mass fractions. Therefore, this parameter has been added as a constraint to the optimization problem. Results from all single objective optimizations have been summarized and graphed in a chart to obtain the Pareto Set. The Pareto Set proposes a set of optimal solutions for the multi-objective problems, i.e., a set of optimal process and geometry design input variables that can result in the least possible combination of emissions in the outlet.

Automated multi-objective optimization proposes a reliable and promising method to integrate CAD and CAE tools with optimization methods in an automated fashion to perform faster, more accurate, more efficient and more cost-effective designs in the field of coal combustion and gasification.

LOG IN TO DOWNLOAD