Process Optimization and Robustness Analysis of Ammonia–Coal Co-Firing in a Pilot-Scale Fluidized Bed Reactor

Abstract

A computational fluid dynamics (CFD) model was coupled with an advanced statistical strategy combining the response surface method (RSM) and the propagation of error (PoE) approach to optimize and test the robustness of the co-firing of ammonia (NH3) and coal in a fluidized bed reactor for coal phase-out processes. The CFD model was validated under experimental results collected from a pilot fluidized bed reactor. A 3k full factorial design of nine computer simulations was performed using air staging and NH3 co-firing ratio as input factors. The selected responses were NO, NH3 and CO2 emissions generation. The findings were that the design of experiments (DoE) method allowed for determining the best operating conditions to achieve optimal operation. The optimization process identified the best-operating conditions to reach stable operation while minimizing harmful emissions. Through the implementation of desirability function and robustness, the optimal operating conditions that set the optimized responses for single optimization showed not to always imply the most stable set of values to operate the system. Robust operating conditions showed that maximum performance was attained at high air staging levels (around 40%) and through a balanced NH3 co-firing ratio (around 30%). The results of the combined multi-optimization process performance should provide engineers, researchers and professionals the ability to make smarter decisions in both pilot and industrial environments for emissions reduction for decarbonization in energy production processes.