Simulation of a Commercial-Scale Entrained Flow Gasifier Using a Dynamic Reduced Order Model

The development of accurate, flexible, and robust dynamic reduced order models (ROMs) of entrained flow gasifiers (EFGs) is an important step toward greater commercialization of that technology. Previous work by the authors described the development, validation, and sensitivity analysis of such a ROM.(1, 2) This paper presents the results of dynamic simulation of a commercial-scale General Electric (GE or Texaco) gasifier and syngas cooling system. The base case for simulation is introduced, and the ROM is used to simulate six cases of dynamic gasifier operation. The objective of this work is to develop a computationally efficient simulator to assess steady-state and dynamic performance of entrained flow gasifiers under a wide range of realistic operating conditions. The six cases simulated are (1) removal of fluxant, (2) load following, (3) feed switching, (4) coal–petroleum coke cofiring, (5) coal–biomass cofiring, and (6) gasifier cold start. The results of dynamic simulation show that slagging properties (viscosity, temperature, and layer thickness) are always the last variables to reach steady state. This is primarily due to the large heat capacity of gasifier refractory walls. For the fuel-switching and cofiring cases, slag viscosity is found to be extremely sensitive to feedstock ash composition and gasifier refractory wall temperature. For cases 1–5, syngas production is predicted to reach steady state with very little lag. The ROM predicts gasifier cold start to occur over a time-scale of 50–60 h. The main reason for such a long start-up time is the requirement to limit maximum rates of heating of refractory faces. The simulation result agrees with industrial experience of start-up times of 2–3 days. All dynamic simulations took 15–45 min on a desktop personal computer, with the exception of the gasifier cold start case, which took 6 h.