Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Liquid-Fueled Reactor Network Model for Enhanced NOx Prediction in Gas Turbine Combustors

Philip John, Sourav Saha, Opeoluwa Owoyele

TL;DR

This work addresses the challenge of predicting NOx emissions in liquid-fueled gas turbine combustors with reduced computational cost. It introduces a liquid-fueled reactor network (LFRN) that embeds specialized evaporator/breakup and mixer reactors into a chemical-reaction-network framework, calibrated against high-fidelity CFD data using data-driven clustering and particle swarm optimization. The LFRN markedly improves NOx accuracy over gaseous CRNs, achieving NO and NO2 predictions within approximately 14% and 3% of CFD across varied inlet temperatures and fuel flows, while running in seconds on a single CPU core. This approach offers a powerful, efficient tool for rapid emissions assessment and design-space exploration, complementing CFD in the development of next-generation liquid-fueled gas turbine combustors.

Abstract

This study introduces a liquid-fueled reactor network (LFRN) framework for reduced-order modeling of gas turbine combustors. The proposed LFRN extends conventional gaseous-fueled reactor network methods by incorporating specialized reactors that account for spray breakup, droplet heating, and evaporation, thereby enabling the treatment of multiphase effects essential to liquid-fueled systems. Validation is performed against detailed computational fluid dynamics (CFD) simulations of a liquid-fueled can combustor, with parametric studies conducted across variations in inlet air temperature and fuel flow rate. Results show that the LFRN substantially reduces NOx prediction errors relative to gaseous reactor networks while maintaining accurate outlet temperature predictions. A sensitivity analysis on the number of clusters demonstrates progressive convergence toward the CFD predictions with increasing network complexity. In terms of computational efficiency, the LFRN achieves runtimes on the order of 1-10 seconds on a single CPU core, representing speed-ups generally exceeding 2000\texttimes compared to CFD. Overall, the findings demonstrate the potential of the LFRN as a computationally efficient reduced-order modeling tool that complements CFD to enable rapid emissions assessment and design-space exploration for liquid-fueled gas turbine combustors.

A Liquid-Fueled Reactor Network Model for Enhanced NOx Prediction in Gas Turbine Combustors

TL;DR

This work addresses the challenge of predicting NOx emissions in liquid-fueled gas turbine combustors with reduced computational cost. It introduces a liquid-fueled reactor network (LFRN) that embeds specialized evaporator/breakup and mixer reactors into a chemical-reaction-network framework, calibrated against high-fidelity CFD data using data-driven clustering and particle swarm optimization. The LFRN markedly improves NOx accuracy over gaseous CRNs, achieving NO and NO2 predictions within approximately 14% and 3% of CFD across varied inlet temperatures and fuel flows, while running in seconds on a single CPU core. This approach offers a powerful, efficient tool for rapid emissions assessment and design-space exploration, complementing CFD in the development of next-generation liquid-fueled gas turbine combustors.

Abstract

This study introduces a liquid-fueled reactor network (LFRN) framework for reduced-order modeling of gas turbine combustors. The proposed LFRN extends conventional gaseous-fueled reactor network methods by incorporating specialized reactors that account for spray breakup, droplet heating, and evaporation, thereby enabling the treatment of multiphase effects essential to liquid-fueled systems. Validation is performed against detailed computational fluid dynamics (CFD) simulations of a liquid-fueled can combustor, with parametric studies conducted across variations in inlet air temperature and fuel flow rate. Results show that the LFRN substantially reduces NOx prediction errors relative to gaseous reactor networks while maintaining accurate outlet temperature predictions. A sensitivity analysis on the number of clusters demonstrates progressive convergence toward the CFD predictions with increasing network complexity. In terms of computational efficiency, the LFRN achieves runtimes on the order of 1-10 seconds on a single CPU core, representing speed-ups generally exceeding 2000\texttimes compared to CFD. Overall, the findings demonstrate the potential of the LFRN as a computationally efficient reduced-order modeling tool that complements CFD to enable rapid emissions assessment and design-space exploration for liquid-fueled gas turbine combustors.
Paper Structure (21 sections, 31 equations, 12 figures, 1 table)

This paper contains 21 sections, 31 equations, 12 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methodology
  3. CFD Numerical Setup and Computational Domain
  4. Reactor Network Development
  5. Droplet breakup
  6. Droplet evaporation
  7. Thermal Evolution: Uniform Temperature Approach
  8. Integrated Process Coupling
  9. $k-$means Clustering and Optimization
  10. Computational Implementation
  11. Results and Discussion
  12. Comparing Droplet Breakup and Evaporation Timescales
  13. CFD Flow Field Characteristics and Flame Stabilization
  14. Reactor Network Performance Validation
  15. Sensitivity to Air Inlet Temperature
  16. ...and 6 more sections

Figures (12)

  • Figure 1: Schematic of the liquid-fueled swirl-stabilized combustor, illustrating the central fuel injector, surrounding swirled air inlet, and can-shaped combustion chamber.
  • Figure 2: Seven-cluster reactor network configuration showing individual reactors and mass transport connections between clusters.
  • Figure 3: Evolution of droplet radius during spray processes: breakup only (evaporation disabled, ) and evaporation only (breakup disabled, ).
  • Figure 4: Velocity flow field, Temperature, OH mass fraction, and NO mass fraction contours obtained from the CFD simulation.
  • Figure 5: $k$-means clustering partitioning of the combustor into seven zones: evaporation and breakup (), mixing (), flame 1 zone (PSR ), flame 2 zone (PSR ), recirculation zone 1 (PSR ), recirculation zone 2 (PSR ), and post-flame zone (PFR ).
  • ...and 7 more figures