Abstract
As emissions regulations become increasingly stringent, the natural gas industry is seeking to retrofit their integral compressor engine fleet with real-time emissions control technology. The primary pollutants of concern for these engines are oxides of nitrogen (NOx) and methane. Unfortunately, there is often a tradeoff in abating these two chemical species. This paper discusses a potential control strategy for reducing both simultaneously.
The strategy proposed in this work is two-tiered, consisting of slow-speed and cycle-by-cycle control of the fuel-air ratio. Its success hinges upon the ability to rapidly diagnose, learn, and predict the cycle-by-cycle thermodynamics and emissions of the engine. This work describes a zero-dimensional simulation created for this purpose. Simulated results for in-cylinder pressure and emissions are compared to experimental data for an actual engine.
The simulation achieved very good agreement with the measurements and performed very well as a diagnostic tool, which bodes well for both the first and second tiers of the proposed control strategy. Future work will involve cycle-by-cycle NOx modeling and investigating how the simulation can be used to learn and rapidly predict cycle-by-cycle composition.
