High-pressure hydrogen storage requires up to 700 bar to achieve sufficient volumetric energy density, with reciprocating compressors preferred for their high flow capacity and oil-free operation. Meeting purity requirements while handling hydrogen's low molecular weight, high pressures, and dry seals presents major challenges for sealing elements. As a result, nonlubricated compressor technology must extend beyond conventional API 618 applications. Predicting leakage, piston ring wear, and seal performance under nonlubricated conditions is therefore essential to ensure the reliability, efficiency, and safety of hydrogen compression systems. To address these challenges, a dedicated test campaign was developed to evaluate sealing performance under representative operating conditions using a unique setup. The setup features a special single-acting, nonlubricated cylinder mounted on a vertical compressor operating with helium at pressures above 200 bar. Extensive instrumentation was implemented, including 45 dynamic pressure probes positioned along the cylinder liner to capture instantaneous pressure profiles in the gaps between seals, piston, and liner, as well as six dynamic pressure probes positioned along the pressure packing. Since fixed sensors measure within a moving volume, a tailored procedure was established to reconstruct pressure signals inside the moving gaps. The results reveal how pressure drop is distributed among sealing elements throughout the compression cycle, depending on the operating conditions, and highlight the distinct operating principles of different seal typologies.
Natural Gas/Ammonia (NG/NH3) fuel blends are a promising route to reduce CO2 emissions from gas turbines while leveraging existing NG infrastructure, but their auto-ignition behavior at elevated pressures is not yet well constrained. This work reports high-pressure shock tube ignition delay time (IDT) measurements for NG/NH3/Air mixtures at gas turbine-relevant conditions. Experiments were performed at pressures of 5, 10, and 25 bars, temperatures from 1500 to 2100 K, with equivalence ratios of Phi = 1.0, 1.2, and 1.4, and argon-diluted fuel compositions spanning pure NH3 to NG-rich blends. Experiments were carried out in two shock tube facilities at the University of Central Florida, with ignition being monitored via OH* emission chemiluminescence. The experimental data is compared against simulations using a widely employed nitrogen chemical kinetic mechanism from Glarborg et al., a mechanism by the CRECK modeling group, and a newly developed University of Central Florida (UCF) NG/NH3 mechanism. Across all mixtures, IDTs decrease monotonically with increasing temperature and pressure. Increasing pressure from 5 to 10 and 25 bars reduced IDTs for similar temperatures by up to 40% and 60%, respectively, depending on mixture composition. The addition of natural gas (NG) substantially shortens the IDT relative to pure NH3, with the largest incremental benefit at the lowest NG fractions and diminishing returns at higher NG fractions. The pressure sensitivity of ignition is strongest for NH3-rich mixtures and weakens as the NG fraction increases, while changes in equivalence ratio have a comparatively modest effect over the range of Phi = 1.0-1.4. The mechanisms reproduce the overall temperature and composition trends but show systematic deviations for NG-rich mixtures and some disagreements at specific pressure and temperature points in NH3-rich mixtures. Sensitivity analyses based on OH* production indicate that ignition is primarily controlled by CH3 and C2 radical chain-branching pathways in mixtures with NG, while NH3 chemistry mainly influences ignition through radical scavenging and NOx-forming channels. The present measurements provide new high-pressure validation targets for NG/NH3 kinetic mechanisms and help identify specific reaction classes that require further refinement for predictive simulations of low-carbon gas turbine combustion.
Predictive models for aircraft engines are being developed to forecast engine health conditions based on available operational data. In-service engine data will be the most valuable information to build such a model but may not always be accessible or sufficient. A pre-trained model can be a possible alternative, if it can be extended to new and other engine families. The objective of this paper is to demonstrate knowledge transfer by evaluating a predictive engine health model across different applications of engines. A predictive engine health model was developed using Machine Learning techniques applied to engine health monitoring (EHM) data gathered from the fleets of three different turbofan engines. Each model was built exclusively by training on the EHM data from each fleet of turbofan engines. The pre-trained models made predictions for their source (baseline case), and for other fleets of turbofan engines (transfer knowledge cases). The results show that the transfer knowledge cases have mean root-mean-square-error (RMSE) values, between 5 degrees C and 8 degrees C, which is about 2 times higher than the baseline case. The baseline case, represented by a smaller interquartile range (IQR), between 0.4 degrees C and 1.6 degrees C, have less variation in the prediction results than that of the transfer knowledge cases which have IQR values ranging from 1.6 degrees C to 6 degrees C. Such results indicate a generative predictive engine health framework may be developed with the capacity to be scaled across multiple gas turbine engine classes and applications.
Hydrogen is a key pathway for gas turbine decarbonization and optionality in the electricity and industrial sectors where energy demand is rising rapidly. Several demonstrations have shown that hydrogen blending into natural gas is successful in currently operating gas turbine hardware. However, due to significant differences in key flame properties like flame speed and ignition delay time between hydrogen and natural gas, the use of hydrogen can affect operability in gas turbine combustors. Issues related to flameholding have been addressed through the development of fuel-flexible combustors with multi-element arrays where many smaller flames allow for better flashback resistance than traditional combustor designs. However, issues related to thermoacoustic instability may still arise in these multi-element configurations. The goal of this work is to understand fundamental flame response behaviors in several multi-element configurations with blends of natural gas and hydrogen ranging from 100% natural gas to 100% hydrogen. To achieve this goal, flame transfer functions (FTFs) have been measured over a range of operating conditions and fuel blends. Global OH* chemiluminescence is used as a marker of heat release rate oscillations and a two-microphone method is used to measure the input velocity oscillation amplitude over a range of frequencies between 100 and 1000 Hz. Results show marked differences in the flame response as the fuel composition is varied, including variations in both the peak response Strouhal numbers as well as the gain of the FTF. High-speed OH* chemiluminescence imaging is used to explain some of the differences in the flame response with different fuel compositions and multi-element injector configurations.
This study experimentally investigates the impact of biokerosene blends derived from used cooking oil (UCO) on the performance and emissions of a microgas turbine, assessing their suitability as drop-in sustainable aviation fuels (SAFs). Engine tests were performed using jet-A and UCO-based biojet blends at volumetric ratios of 10%, 25%, 50%, and 60% in a fully instrumented and carefully calibrated small-scale turbojet test rig, capturing undiluted emissions as well as detailed performance parameters. Across the full operating range, thrust, fuel consumption, air mass flow, compressor pressure ratio, combustor pressure losses, exhaust gas temperature (EGT), and gaseous emissions (CO2, O2, NOx, CO) were systematically analyzed. Results show that overall engine performance remained effectively unchanged, with observed variations remaining within typical measurement uncertainty. Importantly, increased biojet content correlated with reductions of up to 16% in NOx and 12% in CO emissions, effects linked to enhanced combustion uniformity quantified via circumferential EGT deviation. This work therefore presents the first experimental demonstration of this correlation in a microgas turbine, alongside the first combustor pressure loss data for multiple biojet/jet-A blends. The findings clearly indicate that UCO-derived biojet blends up to 60% offer tangible environmental benefits without compromising engine performance, while also providing novel insights into fuel-air mixing effects directly relevant for future sustainable aviation fuel (SAF) deployment.
