.
The need for decarbonization has become an urgent objective being pursued across borders and sectors. The energy sector is one example to follow, with the penetrations of renewable energy sources increasing year after year and already having an impact on carbon emissions. Despite their benefits, as these penetration levels grow, grid frequency is likely to be affected, namely in small, isolated power systems. Part of this instability is caused by the different types of connections to the grid used by most renewable energy sources. These converter-connected technologies do not contribute to the grid's inertia the same way synchronous generators do, therefore the difficulties in keeping the frequency under control. Focusing on the behaviour of the system's frequency, following some disturbances, a grid model of the Portuguese Terceira Island is created to assess the potential benefits of using hydrogen technologies (electrolysers and fuel cells) to provide grid services. In this paper, two scenarios are analyzed, the system's frequency response to large imbalances and steady-state imbalances stemming from load and wind power forecasting errors. The results reflect how the frequency behaviour is affected by the amount of the system's inertia, synchronous generators characteristics, and the presence of electrolysers and fuel cells. Hydrogen systems impact positively on the frequency containment after large disturbances, the impact related to the response to steady-state imbalances being lesser and prone to affect the stacks lifespan.
The New York Power Authority (NYPA), EPRI, and General Electric (GE) developed and executed a pilot project focused on hydrogen-fueled power generation. As part of the Low-Carbon Resources Initiative (LCRI), the companies jointly conducted a hydrogen blending project at NYPA's Brentwood Power Station. This collaborative effort demonstrated the burning of a hydrogen-natural gas blend on an LM6000 gas turbine (GT) to identify the resulting impact on combustion emissions (CO2, NOx, CO) and GT operation. The GT was operated on hydrogen blends ranging from 5 to 44% by volume. The successful test represents the first utility-scale hydrogen blending project in the state of New York, which is mandating a zero-emission electricity sector by 2040 and calling for an orderly and just transition to clean energy for a economy-wide carbon neutrality through the Climate Leadership and Community Protection Act.
The previous studies by the authors [1–2] focused on understanding the breakup mechanism of single or multiple liquid jets in non-uniform crossflows. Two different numerical approaches were taken; namely, the low-fidelity extended Madabhushi model and the high-fidelity VOF-DPM multiscale approach. These found that the high-fidelity approach, even without any fine-tuning, could predict the breakup mechanism and spray characteristics. Also, it was possible to fine-tune the parameters of the extended Madabhushi model based on experimental measurements or high-fidelity VOF-DPM results, whichever is more readily available. The present study is a continuation of the previous work where the applicability of the two spray generation approaches is used inside a combusting chamber where the crossflow is considerably hotter, causing the spray to evaporate and burn. Various unsteady LES simulations are performed with pure diesel and water-diesel mixture and injected in a heated nonuniform crossflow of 350°C inside a chamber at a pressure of 50 psi with a Momentum ratio (J) of ∼1.5. Due to experimental challenges, the momentum ratio was lowered (from the original planned J of 7.5) to have a stable flame and prevent blowout. As a result, the flame is near the bottom wall. This does create a modeling challenge to account for the heat loss effects. Due to the lack of droplet data such as Sauter mean diameter, volume flux, etc. for this combusting case, the authors rely on the comparison made by Feiz et al. [2]. Like previous studies by the authors, a modified version of the Madabhushi model proposed by Lambert et al. [25] is used here to simulate the jet breakup in a reacting flow simulation to get the spray regime. Alternatively, the authors have also generated the initial droplet data by running a VOF-DPM multiscale simulation with the LES turbulence model and explicit VOF for interface tracking. All the model parameters, whether those associated with turbulence modeling or with multiphase VOF modeling were retained at default values. Subsequently, reacting simulations using the single-step global mechanism of Kerosene (using C12H23 as a surrogate) for pure fuel injection (dry) and fuel+water injection (wet) are performed. Turbulence chemistry interaction is modeled using Eddy-Dissipation Model to calculate the flame shape and combustion products. Model tuning has been conducted using the University of Cincinnati Research data specifically designed for this configuration in partnership with General Electric Company. The Smagorinsky subgrid scale model with Wall adapting local eddy viscosity (WALE) near wall model is used for LES. The droplets are tracked using Ansys Fluent Discrete Particle Model (DPM). In this effort, the flame shape and extent of the flame near the wall have been captured and compared to experiments. Overall, the liquid penetration and flame shape are within reasonable accuracy.
