Alexandros G. Venetsanos

The DISCHA engineering tool has been extended to perform liquid hydrogen (LH2) tank-to-tank transfer operations. The tool uses Helmholtz free energy-based equations of state for hydrogen’s physical properties. The tanks can be modelled either with single-zone or multizone models, including heat transfer through the tank walls. The flow through the transfer lines is modelled using a method that accounts for phase change, friction, area change and heat transfer through the pipe walls. The maximum (choked) flow is calculated using the Possible Impossible Flow algorithm. Pumps can also be added to the transfer line. The present work presents the DISCHA tool’s recent validation as well as an investigation of the parameters that affect the tank-to-tank transfer most.

Reliable and timely estimation of air pollution exposure at high spatial and temporal resolution remains challenging in complex urban environments, where pollutant concentrations vary due to traffic emissions, urban morphology, and meteorological conditions. This study presents a physics-informed machine learning framework for near-real-time estimation of NO₂ concentrations at fine spatial scales. The approach combines a limited set of steady-state computational fluid dynamics (CFD) simulations with operational meteorological and air-quality data. CFD simulations under specific wind directions are first used to characterize site-specific dispersion patterns. These outputs are then scaled using hourly meteorological observations to generate physics-based concentration descriptors. A machine learning predictor, implemented using Random Forest and Extreme Gradient Boosting, is trained to refine these estimates by incorporating additional environmental and observational features. The method is applied to a 1 km × 1 km urban district in Antwerp, Belgium, within the FAIRMODE intercomparison framework. Validation against measurements from 105 passive samples collected over one month shows substantial improvement compared to standalone dispersion modeling, with coefficients of determination up to R² = 0.965 and reduced bias across locations. These findings demonstrate that integrating physical modeling with machine learning enables accurate and computationally efficient high-resolution exposure assessment in urban settings.

Reliable and timely estimation of air pollution exposure at high spatial and temporal resolution remains challenging in complex urban environments, where pollutant concentrations vary due to traffic emissions, urban morphology, and meteorological conditions. This study presents a physics-informed machine learning framework for near–real-time estimation of NO₂ concentrations at fine spatial scales. The approach combines a limited set of steady-state Computational Fluid Dynamics (CFD) simulations with operational meteorological and air-quality data. CFD simulations under specific wind directions are first used to characterize site-specific dispersion patterns. These outputs are then scaled using hourly meteorological observations to generate physics-based concentration descriptors. A machine learning predictor, implemented using Random Forest and Extreme Gradient Boosting, is trained to refine these estimates by incorporating additional environmental and observational features. The method is applied to a 1 km × 1 km urban district in Antwerp, Belgium, within the FAIRMODE intercomparison framework. Validation against measurements from 105 passive samplers collected over one month shows substantial improvement compared to standalone dispersion modeling, with coefficients of determination up to R² = 0.965 and reduced bias across locations. These findings demonstrate that integrating physical modeling with machine learning enables accurate and computationally efficient high-resolution exposure assessment in urban settings.

Accidental release from a hydrogen car tank in a confined space like a tunnel poses safety concerns. This Computational Fluid Dynamics (CFD) study focuses on the first seconds of such a release, which are the most critical. Hydrogen leaks through a Thermal Pressure Relief Device (TPRD), forms a high-speed jet that impinges on the street, spreads horizontally, recirculates under the chassis and fills the area below it in about one second. The “fresh-air entrainment effect” at the back of the car changes the concentrations under the chassis and results in the creation of two “tongues” of hydrogen at the rear corners of the car. Two other tongues are formed near the front sides of the vehicle. In general, after a few seconds, hydrogen starts moving upwards around the car mainly in the form of buoyant blister-like structures. The average hydrogen volume concentrations below the car have a maximum of 71%, which occurs at 2 s. The largest “equivalent stoichiometric flammable gas cloud size Q9” is 20.2 m³ at 2.7 s. Smaller TPRDs result in smaller hydrogen flow rates and smaller buoyant structures that are closer to the car. The investigation of the hydrogen dispersion during the initial stages of the leak and the identification of the physical phenomena that occur can be useful for the design of experiments, for the determination of the TPRD characteristics, for potential safety measures and for understanding the further distribution of the hydrogen cloud in the tunnel.

In April 2018, the International Maritime Organisation adopted an ambitious plan to contribute to the global efforts to reduce the Greenhouse Gas emissions, as set by the Paris Agreement, by targeting a 50% reduction in shipping’s Green House Gas emissions by 2050, benchmarked to 2008 levels. To meet these challenging goals, the maritime industry must introduce environmentally friendly fuels with negligible, or low SO_X, NO_X and CO₂ emissions. Ammonia use in maritime applications is considered promising, due to its high energy density, low flammability, easy storage and low production cost. Moreover, ammonia can be used as fuel in a variety of propulsors such as fuel cells and can be produced from renewable sources. As a result, ammonia can be used as a versatile marine fuel, exploiting the existing infrastructure, and having zero SO_X and CO₂ emissions. However, there are several challenges to overcome for ammonia to become a compelling fuel towards the decarbonisation of shipping. Such factors include the selection of the appropriate ammonia-fuelled power generator, the selection of the appropriate system safety assessment tool, and mitigating measures to address the hazards of ammonia. This paper discusses the state-of-the-art of ammonia fuelled fuel cells for marine applications and presents their potential, and challenges.