Publications
-
Towards Eco-Efficient AI: Hybrid Data Strategies for BIPV Energy Prediction
Accurate prediction of energy production from building-integrated photovoltaic (BIPV) systems is essential for optimizing building energy use and supporting decarbonization strategies. Synthetic data provides a valuable foundation for model development, particularly when real measurements are limited, but validation on operational systems remains critical for reliable deployment. In this study, we propose a hybrid data approach, […]
-
Reliability based design optimization for multiaxial fatigue damage analysis using Matsubaraˆas criterion developed in the frequency domain of structures under random vibrations
This study aims to an optimal design of a structure subject to random vibrations, considering fatigue damage, that guarantees a required reliability level under uncertainties parameters. In practical applications, structures are exposed to multiaxial random loading. While numerous fatigue life prediction criteria have been developed primarily in the time domain, frequency domain methods are preferred […]
-
Potential of Generative Artificial Intelligence in Knowledge-Based Predictive Maintenance for Aircraft Engines
Predictive maintenance based on remaining useful life (RUL) estimation is widely recognized as a promising strategy for monitoring the health of critical systems such as aircraft engines, anticipating failures, and optimizing maintenance planning. A variety of approaches have been proposed in the literature, including data-driven, physics-based, and knowledgebased methods. Among them, deep learning-based methods have […]
-
Industrial Metaverse Architecture in the Automotive Sector
The transition from Industry 4.0 to Industry 5.0 aims to develop production systems that are more people-centered, sustainable, and resilient. An emerging concept that supports this evolution is the Industrial Metaverse, which integrates technologies such as the Industrial Internet of Things (IIoT), Big Data, and virtual environments to bridge physical and digital worlds. As this […]
-
A Hybrid Approach to Building Thermal Modeling Using Physics-Based Machine Learning
Buildings account for approximately 30% of primary energy consumption, mainly due to Heating, Ventilation, and Air Conditioning (HVAC) systems. Reactive controllers can be used to manage these systems optimally, however, their performance depends strongly on the accuracy of building thermal models. In this study, a hybrid physics-informed machine learning (PIML) approach is proposed to improve […]
-
On-Policy vs. Off-Policy HVAC Control: Comparing PPO and SAC–Gumbel in EnergyPlus
We compare two reinforcement learning methods for HVAC control in a university amphitheater simulated in EnergyPlus: Proximal Policy Optimization (PPO, on-policy) and SAC-Gumbel (off-policy). We run two experiments. First, a weekly adaptation test trains each agent for 50 episodes using the first week of January in Luxembourg. Second, a year-long generalization test trains on a […]
-
Bio-stabilised earthen materials: a perspective on the potential contribution to climate change adaptation and mitigation
With its potential for low embodied carbon, bio-stabilised earth offers a sustainable alternative to traditional carbon-intensive building materials. Moreover, unlike lime and cement stabilisation, bio-stabilisation methods hold the promise to improve the durability of earthen materials while retaining their moisture buffering capacity and recyclability. Despite the promising characteristics of bio-stabilised earth, research on the topic […]
-
Combining Client-Based Anomaly Detection and Federated Learning for Energy Forecasting in Smart Buildings
In today’s interconnected world, energy consumption forecasting faces challenges due to client-side anomalies in time-series data. Federated Learning (FL) offers a decentralized solution by forecasting without directly accessing user data. However, the effectiveness of the global model can decline if local anomalies are not properly managed. We propose our lightweight framework EIF-FL: Elliptic envelope and […]
-
DistillH-Mamba: A Hypergraph-Mamba-Based Knowledge Distillation Model for Efficient Impact Fall Detection
Falls among the elderly represent a significant public health concern due to their prevalence, consequences, and societal burden. While deep learning has improved fall detection, accurately identifying impact moments (when an individual hits the ground) remains challenging. Additionally, current algorithms often rely on complex models with high computational demands, limiting real-time deployment feasibility. In this […]
-
A Stochastic Model for the Bike-sharing
Bike Sharing Systems (BSS) offer a sustainable and flexible solution to urban mobility, but their rapid growth as a viable and popular transportation alternative has exposed major challenges. Due to asymmetric user flows throughout the day they suffer from chronic imbalances in bike distribution, badly impacting both the system reliability and user satisfaction. In this […]
-
Evaluating Robustness of 3D Gaussian Splatting–Based 6D Camera Pose Refinement Under Degraded Conditions for Lightly Textured Industrial Synthetic Objects
In this paper, 6D camera pose refinement is explored using 3D Gaussian Splatting (3DGS) on lightly textured industrial object datasets. The study employs datasets generated with Unity 3D rendering software, featuring objects such as a bicycle, MiR robot, Tiago robot, and UR robotic arm, each captured with ground-truth intrinsic and extrinsic camera parameters. A 3DGS […]
-
Attention Makes HVAC Control More Efficient
Heating, ventilation, and air-conditioning (HVAC) systems account for around 16.4% of global final energy consumption and about 14% of global operational CO2 emissions. Controlling them is a partially observable, sequential decision problem: relying solely on instantaneous sensor readings as inputs overlooks the full sequence of past conditions that shape future dynamics. To tackle this challenge […]