Scholarly Works
Research Output
Recent Publications
Vessel-Type Electrical Load Forecasting for Cold Ironing Berths: A Data-Driven Approach
Publishing year: 2026
Other
Preprints, 2026, 2026
Google Scholar
The P2P-Ironing Platform: Evaluating the Software Quality and User Acceptance of a Decentralized Cyber-Physical Marketplace for Shore-to-Ship Power
Publishing year: 2026
Other
Google Scholar
Hierarchical energy management for renewable energy communities using MPC and rule-based control with peer-to-peer energy trading
Publishing year: 2026
Journal article
Energy Conversion and Management, 2026, Volume 366
Scopus
Islanded and weakly interconnected power systems face increasing operational challenges as the penetration of renewable energy grows, particularly in environments with limited flexibility and reserve support. This study investigates a residential energy community in Cyprus using real-time prosumer data and proposes a hierarchical decentralized Energy Management System (EMS). The EMS was initially implemented as a rule-based Minimum Viable Prototype (MVP), enabling practical validation of system operation and data integrity under realistic conditions. Building on this foundation, the final EMS adopts a three-layer control architecture in which rule-based household storage operation and peer-to-peer energy trading are complemented by a community-level Model Predictive Control (MPC) strategy for shared battery management. The MPC leverages short-term net-energy forecasts to proactively schedule charging and discharging of the community battery with the objective of reducing grid energy imports and improving local renewable energy utilization, while respecting battery operational limits and regulatory constraints. Simulation results based on measured data demonstrate consistent improvements over the rule-based baseline, achieving up to 23.4% reduction in electricity cost, approximately 6%–7% reduction in grid energy imports, and a 1%–2% increase in community self-sufficiency. Although the reduction in imported energy is moderate, its cumulative impact over long-term operation lea...
Dynamic Response and Temporal Encoding Behavior of a Copper-Inducible Living Biosensor
Publishing year: 2026
Conference paper
2026 IEEE BioSensors Conference (BioSensors), 1-4, 2026, 2026
Google Scholar
Development of a “Transmitter” Cell in a Yeast-Based Testbed for Molecular Communications
Publishing year: 2026
Conference paper
Nanocom 2026 Proceedings of the 13th ACM International Conference on Nanoscale Computing and Communication, 2026
Scopus
Pages 123-124
The development of a living End-to-End (E2E) testbed for investigating and validating Molecular Communication (MC) at the microscale, remains a critical challenge. In this paper, we address this challenge, by reporting ongoing work towards the development and mathematical characterization of a “transmitter” cell, based on the yeast model organism, which can complement our previously developed “receiver” cell, for delivering a complete living E2E system. The significance of such a testbed stems from the genetic amenability of yeast which can facilitate future research activities as well as the high relevance to emerging biomedical applications. Yeast cells have been recently leveraged for the development of relevant biosensors for biomarker detection and information dissemination, being explored for their potential to act as MC receivers when integrated on ingestible smart capsules, while their integration on Organs-On-Chip underpins their role as transmitters relaying locally sensed data to information collection devices.
Towards a general Machine Learning model for energy profile predictions in Smart Energy Communities in Cyprus
Publishing year: 2025
Conference paper
Iop Conference Series Earth and Environmental Science, 2025, Volume 1558, Issue 1
Scopus
Algorithms developed for energy management systems in smart energy communities require accurate forecasting of the energy production from RES and load consumption profiles. The formulation and optimization of such algorithms may be multi objective aiming for example to maximize revenue from excess renewable energy or minimize the energy imported from the grid. This paper presents an approach to predict energy profiles using Long Short-Term Memory (LSTM) based neural network models for a house in a community in Nicosia, Cyprus. The study addresses two key forecasting tasks: predicting photovoltaic (PV) energy production and estimating load demand for individual prosumers or consumers. For PV production forecasting, the model incorporates weather parameters such as solar irradiance, ambient temperature, and humidity. For load forecasting, the model is trained on historical consumption patterns of the prosumer. The models' performances are evaluated using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). Results demonstrate that both LSTM models yield accurate and reliable forecasts suitable for effective energy management in smart communities.
113
Unique publications
23%
Open access (26 publications; 18 without information)
78.8%
With other institutions (89 publications; 18 without information)
H-Index Reports
| Index type | Scopus | Google Scholar | Web of Science |
|---|---|---|---|
| H-index | 19 | 23 | 1 |
| H5-index | 9 | 10 | 1 |
| H10-index | 16 | 19 | 1 |
Publications per year and source
No description yet.
- Faculty
- Faculty of Engineering and Technology
- Department
- Department of Electrical Engineering, Computer Engineering and Informatics
- Title
- Assistant Professor