Damien Pantalos

Developed a multi-market optimization model for Battery Energy Storage Systems (BESS) operating in Belgium. The model enables storage operators to simultaneously access arbitrage markets (Day-Ahead and Imbalance) and ancillary service markets (FCR, aFRR).

Key methodology: Implemented in Pyomo using mixed-integer linear programming (MILP) with 15-minute resolution. The model captures operational constraints, round-trip efficiency losses, and battery degradation effects under multiple cycling scenarios. Market data from Elia (2024) was used to simulate real-world operation and validate the approach.

Key finding: Coordinated multi-market participation significantly improves BESS profitability over single-market strategies. Ancillary services provide stable baseline revenue while arbitrage exploits price volatility for additional gains.

Developed a Multilayer Perceptron (MLP) neural network to forecast electricity demand in Spain's peninsular region for the upcoming 2 hours, segmented into 30-minute intervals. The model was trained on real-world consumption data from February to August 2024.

Key methodology: Implemented a sliding window approach with a 25-hour lookback period (n=50) to capture temporal patterns. Integrated exogenous variables including temperature and wind speed to enhance predictive capabilities. The network architecture featured 3 hidden layers (128→64→32 neurons) with batch normalization and dropout regularization (0.2) to prevent overfitting.

Developed a comprehensive methodology to optimize heat pump sizing for UK residential buildings using load disaggregation techniques on historical energy consumption data coupled with regional weather information.

Key methodology: Implemented Non-Intrusive Load Monitoring (NILM) using NILMTK toolkit and deep learning approaches to disaggregate heating/cooling loads from total household consumption. Integrated weather data correlation to identify temperature-dependent patterns and calculate optimal heat pump capacities for both electrified and non-electrified homes.

Designed an innovative underground vertical farming system for Mars that addresses the unique challenges of Martian agriculture: extreme radiation, temperature fluctuations, and limited water resources. The system leverages underground lava tubes for natural radiation shielding and thermal stability.

Key innovations: Integrated precision farming technology with LED grow lights optimized for plant photosynthesis, closed-loop hydroponic systems for water recycling, and CO₂ extraction from the Martian atmosphere. Surface-mounted PV arrays provide sustainable power for the entire operation.

Developed a comprehensive multi-criteria optimisation framework for designing net-zero energy communities. The model integrates real-time meteorological data via APIs, energy generation/consumption profiles, and techno-economic parameters to evaluate renewable energy configurations.

Methodology: Analysed multiple scenarios combining solar PV, wind turbines, and battery storage systems. Optimization criteria included energy self-sufficiency, levelized cost of energy (LCOE), net present value (NPV), and carbon emission reductions. The tool helps communities identify the optimal balance between investment costs and environmental benefits.