AI-Based Multi-Objective Optimization of Building Retrofit Strategies for Heat Resilience in France: A Cost, Energy, and Thermal Comfort Approach
AI-Based Multi-Objective Optimization of Building Retrofit Strategies for Heat Resilience in France: A Cost, Energy, and Thermal Comfort Approach
baharsalari1
1) B.Sc. in Civil Engineering
IOI Code
:
ACCA-ACZHZB
Publication :
4th International Congress on Science, engineering & New Technologies(secongress.com)
Abstract :
The increasing frequency and intensity of heatwaves in Europe, particularly in France following the severe experience of the 2003 summer heatwave and similar events in recent years, have transformed the issue of building thermal resilience from a secondary concern into a strategic priority in energy and construction policymaking.
This study presents an artificial intelligence-based multi-objective optimization framework for selecting and combining retrofit strategies for existing buildings within the French climatic context. The proposed framework simultaneously considers three conflicting objectives: capital and operational costs, heating and cooling energy consumption, and occupants’ thermal comfort.
The proposed framework is based on dynamic thermal building simulation, using an approach similar to EnergyPlus tools, the integration of current and future climate data under climate-change scenarios through the 2050 and 2100 horizons, and the coupling of a Non-dominated Sorting Genetic Algorithm II (NSGA-II) with artificial neural networks as a surrogate model to reduce computational costs.
The retrofit strategies investigated include phase change materials (PCMs) in walls and floors, dynamic shading systems with angle control, a smart energy management system (EMS) for nighttime natural ventilation, and low-emissivity glazing with a low thermal transmittance coefficient.
The results obtained from a representative case study in three representative French climates—Paris, Lyon, and Marseille—indicate that the optimized combination of these strategies can reduce the Indoor Heat Degree (IOhD) index by more than 90% compared with the baseline condition, while maintaining a reasonable economic payback period.
The resulting Pareto front enables multi-criteria decision-making for designers, building managers, and policymakers.
This paper also discusses implementation challenges, modeling limitations, and future research directions in the field of data-driven and artificial intelligence-based thermal resilience.
Keywords : building retrofit multi objective optimization artificial intelligence nsga ii genetic algorithm phase change materials smart energy management france