The Renovation Explorer - Sensitivity Analysis and Selection of Parameters Report V2.0: Model verification and scenario reduction

Improving the sustainability of existing houses is critical to achieve climate goals. The Renovation Explorer (De Renovatieverkenner) project aims to develop simulation models that provide reliable and practical insights into the most cost-effective and sustainable renovation solutions for the existing house stock. To develop these models at the right level of detail, it is important to understand the influence of various parameters (house
characteristics, renovation measures, and user behaviour) on building performance.
The version 1.0 of this deliverable (previous version) focused on quantifying how key parameters affect simulated key performance indicators (KPIs) (heating demand and overheating). The study served three primary goals: 1) Prioritizing follow-up research by identifying highly influential input parameters that warrant greater model complexity and input certainty in future studies; 2) Creating a parameter database for housing variations, user behaviour, and renovation measures, with sampling weighted by parameter impact to improve prediction accuracy for the Renovation Explorer development; 3) Understanding uncertainty propagation from input parameters to simulated performance outcomes.
Version 2.0 expands this analysis to include four critical performance indicators: heating demand, overheating risk, peak energy demand, and indoor air quality (CO₂ exposure). The analysis now incorporates renovation packages, occupant behaviour profiles, and future climate scenarios to provide comprehensive guidance for long-term housing strategies.
The objectives of this report are:
• to verify and ensure the simulation model accurately captures the building physics.
• to identify the low-impact input parameters to reduce the number of scenarios (e.g. occupant behaviour or future weather) or renovation packages to simulate.
The model was verified, with simulated outputs aligning well with theoretical expectations across all KPIs. This confirms the computational framework's reliability and accuracy. Heating Demand is mainly driven by the building envelope, orientation, ventilation, internal heat gains, climate, thermal mass, and occupant behaviour. These inputs consistently ranked high in importance plots. Overheating is strongly influenced by building orientation, ventilation, internal gains, and user behaviour, especially window and shading use. These parameters also ranked high in the importance analysis. CO₂ Exposure is controlled by ventilation and occupancy patterns. Key influencing factors include window schedules, occupant numbers, ventilation types, and airtightness, providing clear guidance for optimizing indoor air quality. OPP is dominated by heating-related parameters (excluding gas boiler scenarios). Heating setpoint, system type, heated zones, and ventilation schedules are critical drivers. PV systems had minimal influence on peak loads, highlighting the need for energy storage or demand-side management to address peak demand.
To streamline simulation efforts and reduce computational demand, several input parameters were removed based on sensitivity analysis and redundancy in results. Specific window ventilation schedule is removed that showed similar performance trends across all KPIs. Future weather scenario (2100) is excluded due to limited impact outside of overheating and high uncertainty over long-term projections. The sampling frequency of insulation R-values (floor, wall, roof) is reduced for high R-values, where sensitivity levels off. Heating setpoint (N17_D20) is removed due to similarity with N17_D19. Heated Zones (0F, 1FS, 2F) are removed due to overlapping outcomes. These reductions preserve analytical accuracy while improving computational efficiency.