Digital Twin Framework
Data-Driven Digital Twin
Description
The Data-Driven Digital Twin leverages real-life data collected from pilot sites to learn, model and predict energy-related building operations and occupant behaviour, particularly when detailed information is unavailable from static sources like BIM files.
Based on machine learning algorithms, it simulates the usage patterns of building systems – such as HVAC and electric water heaters – and generates dynamic behavioural profiles including occupancy and comfort preferences. It also supports human-centric KPI calculation and predictive maintenance, complementing and enhancing the accuracy of the simulations produced by the Physics-Based Digital Twin (below).
Developers
Hypertech Sustainability Research & Technology Center
www.hsrt.gr
Features & Benefits
The Data-Driven Digital Twin includes four core sub-components:
- ML-Based Asset Profiling Engine
- Occupancy & Comfort Profiling Engine
- Human-centric KPI Calculator
- Predictive Maintenance Analytics module
These components enable continuous learning from actual energy consumption and environmental data, allowing for precise profiling of loads and occupant behaviour. This results in improved simulation accuracy, enhanced comfort modelling and early detection of HVAC faults for proactive maintenance.
Facility managers, ESCOs and researchers benefit from its ability to provide actionable insights without relying solely on detailed technical documentation – making it ideal for retrofitting scenarios or incomplete digital models.
Physics-Based Digital Twin
Description
The Physics-Based Digital Twin is one of two core sub-components within the Digital Twin Framework. The main functionality is a comprehensive dynamic simulation software for Building Design and Energy Performance analysis.
Developers
Integrated Environmental Solutions (IES)
www.iesve.com
Features & Benefits
The advanced building performance simulation software enables the creation of physics-based digital twins for energy and sustainability analysis. Its features allow accurate modeling of building behavior and evaluation of renovation scenarios.
