Digital Twin Technology
A real-time, data-driven virtual representation of the METU RealLAB.
What is Digital-Twin Technology?
The Digital Twin of METU RealLAB is a real-time, data-driven virtual representation of the RealLAB laboratory located in the Middle East Technical University, Department of Electrical and Electronics Engineering.
This digital twin has been developed and operated within the scope of the TÜBİTAK 1004 – SÜİT (Sürdürülebilir Kentler için İleri Teknolojiler) Program, where A-403 serves as a reference real-environment testbed for digital twin technologies.
Unlike simulation-based models, the RealLAB Digital Twin continuously mirrors the state of a physically occupied laboratory, enabling realistic monitoring, control, and long-term experimentation under actual operating conditions.
Real Time Monitoring
Environmental data such as energy consumption, air quality, temperature, humidity, lighting of smart buildings can be monitored in real time. In this way, more effective management is provided.
Maintenance
Environmental data such as energy consumption, air quality, temperature, humidity, lighting of smart buildings can be monitored in real time. In this way, more effective management is provided.
Energy Efficiency
Environmental data such as energy consumption, air quality, temperature, humidity, lighting of smart buildings can be monitored in real time. In this way, more effective management is provided.
Simulation and Optimisation
Simulations can be performed to test how the building will operate in different scenarios and improve performance. This allows for better design and operation decisions.
Remote Management
Smart buildings can be easily controlled and managed over the internet thanks to the digital twin. This reduces operational costs, especially for large complexes.
Sustainability
It is beneficial in terms of efficient use of resources, reducing carbon footprint and complying with green building standards.
Role within TÜBİTAK 1004 – SÜİT
Within the SÜİT Program, METU RealLAB plays a critical role as:
- A living laboratory for validating digital twin concepts
- A pilot environment for smart building and energy-aware control scenarios
- A benchmark space for comparing rule-based and learning-based control strategies
- A data source for reproducible, long-term research on sustainable indoor environments
The laboratory enables the transition of digital twin technologies from conceptual designs to TRL 5–6 real-world deployments, supporting the program’s objectives of sustainability, scalability, and real impact.
What the Digital Twin Represents
The digital twin integrates heterogeneous data streams to represent both environmental and operational states of the laboratory, including:
- Indoor temperature and relative humidity
- Indoor CO₂ concentration (air quality)
- Occupancy and human presence indicators
- Electrical energy consumption (multi-phase monitoring)
- HVAC operating modes and temperature setpoints
- Ventilation and window-fan actuation states
All measurements are time-stamped, archived, and synchronized with the virtual model, enabling both real-time visualization and retrospective analysis.
Technology Stack and Platform
The Digital Twin of METU RealLAB is powered by JeoIT and built on the Evolved.City IoT and digital twin platform.
Key characteristics of the platform include:
- Secure IoT data ingestion from edge devices
- Real-time digital twin state synchronization
- Time-series data storage and querying
- API-based integration with external systems
- Web-based visualization and analytics
This infrastructure ensures that the digital twin remains modular, extensible, and interoperable across different projects within the SÜİT ecosystem.
System Architecture
The RealLAB Digital Twin follows a distributed edge–cloud architecture:
- Edge Layer: Raspberry Pi–based edge nodes collect sensor data and execute local control logic.
- Communication Layer: MQTT and ZeroMQ protocols provide reliable, low-latency data and control exchange.
- Digital Twin & Data Layer: The Evolved.City platform maintains synchronized digital representations and historical records.
- Visualization & Interaction Layer: A web-based dashboard enables monitoring, analysis, and human-in-the-loop supervision.
This architecture supports continuous operation and long-running experiments without disrupting daily laboratory use.
Intelligent Control and Experimentation
The Digital Twin supports multiple control paradigms evaluated under identical real-world conditions:
- Rule-based control strategies
- Edge-based reinforcement learning controllers
- Cloud-assisted learning and optimization
- Manual and supervisory override modes
This capability makes METU RealLAB a controlled yet realistic benchmark environment for studying comfort–energy–air quality trade-offs.
Research and Educational Impact
The Digital Twin of METU RealLAB contributes to:
- Research on smart buildings and sustainable campuses
- Development of digital twin methodologies under real constraints
- Validation of edge AI and learning-based control
- Graduate and undergraduate education using real sensor data
Students and researchers interact directly with a functioning cyber–physical system, strengthening hands-on learning and reproducible research practices.
REAL Lab
Bridging the gap between simulation and reality using Reinforcement Learning.
Quick Links
Contact Us
A-403, Dept of EEE, METU
Ankara, Turkiye