Turning building data into intelligent energy control
Source: https://www.tue.nl/en/news-and-events/news-overview/06-03-2026-turning-building-data-into-intelligent-energy-control Parent: https://www.tue.nl/en/research
Share
Turning building data into intelligent energy control
March 6, 2026
Lasitha Rathnayaka Mudiyanselage defended her PhD thesis at the Department of Built Environment on March 5.
Photo by Rukshan Wijekoon
Buildings play a central role in today’s energy and climate challenges. They account for a large share of global energy consumption, electricity use, and greenhouse gas emissions, and their demand continues to grow as heating, transport, and other services become increasingly electrified. How buildings use energy therefore has a direct impact on electricity grid stability, energy costs, and the speed of the transition to a low‑carbon society.
Lasitha Rathnayaka Mudiyanselage’s PhD thesis addresses a key barrier that limits our ability to manage building energy intelligently at scale: the lack of interoperable, reusable software systems for smart energy control.
Data fragmentation as a major obstacle
A major reason for this challenge is data fragmentation. Buildings contain multiple independent information systems—such as building management systems, Internet of Things (IoT) sensors, and digital building models (BIM). These systems often use different formats, naming conventions, and data structures, many of which are tailored to specific vendors or applications.
As a result, smart energy applications typically have to be custom engineered for each building, making them difficult to scale and slowing innovation.
Rathnayaka Mudiyanselage’s research investigates how semantic technologies can help overcome these challenges. Ontologies provide a shared, machine‑readable way to describe building components, data points, and their relationships. By giving data explicit meaning rather than just labels, ontologies enable software applications to automatically discover, interpret, and reuse building data across different systems and buildings.
A semantic architecture for smarter buildings
The thesis proposes a reference architecture for smart buildings that combines service‑oriented software design with ontology‑based data integration. In this architecture, heterogeneous data sources—such as building management systems, IoT platforms, and BIM models—are connected through a common semantic layer.
According to Rathnayaka Mudiyanselage, this semantic layer forms a critical bridge between physical building systems and advanced control applications, allowing data to be accessed and combined in a scalable, consistent way.
Making Model Predictive Control portable
Building on this foundation, the thesis focuses on Model Predictive Control, an optimization technique that plans building energy use ahead of time based on forecasts of demand, weather, and system constraints. While Model Predictive Control has demonstrated strong potential for improving energy efficiency and flexibility, it is usually highly customized and difficult to reuse.
This work introduces a modular design approach in which control algorithms are decomposed into reusable services semantically standardized data requirements. A semantic portability service is developed to validate and configure these control applications for different buildings.
Real‑world demonstrations
The proposed methods are demonstrated through real-world case studies, including the control of electric vehicle charging in a solar-powered office building.
Rathnayaka Mudiyanselage’s results show that semantic data models and modular software design has the potential to reduce the manual effort required to deploy advanced energy management applications, while maintaining performance and reliability.
Supporting a more sustainable built environment
The societal impact of this work lies in enabling scalable and intelligent energy management in buildings. By lowering the barriers to deploying advanced control strategies, Rathnayaka Mudiyanselage’s research contributes to:
- more efficient building energy use
- better integration of renewable energy
- improved grid flexibility
In the long term, this helps reduce carbon emissions, lowers energy costs, and accelerates the digital transformation of the built environment.
This thesis was carried out under the project Brains4Buildings' Energy Systems – Work Package 4: Data Integration for smart communication.
Title of PhD thesis: Semantics for Smart Building Systems: Towards Portable Demand Side Management Applications in Buildings Supervisors: Pieter Pauwels and Ekaterina Petrova.
Media contact
Joana Borges
(Communication Advisor)