Data driven lab for building energy systems
A database, which contains operation data from more than 4000 heat pump installations throughout Sweden, can be potentially exploited by end user applications to allow manufacturers, utilities, customers or third parties to perform data monitoring and analysis. However the database suffers from incompleteness, inconsistency, lack of accuracy or sensor calibration issues. To appropriately utilize the database, we will integrate other sources such as models and lab measurements to turn the low quality data into useful information. We will develop a data-driven lab which will act as a virtual platform to improve the control strategies, fault detection and performance degradation.
Funded by: AIT(Austria Institute of Technology), CSC(China Scholarship Council).
Background
Every building heating system potentially generates a large amount of data every day that can be stored in databases and exploited by end user applications to allow manufacturers, utilities, customers or third parties to perform data monitoring and analysis.
At the current situation, this highly valuable and abundant data with billions of entries every month is merely collected by heat pump manufacturers and is neither processed nor used for any innovative service. The use of these data is nowadays mostly limited to the instantaneous visualization of the measurements and only a shallow and limited analysis is performed to help the development of enhanced devices or improved control strategies.
The collected data, if processed and coupled to other sources of information in an appropriate way, encompasses a lot of valuable information about local outdoor and indoor climate, occupancy and activities in the built-environment, people behavior, building characteristics, components and system efficiencies, and the early symptoms of the possible fault or performance degradation in the built-environments.
Within this project, KTH and AIT will develop an intelligent data driven virtual lab which acts as the access point for all the collected data and provides innovative services to OEMs, entrepreneurs, researchers, relevant third parties such as insurance companies and last but not least to end-users. The measured data, combined with models and lab measurements, will be used as a source to improve the control algorithms and support the operation and maintenance of the system.
Aim and objectives
The project ultimately aims at reducing the energy use, operating cost, and CO2 emission from the heating systems in the built-environment. The aims will be achieved through design and development of a data-driven lab which will act as a virtual platform to improve the control strategies, fault detection and performance degradation in the heating systems.
