29 January 2024

Annex 56 Digitalization and IoT for Heat Pumps Task 3: Data analysis


Today, more and more devices are connected to the Internet and can interact due to increasing
digitalization – the Internet of Things (IoT). In the energy transition, digital technologies are intended to
enable flexible energy generation and consumption in various sectors, thus leading to greater use of
renewable energies. This also applies to heat pumps and their components.
The IoT Annex explores the opportunities and challenges of connected heat pumps in household
applications and industrial environment. There are a variety of new use cases and services for IoT
enabled heat pumps. Data can be used for preventive analytics, such as what-if analysis for operation
decisions, predictive maintenance, fine-tuning of the operation parameters and benchmarking.
Connected heat pumps allow for demand response to reduce peak load and to optimize electricity
consumption, e.g. as a function of the electricity price. Digitalization in industry can range from
automated equipment, advanced process control systems to connected supply value chains. IoT
enabled heat pumps allow for integration in the process control system and into a high level energy management system, which can be used for overall optimization of the process. IoT is also associated to different important risks and requirements to connectivity, data analysis, privacy and security for a variety of stakeholders. Therefore, this Annex has a broad scope looking at different aspects of digitalization and creates a knowledge base on connected heat pumps. The Annex aims to provide information for heat pump manufacturers, component manufacturers, system integrators and other actors involved in IoT. The Annex is structured in 5 tasks:
Task 3 – Data analysis
This task gives an overview on data analysis based on examples of IoT products and services,
Different targets for data analysis are derived, data analysis methods are categorized and assessed,
starting with visualization and manual analysis reaching to machine-learning algorithms. The report provides insights in the pretreatment of data, the use of data models, meta data and BIM (building information modeling).