Cognitive Buildings @ IBM Research - Ireland       

links

Cognitive Buildings @ IBM Research - Ireland - Projects


Joern Ploennigs is the principal investigator for two EU-funded H2020 projects focused on cognitive building applications. The first of these two projects is called Tribute. Joern and his Cognitive IoT team is studying how occupants use their buildings to identify ways to reduce energy consumption. The project collects and mines a building’s energy data  and compares the performance against its intended efficiency from building energy performance simulations. This toolkit will evolve the building model from “as built” to “as is” and then to “as it will be” to predict the best way to minimize a building’s energy impact.

The second H2020 project is TOPAs (Tools for Continuous Building Performance Auditing). After the three-year study, TOPAs will provide decision support tools for building and facilities managers to more effectively manage their site, such as providing visibility on how decisions impact energy related occupant comfort and health. Joern’s team is particularly interested in the energy flow within a campus, and how machine learning can correctly predict energy consumption and detect and diagnose energy waste.

The IBM Research - Ireland lab facility is a use case or a “living lab” for both projects providing real-time data from thousands of sensors on heating, cooling, lighting, water, electricity, and presence. This data is used to create comprehensive energy usage models and real-time reports, providing deeper insight into how the space is being used, how the infrastructure is performing, and where resource usage can be optimized.

Recently, the team have begun scaling the "living lab" to a new level of cognitive operations for large groups of buildings or campuses. Joern has created a demo which builds on IBM’s Watson IoT platform to create eight cognitive IoT use cases with several buildings, including a datacenter at the IBM Technology Campus in Dublin, Ireland – a rich test environment with more than 3,000 sensors. These use cases show how cognitive operations solutions will drive predictive maintenance and integrate new immersive virtual and augmented reality interfaces.