Johnson Controls selected GraphDB to provide semantic data creation and management for their Metasys system – a Top-5 Integrated Building Management System.
“Buildings that almost think for themselves” is how a 1984 NYT article bravely called intelligent buildings as if looking into the future of the building industry and more specifically of the building automation sector.
And they were right. The increasing demand for sustainable designs, efficient management, commercial buildings energy savings and flexibility did direct the building industry towards developing smarter building automation systems (BAS). This in turn led to better occupants and management experiences while saving costs and energy.
Today, intelligent buildings like Oakland City Center collect temperature and humidity data, evaluate it and control the heating, ventilation and air conditioning (HVAC) system for cost-efficient and sustainable energy use. There is also Frasers Tower in Singapore, whose 179 Bluetooth Beacons and 900 lighting, air quality and temperature sensors gather data to enable maximum efficiency of physical space resources and optimal productivity for the occupants. A factory floor in Dresden, for instance, was filled in with sensors for personnel, machine movement and is now helping in optimizing production, keeping up with safety regulations while at the same time ensures continued maintenance of the equipment.
All these soaring numbers of smart buildings and controllers across physical spaces and the vast amounts of data that feed them challenge the building industry. It needs to find sustainable ways to manage physical spaces and the data they consume and produce. And, more importantly, to improve and extend the way data is connected and further managed.
Let’s face it, when you have to automate a building with more than 10 thousand connected devices, which stream and exchange data, you need to find a way to integrate this data smoothly and then manage it efficiently..
Knowledge graphs built with semantic technology have paved the way to seamless integration and efficient future-proof data management.
Sensors, controllers, actuators and other devices all exchange data at unprecedented levels. Left without context, this data cannot become actionable information.
Take for example the most common use of automation functions – lighting control. Reportedly, it is one of the easiest to start with when saving on energy costs. But to take the best advantage of this subsystem, one must optimize the way the data it produces is handled and contextualized.
Case in point, the lighting in an office space can be efficiently automated and scheduled by integrating data coming not only from the light controlling devices but also from photosensors, occupancy sensors, etc. In such a scenario, when the optimal light for each room is automatically turned on, it takes into account all relevant data such as the levels of light coming from outside, the occupancy data, etc. Such optimization is only possible when all this data is properly modeled, described and semantically integrated.
The more formalized, standardized and properly integrated the data is, the more efficient it is to monitor, control and manage the systems that generate it.
Knowledge graph technologies, which have been on the rise and maturing as an enterprise solution for the last few years, allow seamless data integration and easy data management. They also offer the means for meeting the current challenges while tapping into the potential of future opportunities for even smarter building automation systems.
Some companies from the building automation industry already have taken the semantic technology road ahead. Take Johnson Controls for example. In order to enhance their Building Management System, called Metasys, the company added a semantic layer to it.
The modeling and interlinking of the data involved in the process of building a knowledge graph helped Johnson’s Controls ingest heterogeneous data and reap the benefits of a data-centric approach to building automation. Thanks to this newly enhanced system powered by a knowledge graph, their clients could enjoy safer and smarter buildings that are also easier to manage.
In other words, the better the linking of the physical and digital layers of the devices and their data, the smarter the thinking of the building.
The whys behind making a building smarter are quite obvious. The more layers talk to each other, the more space for automation, monitoring and efficiency management. In practical terms, this means richer dashboards, better integrated systems, enriched user interfaces, safer maintenance and optimized energy consumption.
Adding a semantic layer to the data produced by systems such as HVAC, lighting, access control and so on significantly improves energy efficiency, cost optimization, occupants’ experiences, etc. But only if the data integration is done right.
To transition from “almost thinking for themselves” to “really thinking for themselves”, buildings need management and automation solutions that talk to each other in a uniform language. This means using universal names for identifying physical objects in a standardized way. It enables a flexible and dynamic abstract model out of the BAS concepts and their instances, and uses this model to access, manage and control these systems virtually.
The standardized mapping between physical and virtual objects and processes gives context to the data building automation systems produce. And it is exactly what turns this data into actionable information.
Knowledge graphs are a proven solution to the challenge of interlinking and optimally utilizing data coming from disparate sources – think real-time sensor data, occupancy data, manufacturer’s data, etc. Their modeling flexibility, data normalization capacity and interconnectedness make it easy to ingest heterogenous data and describe it with uniform metadata.
In the building industry domain there already are open-source ontologies (such as Brick to name one), which help companies in the Industry sector model their buildings in a knowledge graph.
Putting standardizeа semantic data in context, knowledge graphs enhance building automation processes by enabling:
As the levels of automation in commercial and residential buildings increase, so does the demand for proper integration of the data their “thinking for themselves” entails.
Modeling, normalization and data integration is the way forward to harvesting the data streams building automation systems produce. By using knowledge graphs, these systems are enriched with an abstracted level, a sort of a digital twin, of each and every controller, space and event in a building. This enables one to have multiple views on the data, navigate data sensors and devices, query them effortlessly, perform diagnostics from a single access point, practice preventive maintenance and more.
All in all, reaping the benefits of informed, sustainable strategies for the future of the building automation industry and the smarter day-to-day functioning of our buildings looks easier and more efficient with a knowledge graph.