Ontologies for Formalizing the Process of Configuring and Deploying Building Management Systems

For tackling the current energy crisis and increasing energy demand worldwide, it has become urgent to rationalize energy use and decrease energy waste. Buildings are responsible for about one third of the global energy consumption in the world, and 30% of this energy is even wasted due to e.g. malfunctions of heating, ventilation and air conditioning systems (HVAC), wrong operation strategies or unaware user behavior. Building management systems (BMS) can bring support through optimal control and continuous monitoring of building energy systems, but their design and configuration is a complex process that requires much expert knowledge and labor cost. This paper presents a method relying on knowledge graphs to automate this process. Starting from a semantic description of a building, it characterizes its energy system to identify applicable monitoring and control tasks. Following a risk paradigm, it then configures and deploys these tasks as BMS functions selecting for each their required sensor data. The industrial adoption of the method is foreseen as an expert system addon to BMS that would accelerate and scale up their deployments for energy saving.