An integrated performance analysis framework for HVAC systems using heterogeneous data models and building automation systems

More than 20% of the energy consumed by heating, ventilation and air-conditioning (HVAC) systems is wasted due to undetected faults in these systems. In the past three decades, researchers have developed hundreds of computer algorithms to automatically and continuously analyze their energy performance. However, due to the complex information required by these algorithms, it is very difficult for facilities operators to deploy them in real-world buildings. This paper presents an integrated performance analysis framework (IPAF) that can be used to integrate heterogeneous data models of the building and HVAC systems and the dynamic data from embedded sensors and controllers. This framework facilitates the deployment of performance analysis algorithms in different buildings and HVAC systems by automatically providing the required information. We developed and tested our proposed framework using four different types of algorithms in a real-world facility. The IPAF is able to integrate three heterogeneous data models with 85% of precision and 91% recall. The precision and recall for retrieving data required by the four different types of algorithms are both 100%.