Implementation of IT Infrastructure for Model based Real-time HVAC Diagnostics

Abstract

The objective of this work is to design and test a real-time building energy model framework for monitoring and mea- suring the performance of existing commercial buildings. Existing commercial buildings often perform below their design potential due to lack of appropriate maintenance. This situation can be improved by incorporating a system that continuously monitors and evaluates performance of the building by comparing real-time data with results predicted by an accurate model. This paper describes the framework for a simple system that could be applied to a wide range of buildings with varying levels of controls and measured sensor data. Tests were performed to demonstrate that the availability of predicted model data can lead to the detection of faults with building subsystems that would otherwise go unnoticed. It was also demonstrated that there are several possible options to provide necessary inputs to the model so that even low data availability can produce useful results.