ZandrEA Software in Research to Automate Fault Detection and Diagnostics of Mechanical Systems in Large Commercial Buildings -- A Primer

Published

April 10, 2025

Author(s)

Daniel Veronica

Abstract

Mechanical systems in large commercial buildings, particularly their heating, ventilating, and air-conditioning (HVAC) systems, routinely waste significant amounts of energy and thus levy significant unnecessary economic costs upon building owners. This happens due to "faults"—improper design or installation, improper control actions, or operational equipment malfunctions—occurring within the HVAC systems. Saying a commercial building is "large" refers not only to its size, but that it has an automatic control system continually generating large amounts of data from operation of its HVAC system. Those data are resources that software exercising well-developed techniques of artificial intelligence can leverage for automated fault detection and diagnosis (AFDD) done in real time. Ongoing research and development is necessary to ensure novel, "cutting-edge" ideas that can improve AFDD actually become, as just stated, "well-developed techniques" in their effectiveness. That requires software having distinctly research-oriented capabilities that go beyond what is available from AFDD software commercially marketed to building owners. This document first introduces, and then broadly details, new software having the research-oriented capabilities the AFDD industry needs. The software is offered in the public domain as the foundation for an ongoing, collaborative, open-source software development project named ZandrEA. The ZandrEA project welcomes collaboration from stakeholders throughout the academic and industrial communities interested in advancing the technology of AFDD, and in supporting the transfer of that technology into marketable products and services.

Citation

Technical Note (NIST TN) - 2337

Report Number

2337

NIST Pub Series

Technical Note (NIST TN)

Pub Type

NIST Pubs

Download Paper

https://doi.org/10.6028/NIST.TN.2337

Local Download

Keywords

afdd, automated fault detection diagnostics, Bayes theorem applied, Bayesian inference, expert system, fdd, hvac, intelligent buildings, open-source software, naive Bayes classifier.

Citation

Veronica, D. (2025), ZandrEA Software in Research to Automate Fault Detection and Diagnostics of Mechanical Systems in Large Commercial Buildings -- A Primer, Technical Note (NIST TN), National Institute of Standards and Technology, Gaithersburg, MD, [online], https://doi.org/10.6028/NIST.TN.2337, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=958096 (Accessed April 30, 2025)