|Title||Characterization and Survey of Automated Fault Detection and Diagnostics Tools|
|Year of Publication||2017|
|Authors||Jessica Granderson, Rupam Singla, Ebony Mayhorn, Paul Ehrlich, Draguna Vrabie, Stephen Frank|
|Institution||Lawrence Berkeley National Laboratory|
It is estimated that 5%–30% of the energy used in commercial buildings is wasted due to faults and errors in the operation of the control system. Tools that are able to automatically identify and isolate these faults offer the potential to greatly improve performance, and to do so cost effectively. This document characterizes the diverse landscape of these automated fault detection and diagnostic (AFDD) technologies, according to a common framework that captures key distinguishing features and core elements.
To understand the diversity of technologies that provide AFDD, a framework was developed to capture key elements to distinguish the functionality and potential application of one offering from another. The AFDD characterization framework was applied to 14 currently available technologies, comprising a sample of market offerings. These 14 technologies largely represent solutions that integrate with building automation systems, that use temporary in field measurements, or that are implemented as retrofit add-ons to existing equipment. To characterize them, publicly available information was gathered from product brochures and websites, and from technical papers. Additional information was acquired through interviews and surveys with the developers of each AFDD tool. The study concludes with a discussion of technology gaps, needs for the commercial sector, and promising areas for future development.
|LBNL Report Number|| |