When it comes to building management, there are many different ways that facilities across the country are moving toward a future of smarter technology and more connected digital systems. As building management software becomes more sophisticated, facility managers can more easily view the operational efficiency of their energy network.
Many of these solutions focus on the use of real-time data and alerts to identify problems that can make it harder to maintain occupant comfort, reduce energy costs and improve maintenance outcomes. As these practices have evolved, the term fault detection and diagnostics (FDD) has been increasingly applied to this process of identifying, isolating, and diagnosing building performance issues.
Since the early days of computerized building automation systems, operators have used the data and alerts to identify issues impacting facility performance. These initial versions of fault detection and diagnostics have been used for decades within the commercial HVAC space.
However, two recent trends have brought FDD back to the industry’s forefront: increasing awareness of the value that operational data can provide to an organization, and the growth of advanced data analytics methods such as machine learning. Thanks to these developments we have seen a new term coined: automated fault detection and diagnostics (AFDD). With an AFDD system, data is constantly updated showing locations within the system where equipment might not be functioning at an optimal level.
Looking at facility data can show how specific pieces of equipment are operating at any moment in time. The data sources can vary based on the type of facility. In a typical office building, sensors and control points in the building automation system provide a good snapshot of the performance of the HVAC systems.
The most common use for this data is to detect and diagnose “faults” showing precisely when a piece of equipment is not operating within expected parameters.
A case in point: a state energy office in California employed FDD technology at its headquarters in 2018, discovering flaws in the design of the heating and cooling system that was creating both energy inefficiencies and space comfort issues. With a problem diagnosis provided by the FDD tool, the design flaw was fixed, improving energy efficiency by 13 percent and reducing occupant comfort complaints by 30 percent, all at relatively low cost.
Unlike the building automation systems themselves, FDD tools are not intended to alert building managers to failure conditions or deviations in critical spaces. Instead, FDD is intended to identify a condition that deviates from the intended level of performance. FDD tools can also trigger faults when energy consumption is higher than expected based on the facility’s operating parameters, time of day, and weather conditions (a practice called “energy anomaly detection” or EAD). Once a potential issue is identified, notices are uploaded onto system monitors for managers to prioritize issues based on their operational objectives and potential cost impacts.
Modern FDD tools rely on a data connection to a building automation system, so the time to implement FDD will vary based on the condition and level of complexity of the subject facility. Here’s how the process usually goes:
The most important step in the process is the “training” of the tool: continuous improvement of the building model and rules to ensure the system is generating accurate and actionable insights. To learn more about successfully implementing smart building technology, watch the video or download the report below.