7 Cannabis Facility Maintenance Best Practices
Sustaining a cannabis cultivation operation means creating the right conditions for the plants to...
The fault detection and diagnostics (FDD) features in building analytics software are powerful tools to eliminate energy waste and reduce costs. A 2020 report by the U.S. Department of Energy’s Smart Energy Analytics Campaign found that FDD produces a median energy savings of 9%. In commercial properties and public buildings, this adds up to significant savings.
A majority of FDD available in the market today is rule-based and straightforward to implement. Rule-based FDD specifies expected conditions for building systems with clear limits or ranges of performance. However, rule-based FDD has several disadvantages:
Fault detection using machine learning (ML) algorithms overcomes these challenges through analysis of real-time and historical building data. ML-driven analytics proactively identifies the issues with your building system and allows you to take action before occupants complain, equipment fails, or energy is wasted. Building owners who invest in fault detection using machine learning see lower energy bills while prolonging the life of the equipment and improving comfort for occupants.
FDD should identify problems, notify operators, and recommend solutions when equipment is operating outside of expected performance criteria. Smart building software with machine learning does this in the best way possible.
Limiting False Alarms
The limited analytic capabilities of a building management system (BMS) often fail to identify the malfunctioning of a single component in a complex building network and instead trigger multiple alarms. Unprioritized and false alarms delay identification of the actual point of failure. Machine learning prevents this by aggregating real-time and historical data to provide context to alarms and suggest effective solutions.
A smart building platform using machine learning for fault detection defines rules and automated response actions to limit nuisance alarms. Building owners minimize reliance on contractors to identify the root cause of alarms, and maintenance visits are more focused and cost-effective. Limiting false alarms also prevents premature upgrades or replacement of related equipment, eliminating unnecessary labor and material expenses.
Effective preventive maintenance of sophisticated building systems is a major challenge for building owners without smart analytics. Conventional rule-based strategies in fault detection models do not provide enough opportunities for preventive maintenance, resulting in increased downtime and energy wastage. Fault detection using machine learning, however, flags any deviations from historic trends and predicts equipment failures long before they occur. This means maintenance is guided by objective data, not arbitrary schedules, and optimal performance is easier to maintain.
In conventional systems, there isn’t a lot of validation of preventive maintenance. During a scheduled maintenance visit, the crew quickly inspects the equipment and moves on if there are no obvious problems. Not only does this waste resources on unnecessary visits, it’s also a highly unreliable way to detect and prevent malfunctions. With ML-based maintenance, crews are informed of specific issues and equipment that need to be checked proactively.
Improving Energy Efficiency
HVAC, lighting, refrigeration, and hot water heating are key sources of energy consumption in commercial and public buildings and directly impact the cost of operations. Fault detection using machine learning identifies the variables in a building’s energy consumption patterns, detects anomalies, and provides detailed insights about suboptimal equipment performance. As a result, your ability to optimize efficiency and reduce energy consumption is dramatically expanded.
Such capabilities mean machine learning should lie at the core of your BMS and allow you to integrate mechanical maintenance and building automation. All you need is the right smart building platform.
onPoint is a cloud-based smart building platform with state-of-the-art FDD capabilities. This innovative software has dedicated strategies to limit false positives and false alarms, including analyzing minimum duration and threshold. But it also goes beyond that.
onPoint analyzes data points and filters out small deviations to ensure data integrity. It also applies ML-based analytics to various data points and creates a model of behavior for every piece of equipment installed in a building. This provides powerful predictive capabilities and allows you to isolate the cause of problems even in complex systems of equipment, sensors, and devices.
Fault detection using machine learning simplifies building maintenance, makes maintenance visits more valuable, and reduces immediate and long-term costs. The result is a more efficient operation and a better building.
Laura draws on her experience in commercial real estate to cover trends in occupancy, indoor air quality and operational efficiency.