predictive maintenance in HVAC:
the new normal?
HVAC systems account for approximately 40% of total energy usage in buildings worldwide, and interlinked HVAC units in built environments require a well-orchestrated maintenance strategy for efficient energy conservation efforts. The traditional approaches of maintenance — reactive, scheduled, and preventive — have limitations in accurately predicting issues arising from complex modern HVAC systems.
Predictive maintenance, using machine learning-led analytics, can predict equipment failure risks before the issue occurs, enabling precise tracking of HVAC equipment runtime, timely pre-failure interventions to ensure reliability, and reduced downtime.
The article outlines seven benefits of predictive HVAC technologies, including data-driven decision-making, reduced downtime, improved equipment performance, upgraded inventory management, energy optimisation, increased cost-effectiveness, and enhanced safety. We also discuss three predictive maintenance techniques for HVAC systems, which are vibration analysis, thermal imaging, and oil analysis, and how these techniques can work together to give a comprehensive picture of HVAC equipment health.
Heating, Ventilation, and Air Conditioning (HVAC) systems are the heart and soul of every building. Responsible for maintaining good indoor air quality (IAQ), HVAC systems, particularly air conditioners and electric fans, account for around 40% of the total energy usage in buildings worldwide. Additionally, in today’s Industry 4.0 world, HVAC units in built environments are increasingly interconnected, and a breakdown in one system affects the entire interlinked device ecosystem. An efficient and well-orchestrated maintenance strategy can result in significant benefits for HVAC systems, building management systems, and subsequently, energy conservation efforts as well.
the early days of HVAC maintenance
Traditionally, HVAC maintenance strategies have relied on reactionary forms of control, centered around three approaches:
While each method has its benefits, none can accurately forecast issues arising out of the complex machinery in modern HVAC systems. This is where the predictive maintenance strategy excels.
bringing intelligence to HVAC controls
Predictive maintenance uses device data and machine learning-led analytics to predict when a piece of equipment is at risk of failure long before the issue occurs. This enables maintenance tasks to be scheduled appropriately, allowing for precise tracking of HVAC equipment runtime. As a result, timely pre-failure interventions can be taken to ensure reliability and reduce downtime.
benefits of HVAC predictive technologies
enables data-driven decision-making
A wealth of historical and real-time data from sources like IoT sensors and data analysis software, for each HVAC unit, are collated and analysed enabling data-driven decision making. With timely predictions on machine health, owners and managers can take meaningful actions to optimise HVAC performance.
The data-driven calculations, based on actual equipment performance numbers, allow for maintenance to occur on an as-needed basis, reducing downtime for HVAC units. This is especially important for systems in critical facilities like hospitals, and data centers, where avoiding unnecessary offline times is paramount.
improves equipment performance
HVAC systems are complex networks of interconnected components, where a breakdown in one section can impact the entire system’s performance. E.g., a dirty air filter forces the entire HVAC unit to work 10 times harder to compensate for the diminished airflow, increasing the risk of a system-wide breakdown. With a system’s machine learning algorithms for predictive maintenance, problems can be detected early on, preventing future performance issues, and extending the entire HVAC system’s lifespan.
upgrades inventory management
The accurate tracking of equipment conditions allows managers and operators to request replacement parts only as needed, resulting in a better level of inventory management.
boosts energy optimisation
HVAC performance deficits can trigger serious energy wastage, which a cutting-edge predictive maintenance strategy can circumvent. Data collected is analysed for energy-related operational issues, and stakeholders are notified instantly when problems are identified. As a result, optimal operational performance is restored faster and more easily, leading to a higher degree of energy conservation.
increases cost effectiveness
Predictive analytics pinpoint faults before they occur, making repairs faster and more accurate. There is reduced reliance on unnecessary inspections and monitoring, fundamentally changing how the systems are maintained and drastically reducing overall operational costs.
Safety is paramount in any building, and environmental conditions such as CO2 levels, temperature, and humidity play a crucial role in maintaining occupant comfort. HVAC systems ensure flawless control of these functions, with data-driven predictive maintenance addressing any aberrations long before occupants have a chance to detect any problems.
predictive maintenance techniques for HVAC systems
Vibration analysis is a sensor-led method of monitoring the vibration levels of equipment and identifying potential problems. All rotating machines, including motors, pumps, and fans, produce vibrations; recording and studying the amplitude of vibration at normal frequencies help identify changes in the vibrations. This in turn helps detect issues such as misalignment, imbalance, bearing wear, and other mechanical problems. Vibration analysis is mostly used on steam and gas turbines, motors, paper machines, pumps, rolling mills, compressors, machine tools, and gearboxes.
Thermal imaging, akin to medical imaging, can serve as an effective diagnostic tool for detecting issues in HVAC systems. Infrared cameras are placed on HVAC equipment such as air handlers, chillers, and boilers to measure the infrared energy emitted, allowing the camera to show distinctions between hot and cold areas. Issues like air leaks, insulation problems, and equipment failure can thus be solved in record time.
Oil analysis, mainly used in the oil and gas industry, is a method of monitoring the condition of lubricating oil in equipment and identifying potential problems. This is performed by taking samples of oil from the equipment and then examining them in a laboratory. By analysing the oil samples, it is possible to detect issues such as wear and tear on equipment, contamination, and other problems that can lead to equipment failure.
All three techniques mentioned here are extremely useful in identifying maintenance issues in HVAC systems before they become major problems. However, using these techniques together makes it possible to get a comprehensive picture of HVAC equipment health, which can inform better decisions about the maintenance of HVAC systems, and subsequently, building management systems as well.
The advances in the Internet of Things (IoT) technology and artificial intelligence have made predictive maintenance even more efficient and effective. According to researchers, predictive maintenance has reduced maintenance costs by 35%, boosted the overall output by the same percentage, and decreased the time taken for breakdowns by 45%. As the data from predictive maintenance pours in, contractors and building operators are learning to interpret and leverage this information to help people remain comfortable and safe in all HVAC-controlled spaces.
See how Softdel’s award-winning IoT gateway platform EdificeEdge can deliver predictive and preventive analytics solutions at the edge for all your building needs.