Leveraging AI for HVAC Energy Efficiency: AHU Optimization through Continuous Coil Monitoring

Until recently, established methods for determining coil fouling were based on either differential pressure testing or visual inspection (NADCA), and neither method accurately quantifies the actual coil fouling level. Additionally, these inspection intervals are not standardized and vary based on geographic location, human, and mechanical conditions. This often leads to maintenance uncertainties among the facility operators causing deviations from optimum cleaning frequency. These uncertainties penalize the facility by either increasing the energy costs due to operating an inefficient system or increasing the maintenance costs associated with cleaning coils too frequently.

The presentation will introduce a new coil monitoring technology that combines modern AI machine learning algorithms with traditional HVAC thermodynamic models to determine the true “Coil Fouling Level”. Through an integrated data acquisition system, this technology measures coil fouling level in real-time with great precision. This cloud-based technology also delivers key information about the facility including energy consumption and operational coil capacity. All this information is used to determine a “Coil Service Interval” which indicates the ideal timeframe for subsequent coil service.