Monitoring And Tuning of Air Conditioning Control Systems
Tech ID: 31945 / UC Case 2020-096-0
Current approaches to thermal control and energy management rely on complex thermal models that are subject to inefficiency and environmental uncertainties. These modelling processes depend on input from multiple environmental sources (i.e. temperature, humidity, HVAC air-flow, etc.) to monitor and tune air conditioning control systems. Unfortunately, this method is subject to increased modelling variance that causes energy management systems to overcompensate when heating/cooling in their effort to maintain desired indoor climatic conditions. By experiencing over-heating and/or over-cooling, current thermal management systems demonstrate increased expense and a lack of efficiency.
Researchers at the University of California, Santa Barbara have implemented an entirely data-driven analysis for monitoring and tuning air conditioning control systems to increase their efficiency and improve thermal management efforts. This analysis uses Koopman Mode Decomposition to analyze data from a single controlled thermal zone, reducing modelling uncertainties by eliminating the need for multiple sources of physical input. Koopman Mode analysis simultaneously monitors real-time and long-term feedback to proactively predict models for optimal thermal efficiency. By reducing variance in energy management and actively analyzing data, Koopman Mode analysis demonstrates an effective approach to reducing costs and increasing the efficiency of thermal control systems.
- Reduced heating and cooling costs
- Increased heating and cooling efficiency
- Simplified thermal data analysis
- Predictive Modelling
- Energy Management Systems