Conserve It’s Head of R&D, Michael Berger, presented ‘Real-time Model Predictive Control with Digital Twins and Edge Computing Technologies’ at CLIMA 2022 on the 24th of May 2022.
The presentation covered key Machine Learning and Predictive Optimization features in PlantPRO, such as SmartSequencing and SmartFlow (CW). Michael presented in detail the underlying concepts behind these flagship features that were researched and developed by Conserve It and have been saving energy at numerous sites around the world. The solution automatically learns the performance of the chillers and pumps to create a Digital Twin of the plant. Subsequently, it utilizes the Digital Twin to find in real-time the optimal number of chillers, cooling load distribution amongst chillers and condenser water flow setpoints that minimize holistically the power usage of the plant.
The presentation highlighted real-world results measured on site. Energy savings of up to 7.1% were measured, without replacing any mechanical equipment and only due to applying the optimal control strategies. Real-time operational data show how these features, that are available in PlantPRO right now, are applied in practice and maintain space comfort and reliability in the plant.
PlantPRO operates on site right on the Edge, without needing a remote connection or cloud platform, resulting in cost efficiency and improved cybersecurity profiles and reliability. The presentation highlighted how Conserve It achieved these advanced Machine Learning and Model Predictive Control features on the Edge on a low-cost embedded controls platform.
The concepts and findings are further detailed in a paper published as part of the CLIMA 2022 Proceedings, open-sourced and referenced in the directory of Openbook (www.doabooks.org) and Google Scholar.
Download the 'Real-time Model Predictive Control with Digital Twins and Edge Computing Technologies' paper here.
Download the presentation slides here: