Digital twin with augmented state extended Kalman filters for forecasting electric power consumption of industrial production systems

• DOI: 10.1016/j.heliyon.2024.e27343
• Publication Date: 2024-03-07T18:18:10Z
• Authors: A. Baldassarre; J.-L. Dion; N. Peyret; F. Renaud
• Keywords: Cyber-physical system

The work aims to develop an effective tool based on Digital Twins (DTs) for forecasting electric power consumption of industrial production systems. DTs integrate dynamic models combined with Augmented State Extended Kalman Filters (ASEKFs) in a learning process. connection the real counterpart is realized exclusively through non-intrusive sensors. This architecture enables model development systems (components, machinery and processes) which complete knowledge not available, by identifying model's unknown parameters short online training phases small amounts real-time raw data. ASEKFs track unknowns keeping updated as physical evolve. When forecast needed, current estimates uncertain are integrated into models. These can then be used without predict actual energy use system under desired operating conditions, including scenarios that differ from typical functioning. approach validated offline reference electricity automatic coffee machine, represents test environment blueprint design other appliance observed measuring supply voltage absorbed current. accuracy results analyzed discussed. method developed context prediction optimization manufacturing industry refined management planning.