Modular data model for energy-flexible cyber-physical production systems

Abstract

Due to the massive expansion of renewable energies and the growing importance of carbon pricing, energy prices are rising, and industrial companies must put higher effort in purchasing electrical energy cost-efficiently. Additionally, the increasing share of renewable energies leads to more volatility in the power market, giving companies the opportunity to reduce their energy costs by using energy-flexibility measures. As thermal supply systems offer a great potential for energy flexibility, building automation systems which control the supply systems must be adapted for communication and external access. Intelligent algorithms interacting with the automation can enable factories to fexibilize their energy purchases and to participate in energy markets by using external information, such as energy prices or forecasts of the machine’s energy demand. This requires seamless communication that connects all hierarchy levels of the Reference Architectural Model Industry 4.0 (RAMI 4.0) from connected world to field level. Also functional safety must be ensured. Moreover, energy-flexibility measures must be validated within simulation environments which should show the same data structure. Therefore, a standardized data model that can be applied across different companies is necessary. We propose a concept for a modular and consistent data model for automation, simulation, and communication to enable simple data exchange using external algorithms. Therefore, base classes containing variables for generic actors and sensors are developed and extended by further variables to model systems based on their control mode. Moreover, a safety module to manage external access by algorithms is included. With this data model, industrial energy supply systems can be easily expanded to include additional components. Furthermore, energy supply systems consisting of several components, such as energy converters, pumps and valves can be modeled in an simple and time-efficient way. The data model is applied to a compression chiller at ETA Research Factory.