Energy efficiency of Heating, Ventilation and Air Conditioning systems in production environments through model-predictive control schemes: The case of battery production

Abstract

In the German industry Heating, Ventilation and Air Conditioning (HVAC) systems account for 11 %–20 % of the final energy consumption. Usually HVAC systems are dimensioned based on static extreme values and thus during most of the operation time, HVAC systems tend to be overdimensioned. In addition, the extensive literature review of 40 papers shows that often topics on model-based energy efficiency improvement are commonly performed for office or residential buildings, but rather rarely for production environments. Furthermore, the performance of control schemes for energy efficiency improvement are rarely compared with each other or incompletely compared with each other. In this publication, we use the framework of a Cyber–Physical Production System (CPPS) to compare the performance of four main control schemes, ranging from simple time-based control to model-predictive control, which has not been done in the context of battery production before. Over a fixed observation period, the four control schemes are transparently compared with each other in a case study using the example of the Battery LabFactory Braunschweig (BLB). Compared to the initial state, a significant reduction in final energy demand can be achieved with all examined control approaches, which is even up to 37.29% in case of the model-predictive control. Surprisingly, even simple control approaches, without any predictions, have a good energetic saving potential of up to 20.22% compared to the initial state, which is why they are initially recommended for practical implementation. Moreover, we publish all source code of the control schemes, in order to encourage the implementation of energy efficiency measures in further applications not restricted to battery production.