From Design to Operation: Mixed-Integer Model Predictive Control Applied to a Pumping System

The two most significant life cycle phases of products or systems are the design and operation phase. Both share their incredibly high level of complexity due to the available diversity in components, operating settings and interconnection variants. In the design process, two-stage stochastic optimisation problems have proven to be suitable, in which the operation is anticipated by considering various demand scenarios. Since the operation is characterised by uncertainty and fluctuation, it has to be ensured that the operation is also realised in an optimal way. In this contribution we show how the original planning problem is transformed into a control problem using the example of a pumping system.