A framework for the mixed integer dynamic optimisation of waste water treatment plants using scenario-dependent optimal control

In many real life processes, operational objectives, constraints and the process itself may change over time. This is due to changing product requirements, market demands and other external or internal influences, which constitute a certain scenario. Model-based techniques can provide optimal solutions to the corresponding scheduling and process control problems. This paper focuses on those situations, where the objectives and constraints of plant operation depend on the scenario and therefore change over time. A framework is developed, which enables the modelling and scheduling of different operational strategies on the optimisation horizon. A case study involving a waste water treatment plant is used to demonstrate the approach. Existing expert knowledge is used to relate certain operational objectives and constraints to corresponding scenarios. It is shown that easily interpretable optimisation results are obtained. Also, the results are significantly improved as compared to a weighted average approach only approximating sequential strategies.