Extrapolating performance indicators for annual overflow volume reduction of system-wide real time control strategies

ABSTRACT

Model-based evaluation of real time control and model predictive control strategies is often performed by estimating the yearly reduction in combined sewer overflow (CSO) volumes. This requires sufficiently long data series, but the necessary input data are often lacking in the case of complex control strategies. This article compares a new method to extrapolate yearly CSO volume reduction by simulating a limited number of CSO events. The method showed a good accuracy (2–3% deviation) when applied to a synthetic example. When applied to a real case study in Copenhagen (Denmark), the method showed a tendency to overestimate the performance of the control. The results underline how performance of real time control strategies is strongly affected by yearly variation, non-linearity and interactions among the elements of the system. It is thus suggested to use a mix of different performance indicators when evaluating control performance in conditions of data scarcity.

KEYWORDS:

• Model predictive control
• key performance indicators
• combined sewer overflows

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed https://doi.org/10.1080/1573062X.2021.1948078

Additional information

Funding

This work is supported by the European Regional Development Fund through the NOAH Project [Interreg Baltic Sea Region Programme Grant #R093]. Results for the Lynetten case were generated during the industrial postdoctoral project ‘MOPSUS – Model predictive control of urban drainage systems under uncertainty’ (2012–2014), which was financed by the former Danish National Advanced Technology Foundation.