Reliable, robust and resilient system design framework with application to wastewater treatment plant control
Journal of Environmental Engineering
American Society of Civil Engineers
This is the author accepted manuscript. The final version is available from American Society of Civil Engineers via the DOI in this record.
This paper presents a framework for reliable, robust and resilient system design, addressing the need for acceptable performance not only to be provided under expected conditions, but to be maintained and/or quickly returned to when threats act upon a system. This is achieved through incorporation of multi-objective optimisation, assessment of reliability, robustness and resilience, and filtering and prioritisation of solutions at key intervals. The effectiveness of the framework and benefits of its use are demonstrated with a wastewater treatment plant (WWTP) control strategy design problem. The effects of extreme, non-design loads on highly optimised control strategies have not previously been explored and it is found in this example that options yielding the greatest improvements under design conditions typically provide poor effluent quality robustness and resilience to influent perturbations. By integrating robustness and resilience, solutions can be identified which meet key performance objectives under design conditions whilst also minimising the magnitude and duration of potential failures under extreme conditions. Application of this framework can also yield knowledge of threats to which the system is least robust and/or the requirements most prone to failure under extreme conditions. It is shown here that, when attempting to reduce greenhouse gas emissions by improved WWTP control, decreased influent temperature poses a significant threat and maintaining an acceptable total nitrogen concentration in the effluent under extreme conditions is a key issue.
Thanks are given for the Matlab/Simulink implementation of the BSM2 from the Department of Industrial Electrical Engineering and Automation, Lund University, Lund, Sweden. This work forms part of a 5-year fellowship for the last author funded by the UK Engineering & Physical Sciences Research Council (EP/K006924/1)
Published online: 12 October 2016