Use of Machine Learning for the Smart Design and Real Time Control of Stormwater Storages
Invited plenary talk at NSW Stormwater Conference, franc.sydney 2022.
Urban stormwater systems are stressed and traditional solutions are expensive
Urban stormwater systems are under increasing stress due to densification and climate change, often necessitating expensive pipe upgrades. In addition to the high cost of these upgrades, they are often not feasible due to practical constraints (e.g. proximity of other services, traffic disruption etc.). An alternative is to reduce peak system flows via storage. Traditionally, such storage is placed at catchment outlets, requiring significant volumes and space to be effective.
Smart Design and Real-time Control of Distributed Storages - A solution for future?
The benefits of a given storage volume can be increased significantly by distributing a series of smaller storages throughout catchments at strategic locations, as these not only reduce peak flows locally, but can reduce the coincidence of hydrographs from contributing sub-catchments. Given the large number of potential storage locations and sizes, machine learning methods are used in this research to identify combinations that minimise system peak flows at critical locations under a range of design conditions. By adding controls to the outlets of these storages and operating them in real-time during storm events, also using machine learning methods, system peak flows can be reduced further. In addition, such smart controls provide opportunities for stormwater re-use for urban greening, as well as producing outflows that meet ecological flow requirements.
Case Study Outcomes: Smart stormwater storage reduce peak flows and infrastructure costs
control of distributed stormwater storages is demonstrated for a case study in Unley, South Australia. Results show that smart design of distributed storage can achieve significantly higher peak flow reductions than more commonly used end-of-system storage. These range from 10% for 100 m3 of system storage up to 40% for 700 m3 of system storage – whereas end-of-system storage achieved no peak flow reduction for up to 700 m3 of storage. This reduced infrastructure costs by ~ 30%. The addition of optimised real-time control to distributed storages is able to achieve an additional 10% in peak flow reduction.
Liang R., Thyer M.A., Maier H.R., Dandy G.C and Di Matteo M. (2021) Optimising the design and real-time operation of systems of distributed stormwater storages to reduce urban flooding at the catchment scale , Journal of Hydrology, 602, 126787, https://doi.org/10.1016/j.jhydrol.2021.126787
Thyer M., Maier H., Di Matteo, M.(2019): Evaluating the Benefits of Smart Stormwater Systems: Unley Case Study. The University of Adelaide. Online resource. https://doi.org/10.25909/5de5971f2b4ae
Liang R., Di Matteo M., Maier H.R., Thyer M.A. (2019) Real-time, smart rainwater storage systems: Potential solution to mitigate urban flooding (Open Access), Water, 11, 2428, DOI: 10.3390/w11122428, https://www.mdpi.com/2073-4441/11/12/2428
Di Matteo M., Liang R., Maier H.R., Thyer M.A., Simpson A.R., Dandy G.C. and Ernst B. (2019) Controlling rainwater storages as a system: An opportunity to reduce urban flood peaks for rare, long duration storms , Environmental Modelling and Software, 111, 34-41, https://www.sciencedirect.com/science/article/pii/S1364815218304857