The design of target structures is a topical subject in the water sector. Drinking water companies are busy with the (proactive) replacement of their mains networks. At the same time, they are also doing a lot of work, within the framework of the energy transition, in the subsurface. This presents an ideal opportunity to improve the mains network. It also raises the question about how to optimally achieve a design for a robust and future-proof target structure – i.e., a target structure that performs as well as possible under different conditions, both now and in the future – in order to take maximum advantage of this opportunity.
This project aims to answer the following questions:
- How can target structure designers take into account the uncertainties regarding water demand?
- How can researchers represent performance indicators by means of a probability distribution?
- How can numerical optimisation techniques contribute to the optimal design of a target structure, taking the above two aspects into consideration?
- Water-demand uncertainty and the design of target structures.
The design of target structures is a topical subject in the water sector. Drinking water companies are busy with the (proactive) replacement of their mains networks. The current design process (both in practice and in research) pays little, or insufficiently detailed, attention to the uncertainty regarding the current and future demand for water. Yet water demand is one of the biggest and most significant uncertainties in the design of a mains network. The associated increases or decreases in water consumption can have important consequences for transport and distribution networks. In view of the long lifespan of a drinking water main, it is therefore extremely important that the newly-installed network be robust and future-proof: a mains network that can handle changes in water demand and continues performing as well as possible.
The researchers during the network’s design phase need to take into account the possible changes in water demand (both patterns and peak factors), as well as the network design requirements associated with knowledge uncertainty. This means that, when it comes to the different performance indicators (output parameters calculated per scenario, such as residence time, pressure gradient, costs) and boundary conditions, which are part of the design process, account has to be taken not only of the expected water demand, or peak factor, but also explicitly of the relevant associated uncertainty. Boundary conditions are the concrete requirements that one can impose on the design, that is, upper and lower limits for performance indicators that users can fill in – for example, a minimum node pressure higher than 300 kPa.
Water-demand scenarios and the application of numerical optimisation techniques
Taking explicit account of the water-demand uncertainty means that the researchers consider not one, but several water-demand scenarios when tackling the optimisation problem. They can therefore represent the performance indicators by means of a probability distribution. Such an approach demands the calculation of a large number of scenarios and designs: a task which quickly becomes unmanageable when done manually. It is here that numerical optimisation techniques, as implemented in Gondwana, offer a helping hand. One case study will illustrate this approach.
At the end of this research project a method will be developed for the purpose of explicitly incorporating water-demand uncertainty into the design of target structures, and thus clarify the impact of the uncertainty on various relevant performance indicators. The method is being elaborated by researchers in the Gondwana tool. The drinking water companies will therefore have a method and a tool to (1) test the robustness of the current structure against different possible water-demand scenarios, and (2) design future-proof target structures, in which the design robustness (against an uncertain water demand) is known. Drinking water companies will therefore be able to identify sensitive areas in the current mains networks and design future-proof target structures, so as to maximise the effectiveness of their mains replacement effort.