Optimisation modelling tools and solving techniques for integrated precinct-scale energy–water system planning

This invited review paper presents state-of-the-art applications to energy–water system optimisation with multiple technologies, implying possible (non)monetary benefits/impacts for facility custodians in urban precincts. To solve outstanding problems, optimisation modelling frameworks were examined in the quest for undertaking an integrated precinct-scale energy–water system planning. Underlying optimisation methods were identified for addressing nested long/short-term capital/managerial decisions that ensure cost-effectiveness and system reliability, especially selecting diversified investment portfolios to allocate controllable assets across energy and water resources. Despite the ongoing energy–water nexus investigations, the integrated energy–water system planning seems to lack a general optimisation problem formulation that reveal synergies between energy and water. Considering the dynamic and nonlinear nature of energy–water interactions, mathematical properties must be carefully treated for tackling complex integrated energy–water system planning optimisation problems. To optimise integrated precinct-scale energy–water system plans, classical and heuristic techniques can be jointly implemented for solving (re)formulated problems with hard constraints and desirable cost/resource-saving objectives. Critical energy–water system sizing/scheduling tasks were discussed while covering deterministic and probabilistic procedures to support decision-making under uncertainty. Although different combined stochastic programs have been proposed incorporating actual outcomes and empirical measurements as well as prior information and expert knowledge, inherent energy–water demand/supply uncertainties should be realistically characterised by parameterising representable random perturbation sets. To find robust optimal feasible solutions to sensitive parameters during extreme load/cost volatility situations, historical energy–water usage data can be randomly perturbed for uncertainty realisation in prospective scenarios, capturing spatiotemporal correlations that avoid overconservative worst-case expected values. From the scientific paradigm of physics-informed data-driven optimisation modelling, robust approaches can be adopted for severe societal/natural events to advance state-of-the-art applications in integrated precinct-scale energy–water system planning and engineering design.