Optimising Pumping Activation in Multi-Reservoir Water Supply Systems under Uncertainty with Stochastic Quasi-Gradient Methods

Under conditions of water scarcity, the energy saving in operation of water pumping plants and minimization of water deficit for users are frequently contrasting requirements, which should be considered when optimizing multi-reservoirs and multi-user water supply systems. This problem is characterised by a high uncertainty level in predicted water resources related to hydrologic input variability and water demand behaviour. We develop a mixed simulation-optimisation model using the stochastic quasi-gradient optimisation method to get robust pumping activation threshold values. This method allows solving complex problems, dealing efficiently with large size real cases with considerable number of data parameters and variables. The threshold values are chosen in terms of critical storage levels in the supply reservoirs. The optimal rules are obtained considering both historical and generated synthetic scenarios of hydrologic inputs to reservoirs. Hence, using synthetic series, we can analyse climate change impacts and optimise the activation rules considering future hydrologic conditions. The considered case-study is a multi-reservoir and multi-user water supply system in South Sardinia (Italy), characterised by Mediterranean climate and high annual variability in hydrological inputs to reservoirs. By applying the combined simulation and optimisation procedure, using the stochastic quasi-gradient method, a robust decision strategy in pumping activation was obtained.