Interactive simulation of a conjunctive water supply system

Abstract

This study is directed at the investigation of the management of the metropolitan Adelaide Water Supply system in South Australia using an interactive simulation model. This supply system serves a consumer population of about one million people and is derived from ten major storage dams, and three large pumping systems which draw water from the River Murray approximately 60 kilometres to the east of the ranges in which the dams are located. Without augmentation of supply from the Murray, and relying only on local rainfall, Adelaide in particular and South Australia in general, could maintain neither the level of industry, standard of living, nor development which are currently required for viability as a modern population centre. Due to the irregular distribution of rainfall, both spatially and temporally, the tuning and specification of pumping operations vary over a wide range of possible combinations particularly with the high demand resulting from a usually long and dry summer. No direct optimising algorithm has yet been developed to manage the system for least cost. The relative constancy of most operating and maintenance costs throughout any year leaves the highly variable pumping cost of the River Murray supplies as a most significant cost item in system operation. Due to the complexity of the system and supply irregularities, sound operator experience is shown to be vital for the successful operation of the system. To investigate the effect of this personal experience on the efficiency of operation, and to facilitate the rapid learning and practicing of control skills which are needed to build experience with the real system, an interactive, graphic simulation model is used. This model is not designed to define or optimise a management policy, but rather to provide the rapid, interactive tool by which the human operator may simulate his own desired management strategy over an extended time period, given unknown system inflows and demands. The model also enables the close scrutiny of different individuals' personal approaches to water supply management, and what, if any, similarities or differences become apparent in their performances.