Optimal management strategies for a cascade reservoir system

Forknall, Clayton (2014) Optimal management strategies for a cascade reservoir system. Honours thesis, University of Southern Queensland. (Unpublished)


Download (2MB) | Preview
[img] Other (Email re supporitng files)
Email re supporting files.msg


Water reservoirs have long been used throughout Australia and the globe as a means of both providing communities with water security during periods of limited rainfall, as well as a form of defence against severe flooding. In recent times, the effective management of these water reservoirs has been questioned and is now, more than ever, under scrutiny.

In order to address the issue of reservoir mismanagement, this thesis demonstrates the methods and procedures undertaken in the development, formulation and application of two Mixed Integer Linear Programming (MILP) models that have the ability to determine strategies for the optimal management of a cascade reservoir system, under the two extreme environmental conditions of drought and flood. For
the purposes of this thesis, the unique cascade conguration of a reservoir system was primarily considered; where cascade refers to a multiple reservoir system in which the spill from earlier reservoirs becomes a source of inflows to subsequent reservoirs. Many physical reservoir systems exhibit this type of layout including the Perseverance and Cressbrook system located near Toowoomba, which has been
considered as a case study throughout this thesis.

By applying the drought and flood models to the case study of the Perseverance and Cressbrook cascade reservoir system, it was found that both models provided com-
prehensive approximations of the system behaviours under the differing extreme conditions considered by each model. However, in order to conduct a successful comparison of the management strategies employed by the drought and flood
models, a common set of inflow records upon which both models could be considered was required. Rather than using a portion of the historic inflow records sourced
for the case study considered, time series analysis was employed instead to select a time series model that suitably represented the historic records, and then from this model, an alternate set of inflows was simulated.

Using the simulated set of inflows, a comparison of the management strategies employed by the two MILP models for a drought and flood was conducted; demonstrating both similarities and differences between the optimal strategies employed for the management of the cascade reservoir system. The comparison also revealed that although 'common sense' practices could be employed to operate the cascade
reservoir system, these practices were not optimal and thus did not result in the effective management of the system. Therefore, models like those developed, formulated, and utilised in this thesis are necessary to ensure that a commodity as heavily relied upon and sometimes as potentially dangerous as water is optimally governed and regulated into the future.

Statistics for USQ ePrint 25948
Statistics for this ePrint Item
Item Type: Thesis (Non-Research) (Honours)
Item Status: Live Archive
Additional Information: B. Sci (Honours) thesis. For supporting input/output files, please contact the ePrints Editor at: eprints@usq.edu.au
Faculty/School / Institute/Centre: Historic - Faculty of Health, Engineering and Sciences - School of Agricultural, Computational and Environmental Sciences (1 Jul 2013 - 5 Sep 2019)
Supervisors: Langlands, Trevor
Date Deposited: 09 Sep 2014 00:28
Last Modified: 10 Sep 2014 00:18
Uncontrolled Keywords: reservoir management; reservoir mismanagement; management strategies; cascade reservoir system; drought; flood; Perseverance; Cressbrook
Fields of Research (2008): 01 Mathematical Sciences > 0102 Applied Mathematics > 010206 Operations Research
Fields of Research (2020): 49 MATHEMATICAL SCIENCES > 4901 Applied mathematics > 490108 Operations research
URI: https://sear.unisq.edu.au/id/eprint/25948

Actions (login required)

View Item Archive Repository Staff Only