Research Thesis Topic
Fuzzy-Neural Network Based Flow Prediction and Optimized Operation For River Murray System
The river/catchment flow models of River Murray System are highly complex hydrological models with many uncertain environmental conditions, and the River Murray Operations are very complex functions to direct water releases from storages and control the diversions of water from the River Murray for irrigation and agricultural use, and for consumers in urban areas. Both hydrological modelling and water operation are the key functions of the Murray-Darling Basin Authority (MDBA), which manages the River Murray System in close cooperation with state authorities to ensure reliable water supplies for all users.
The existing river/catchment models used in MDBA have been continuously developed for decades. They fit into the River Murray Operations requirements. The River Murray Operations decisions are made by running a series of river/catchment models sequentially from upstream to downstream, and by taking in a range of technical considerations. The operation by using existing models, however, doesn’t contain any automatic optimization scheme. For example, the existing models used for flow prediction are based on historic data without considering rainfall forecasts, which would impact on the predictive accuracy. A lot of decisions depend upon how an operator responds to the model run results and understands of technical considerations.
The objectives of this study are:
• to investigate alternative ways of flow prediction modelling.
• to develop practical optimization methods for River Murray Operations.
• to develop a user-friendly software tool implementing flow prediction modelling and optimizing operation for River Murray System.
- School of Agricultural, Computational and Environmental Sciences
- Artificial Intelligence and Image Processing
- Electrical and Electronic Engineering
Please review the admission requirements for the academic program associated with this Thesis Topic