Keywords
Data-based modelling; Model Predictive Control; Chance-constrained optimisation problem; Scenario approach; Flood risk mitigation
Location
Session A1: Environmental Fluid Mechanics - Theoretical, Modelling and Experimental Approaches
Start Date
12-7-2016 5:50 PM
End Date
12-7-2016 6:10 PM
Abstract
Here we present findings from on-going research on how to improve the management and efficiency of river operations using data based modelling and control engineering. The objectives of the river operations include minimisation of flow and water level deviations from their required set- points, on-time water deliveries to irrigators and flood risk mitigation. Due to long time delays in rivers, forecasts of flows in tributaries are required, and control strategies for rivers should be able to accommodate such forecasts. Here we propose to use system identification techniques to obtain models of the rivers which are very simple, but sufficient for control design, and a Stochastic Model Predictive Control strategy for control of a river. The control problem is formulated as a chance- constrained optimisation problem and a solution is found using a scenario approach. The developed modelling and control framework has been successfully applied, in simulations, to the upper part of Murray River in Australia. The strategy worked well and achieved the objectives identified by the river operators.
Included in
Civil Engineering Commons, Data Storage Systems Commons, Environmental Engineering Commons, Hydraulic Engineering Commons, Other Civil and Environmental Engineering Commons
System Identification and Control of Rivers
Session A1: Environmental Fluid Mechanics - Theoretical, Modelling and Experimental Approaches
Here we present findings from on-going research on how to improve the management and efficiency of river operations using data based modelling and control engineering. The objectives of the river operations include minimisation of flow and water level deviations from their required set- points, on-time water deliveries to irrigators and flood risk mitigation. Due to long time delays in rivers, forecasts of flows in tributaries are required, and control strategies for rivers should be able to accommodate such forecasts. Here we propose to use system identification techniques to obtain models of the rivers which are very simple, but sufficient for control design, and a Stochastic Model Predictive Control strategy for control of a river. The control problem is formulated as a chance- constrained optimisation problem and a solution is found using a scenario approach. The developed modelling and control framework has been successfully applied, in simulations, to the upper part of Murray River in Australia. The strategy worked well and achieved the objectives identified by the river operators.
