Study On Flood Control Operation Of Pump-Sluice Group Based On Model Predictive Control

Due to the continuous accumulation of the "greenhouse effect",the frequent occurrence of extreme weather,together with the impact of human activities,the urban flooding caused by extreme weather is more serious.Pump sluice is an effective engineering measure to deal with flood disaster.Therefore,how to dispatch the city pump sluice group scientifically is of great significance.Aiming at the problem of flood Control and drainage in Fuzhou urban area,this thesis r established the channel hydraulics Model of Jin ’an River based on the Rainstorm Flood Management Model(SWMM).Based on the safety-economy level,this thesis constructed the optimal dispatching model of flood control combined with pump-sluice group,and introduced Model Predictive Control.Effectively improve the ability of optimizing scheduling model to apply uncertainty.The main research contents are as follows:(1)Study on the hydrodynamic model of Jin ’an River based on SWMM.Aiming at the requirements of fine simulation of the flood process of Jin ’an River,the backbone river in Jiangbei urban area of Fuzhou,ARCGIS was used to process the basic data such as river channel,catchment area and river section in the study area,and establish the channel hydraulics model of Jin ’an River based on SWMM.The real rainfall data in Fuzhou were selected to rate and verify the model parameters.(2)Study on the joint optimal operation model of pump-sluice group flood control.For port of fuzhou city jiangbei light drainage engineering,lute pavilion lake drainage engineering draining of flood control scheduling requirements,fully considering the pumping stations units open-close frequency measure operational maintenance cost,lute pavilion lake with p risk minimum port,light running water lowest,pumping unit scheduling cycle cost minimum,opening and closing the lowest target number of brake pump group joint optimization scheduling model,The multi-objective problem is transformed into a single objective problem based on the penalty function,and the model is solved efficiently based on the artificial bee colony algorithm.(3)Study on optimal dispatching of pump-sluice group combined flood control based on model predictive control.In order to effectively improve the scheduling effect of the combined flood control optimization model of pump-sluice group,based on the model predictive control framework,the rolling optimization and feedback correction mechanism of the combined flood control optimization model of pump-sluice group was proposed,and the combined flood control model of Jin ’an River pump-sluice group based on model predictive control was formed.By combining with the hydrodynamic model of Jin ’an River channel based on SWMM,the data of 24 periods of inlet flow of a typical design flood in 100 years were simulated,and it was verified that the combined flood control optimization scheme of pump and sluice group after MPC rolling optimization had lower peak water level,less pumping and discharging costs,and fewer opening and closing times of sluice pump than the regular scheme.Therefore,the scheduling scheme based on MPC rolling optimization is better than the rule scheduling scheme.It can provide decision support for flood season scheduling in Jin ’an River basin and also provide some ideas for the research of model predictive control strategy in the optimization of pump gate group scheduling.