Research On Distributed Robust Model Predictive Control Of DC Microgrid With Uncertain Structure

With the increasing proportion of DC load,DC microgrid(DC-MG)has gradually become a research focus of scholars due to its broad application prospect in order to reduce intermediate transformation and improve the utilization rate of electric energy.Since the problem of reactive power fluctuation does not exist in DC microgrid,the stability of DC bus voltage has become one of the most important indexes to judge the stable operation of DC-microgrid.During the operation of DC microgrid,due to the influence of load change,maintenance or fault,the distributed generation unit(DGUs)often needs to be plugged in or out,resulting in the change of DC microgrid system structure,which is not conducive to the stability of DC bus voltage.Therefore,it is of great value for the safe,stable and reliable operation of DC microgrid to study the structural uncertainty in DC microgrid.In a DC microgrid system composed of multiple distributed generation units interconnection,the plug in or out of a single distributed generation unit will affect the stable operation of DC microgrid.A distributed robust predictive control strategy is proposed for the uncertainty in DC microgrid.Firstly,equations are written according to KCL and KVL laws to establish a DC microgrid model,and then a corresponding controller is designed for each distributed generation unit to adjust the voltage value of the corresponding distributed generation units common coupling point,so as to ensure the stability of the DC bus voltage when distributed generation units are plugged in or out.It satisfies plug-and-play(PNP)of distributed generation units in DC microgrid system.When the interconnected system is plug in or out of distributed generation units,the stability of the system is not changed,and only the controllers of the adjacent distributed generation units are updated at most.In addition,the design of the local controller is inseparable from the corresponding distributed generation units and the interconnected circuit state.Finally,the simulation is carried out in Matlab,and the simulation results verify the performance of the controller.For the DC microgrid composed of multiple distributed generation units interconnection,the plug in or out of distributed generation units makes the microgrid structure uncertain.Based on the Tube invariant set,this thesis studies the robust predictive control method of microgrid under the condition of structural uncertainty.Firstly,based on the elliptical invariant set,the robust model predictive control(RMPC)algorithm based on the polyhedral invariant set is designed for the nominal system.Then,for the uncertainties of the DC microgrid structure caused by the distributed generation units plug in or out,a Tube invariant set is introduced,and the thesis proposes a control strategy based on the Tube invariant set,which is designed off-line and online in real-time control,so that the state of the DC microgrid will always converge to the terminal interference invariant set when the structure isuncertain,ensuring the bus voltage stability of DC microgrid when distributed generation units are plug in or out.Simulation results verify the feasibility of the control method.