Study On Finite State Model Predictive Control Of Photovoltaic Grid-connected Inverters For Micro-grid

With the global oil,electricity or other natural energy supply shortage and prices moving up caused by economic recession becoming more serious.The objective existence of all nature,society,and self that can be dominated by mankind on earth is seriously threatened.More research power on new energy technology,such as solar photovoltaic inverter technology,is bound to become a trend.And global scholars take more focus on R & D technology based on solar photovoltaic grid connected inverters.Meanwhile,Faced with the increasing number of Distributed Generation(DG),How to apply better,more flexibly and efficiently,solve their access problems.More scholars are devoted to developing and extending the micro-grid.In micro-grid,there are various control methods for the state of each part of the operation clockwise.In this paper,Model predictive control of photovoltaic grid-connected for micro-grid is studied.Firstly,the traditional classical predictive control based on the mathematical conceptual model of the controlled object(model predictive control,MPC)may require intensive computational power and high sampling intervals to evaluate all possible switching states,which may restrict its applicability when the number of voltage levels increases.In this paper,two simplified finite control-set model predictive current/power control strategies are proposed for the three level grid connected converters.More specifically,the method is ingenious to use the prejudgment of vector sector,integrating the principle of SVPWM equivalent transformation,to solve the problem of the reference output voltage vector of the three level grid connected converter,selecting the optimal switch state with the objective function from the switching state subsets of related sectors in which they are located and directly acting on the grid connected converter.Secondly,The high switching frequency of grid-connected inverters(GCIs)will increase the switching losses of power devices,and reduce the efficiency in high voltage and high power applications.In order to avoid this issue,a low switching frequency model predictive direct current control(L-MPDCC)is proposed in this paper.And this algorithm is proposed for traditional two-level inverters and three-level neutral-point clamped converters(NPCs).More specifically,this strategy uses a rolling optimization based on traditional mode which predict the state of the system corresponding to all switching states respectively.In the evaluation function,the reducing switch frequency target item is introduced to select the optimal switching state of the target function and directly acting on the three-phase grid connected inverter.Meanwhile,the proposed algorithm could effectively reduce the switching frequency and reduce the switching losses of power devices with high dynamic response.Finally,the MPC Simulink model,which corresponds to the improved algorithm,is established and also analyzes the state index of the algorithm when it is put into operation.It shows that the method can be used for reference in practical engineering.