Grid-connected Inverter Model Predictive Current Control Based On Event Trigger Mechanism

As an important direction for current development of green buildings,renewable energy buildings have made important contributions to promoting energy conservation and emission reduction in international countries.Similarly,as an important part of building electrical energy conversion,the energy transmission from renewable energy buildings to external grid is directly determined by grid-connected inverters.Therefore,grid-connected inverters play an important role in ensuring normal operations of buildings.For Model Predictive Control(MPC)strategy,its advantages of simple modeling,rapid response and easy multi-objective collaborative control make it widely used in grid-connected inverters.However,when renewable energy buildings are operated in steady-state conditions,MPC strategies still need to be periodically and frequently sampled.Therefore,the switching behavior of grid-connected inverters is tedious,and it is difficult to improve the efficiency.In response to the above problems,the mathematical model of traditional MPC strategy for L-type two-level grid-connected inverter is specifically derived and the dynamic restricted condition of switching times is added to cost function.In some ways,it is helpful to deal with those obvious disadvantages caused by MPC strategy and the switching frequency of power devices can be declined flexibly.However,for current MPC frequency reduced strategy,one objective problem that it is difficult to judge operating states of the system from the weight coefficient and the lower limit of stability cannot be determined is discovered and clarified.At the same time,the weight coefficient is easily affected by sampling accuracy and unstable.Therefore,the problem of poor robustness is caused.Finally,the cost function added with constraints needs to be traversed multiple times,and the additional computational burden of the processor is increased.In order to overcome the above-mentioned shortcomings of traditional frequency reduced strategy,a model predictive control strategy based on Event-Triggered(ET)mechanism is proposed in this paper.First,according to the principle of inequality approximation,the ET boundary of the error function is derived.Secondly,the stability of the proposed ET-MPC strategy is proven.Then,the influence of the ET period and adjustment coefficient on MPC performance is specifically analyzed in the MATLAB/SIMULINK software environment.Finally,the comprehensive performance between proposed ET-MPC and traditional frequency reduced scheme is compared in detail.An experimental prototype of a 0.9 kW L-type two-level grid-connected inverter is built in the laboratory.The experimental results show that with proposed ET-MPC strategy adopted,the stable operation of grid-connected inverter is guaranteed under the condition that the ET and sampling period are both 20 microseconds.At the same time,38%of switching times of power devices under traditional MPC strategy is effectively reduced and the system efficiency is improved.