Research On Optimal Consensus Secondary Control Of Microgrid Based On Event-triggering

The microgrid is an effective integration of distributed generations(DGs),which can reduce the impact of DGs on the main grid and improve the reliability of power supply by unifying the management of DGs.Droop control is widely used in the primary control of microgrids because each DG can achieve rapid frequency/voltage stabilization and initial active/reactive power sharing without relying on communication.However,traditional droop control has issues such as frequency/voltage deviation from the rated value,reactive power allocation affected by line impedance,and not considering the operating cost of microgrids.Therefore,it is crucial to study the secondary cooperative control strategy of microgrids to improve the stability of microgrids and enhance the utilization efficiency of DG.The distributed control based on Multi-Agent System(MAS)theory has the advantages of high reliability,high expandability,and low operation cost,which is suitable for the application scenario where the number of DGs in a microgrid is numerous and relatively dispersed.This paper aims to address reasonable active/reactive power sharing and frequency/voltage restoration in islanded ac microgrids,providing important theoretical and technical support for establishing a distributed cooperative control system and methodology for microgrids.The main research contents of this paper are as follows:(1)To address the issues of unreasonable reactive power distribution affected by line impedance,the contradiction between reactive power distribution control and voltage restoration control,and the redundancy problem of the traditional secondary controller structure,this article converts the two control issues of reactive power proportion allocation and average voltage restoration into a convex optimization problem.Based on the optimal consensus algorithm,a secondary reactive powervoltage controller is designed,greatly simplifying the control structure.Adjacent DGs only need to exchange reactive power distribution information to achieve coordinated reactive power proportion allocation and eliminate the impact of line impedance.At the same time,the average voltage of each DG is regulated to the rated value,avoiding the contradiction between reactive power distribution control and voltage restoration control.(2)To optimize the microgrid operation cost,DGs adopt the droop control based on the increment cost-frequency(IC-w),by establishing a linear relationship between frequency and increment cost(IC),so that the IC can spontaneously converge to the consistent with the synchronization of the ac microgrid frequency,thus realizing the economic distribution of active power among DGs.For the frequency deviation caused by the droop control,by controlling the minimum deviation between the output frequency and the rated frequency to achieve frequency recovery.So,the frequency recovery problem can also be transformed into a convex optimization problem,and the second active power-frequency controller is designed according to the optimal consensus algorithm,adjacent DGs only need to exchange frequency compensation to achieve output frequency restoration to the rated value.(3)Considering the limited communication bandwidth and channel capacity in the actual microgrid,the event-triggered communication mechanism is introduced for the designed controllers,and the corresponding event detector and trigger threshold are designed.The communication between DGs only depends on the state information of themselves and neighbors,which avoids the dependence on global information,reduces the network traffic,and improves the trigger efficiency.(4)Based on Lyapunov stability theory,the stability of reactive power-voltage controller,active power-frequency controller,event-triggered reactive power-voltage controller,and event-triggered active power-frequency controller is analyzed.The minimum time interval of communication between DGs is analyzed and the Zeno phenomenon is excluded.An islanded ac microgrid with 4 DGs connected in parallel is built in MATLAB/SIMULINK environment to verify the effectiveness and stability of the control method proposed in this paper.