在孤岛微电网中，由于分布式电源（Distributed Generation，DG）之间线路阻抗的不匹配，导致了传统下垂控制无法按DGs的容量比例均分其输出的无功功率，从而使DGs之间产生了无功环流，影响了系统的稳定运行。同时，为了更好的利用随处分散的可再生能源，灵活性、可靠性以及可扩展性更高的分布式控制逐渐成为了孤岛微电网的主要控制方式。在分布式控制中，通信网络的带宽和质量都会影响系统的控制效果。因此本文针对孤岛微电网中无功功率不均分的问题和通信网络对于分布式无功均分策略的影响展开了如下研究：

（1）为了实现多DGs并联的微电网系统的稳定运行，首先建立了基于dq坐标系的“下垂控制+电压电流双闭环”的微电网模型，实现了多DGs并联系统电压和频率的稳定以及有功功率的均分。然后详细研究了传统下垂控制，分析了不同下垂系数对于系统稳定性的影响。最后通过分析传统下垂控制的功率均分特性得出了其无法实现无功功率均分的原因。

（2）针对孤岛微电网中传统下垂控制无法实现DGs之间输出无功功率均分的问题，提出了三种自适应无功功率均分策略。首先详细讨论了现有的三种无功功率均分策略的原理。然后提出了三种自适应无功功率均分策略，详细地阐述了其控制过程。所提的控制策略都可以精确地实现DGs之间输出无功功率的均分，并且不需要知道线路阻抗。最后搭建了仿真模型验证了所提控制策略的有效性。

（3）针对所提的无功功率均分控制策略是否具备分布式特性的问题，设计了分布式动态观测器。首先对所提控制策略中需要分布式获取的控制变量进行了分类，确定了观测器的设计要求。然后以分布式一致性算法为基础，设计了分布式动态观测器，并分析了其获取动态信息的能力，确保了DGs可以分布式地获取到所需要的控制信息。最后搭建了仿真模型验证了所设计观测器的有效性以及无功功率均分策略“即插即用”的有效性。

（4）针对所提的分布式无功均分策略依赖通信网络的问题，设计了分布式无功均分事件触发控制器。首先根据事件触发控制的原理设计了分布式无功功率均分事件触发控制器。然后分析了不同类型的事件触发函数的特点，对其存在的不足进行了改进，通过李雅普诺夫函数证明了其稳定性。所提的分布式无功功率均分策略在完成控制目标的同时可以有效地降低DGs之间的通信次数，提高了分布式控制器对于通信网络的利用率并且可以在一定程度的通信延迟下完成控制目标。最后在仿真模型中验证了所设计的事件触发控制器的有效性。

In islanded microgrids, the mismatch of line impedances between Distributed Generation (DG) units leads to the failure of traditional droop control to share the reactive power output according to the capacity ratio of DGs. This results in reactive power circulation between DGs, which affects the stability of the system. Concurrently, to better harness the widely dispersed renewable energy sources, distributed control with higher flexibility, reliability, and scalability has gradually become the primary control method for islanded microgrids. In distributed control, both the bandwidth and quality of the communication network can impact the system's control performance. Therefore, this thesis addresses the issue of uneven reactive power sharing in islanded microgrids and conducts the following research on the impact of communication networks on distributed reactive power sharing strategies:

(1) To achieve stable operation of a microgrid system with multiple DGs in parallel, a microgrid model based on the dq coordinate system with “droop control + voltage and current dual closed-loop” was first established. This model ensures the stability of voltage and frequency and the equal sharing of active power for a system with multiple DGs in parallel. Then, a detailed study of traditional droop control was conducted, analyzing the impact of different droop coefficients on system stability. Finally, by examining the power sharing characteristics of traditional droop control, the reason for its inability to achieve equal distribution of reactive power was deduced.

(2) To address the issue that traditional droop control in islanded microgrids cannot achieve sharing of reactive power output among DGs, three adaptive reactive power sharing strategies were proposed. Initially, a detailed discussion of the principles of the existing three reactive power sharing strategies was provided. Subsequently, Then, three adaptive reactive power sharing strategies were proposed, and their control processes were elaborated in detail. The proposed control strategies can accurately achieve sharing of reactive power output among DGs without the need for knowledge of line impedances. Finally, simulation models were constructed to validate the effectiveness of the proposed control strategies.

(3) To address the issue of whether the proposed reactive power sharing control strategy has distributed characteristics, a distributed dynamic observer is designed. Firstly, the control variables that need to be distributedly obtained in the proposed control strategy are classified, and the design requirements for the observer are determined. Then, based on the distributed consensus algorithm, a distributed dynamic observer is designed, and its ability to obtain dynamic information is analyzed to ensure that DGs can obtain the necessary control information in a distributed manner. Finally, simulation models are constructed to verify the effectiveness of the designed observer.

(4) To address the issue of the proposed distributed reactive power sharing strategy’s dependence on the communication network, a distributed event-triggered controller is designed. Firstly, a distributed reactive power sharing event-triggered controller was designed based on the principle of event-triggered control. Subsequently, the characteristics of different types of event-triggering functions were examined, and improvements were made to address their shortcomings. The stability of the proposed controller was demonstrated using Lyapunov functions. The distributed reactive power sharing strategy not only achieves the control objectives but also effectively reduces the number of communications between DGs, thereby enhancing the utilization rate of the communication network for distributed controllers. Finally, the effectiveness of the designed event-triggered controller was verified in simulation models.