This paper proposes a novel sequential optimization strategy (SOS) for optimal allocation of both active and reactive power among dispatchable distributed generator (DDG) units present in a droop controlled islanded AC microgrid. Active power is optimally dispatched based on the simultaneous satisfaction of an economic objective of total operational cost (TOC) minimization, along with an environmental objective of total operational emission (TOE) minimization, as well as a network-related objective of total active power loss ($P_{loss}$) minimization. A fuzzy-embedded multi-objective particle swarm optimization (FMOPSO) technique is used to solve this optimization problem. Reactive power is optimally allocated depending on the objective of capacity based reactive power sharing, which is solved using the conventional particle swarm optimization (PSO) algorithm. Node voltage deviation index (NVDI) and total active power loss in the network ($P_{loss}$) are used as two main attributes to quantitatively measure the improvement in node voltage profile as a result of optimal sharing of reactive power among the dispatchable distributed generator units. The presence of local heat demand and uncertainties inherent in weather dependent load demand and renewable generation are taken into consideration. A novel mixed probabilistic–possibilistic scenario based approach (MPPSBA) is put forward to model the uncertainties associated with both the electrical and heat load demand and renewable distributed generator (RDG) output. The effectiveness of the proposed methods is demonstrated on a 33-node droop controlled islanded microgrid (DCIMG) test network.

本文提出了一种新颖的顺序优化策略（SOS），用于在下垂控制的孤岛式交流微电网中存在的可调度分布式发电机（DDG）单元之间对有功和无功功率进行最佳分配。基于同时满足总运营成本（TOC）最小化的经济目标，总运营排放（TOE）最小化的环境目标以及与网络相关的总有功功率损耗目标，可以优化分配有功功率（$P_{升Øss}$）最小化。一种模糊嵌入的多目标粒子群优化（FMOPSO）技术用于解决此优化问题。无功功率是根据基于容量的无功功率共享的目标进行最佳分配的，这是使用常规粒子群优化（PSO）算法解决的。网络中的节点电压偏差指数（NVDI）和总有功功率损耗（$P_{升Øss}$）被用作两个主要属性，用于定量测量可分派的分布式发电机组之间无功功率的最佳共享所导致的节点电压曲线的改善。考虑了局部热需求的存在以及与天气有关的负荷需求和可再生能源发电固有的不确定性。提出了一种新颖的基于概率-可能性的混合情景方法（MPPSBA）来建模与电力和热负荷需求以及可再生分布式发电机（RDG）输出相关的不确定性。在33个节点的下垂控制的岛状微电网（DCIMG）测试网络上证明了所提出方法的有效性。