擁有智慧型電網的配電系統可基於虛擬電廠(Virtual Power Plant, VPP)運轉原理進行操作，藉由 蓄電池儲能系統(Battery Energy Storage System, BESS)、各種分散式能源(Distributed Energy Resource, DER)諸如太陽能(Photovoltaic, PV)及風力(Wind Power, WP)、以及需量反應(Demand Response, DR)技術，來最佳化系統運轉效能；然而，DER 設備的安裝和維修費用相當高，VPP 必 須具有高度長期運轉效益。本論文目的首先針對包含有不同類型及數量的插電式電動車(Plug-In Electric Vehicle, PEV)、可執行DR方案的動態負載(Dynamic Loads)、特定容量的PV和BESS 的配 電系統，基於時間電價(Price of Time of Use, TOU)建立一最佳化VPP 運轉模型，接著此問題使用本 文所提基於粒子群優(Particle Swarm Optimization, PSO)演算法結合疊代動態規劃(Iterative Dynamic Programming, DP)演算法求解，來決定各BESS的充放電排程和DR方案以最佳化VPP的運轉效能； 在隨機(Stochastically)考量系統負載、PEV 行為和TOU 下，本論文發展一蒙地卡羅模擬(Monte Carlo Simulation, MCS)程序以評估VPP長期運轉效益並提出未來改進之建議。各種配電系統案例被測試 來驗證所提方法。

Distribution systems with intelligent grid can be operated based on the principle of virtual power plant (VPP) by using battery energy storage systems (BESSs), demand response (DR) technique and different kinds of distribution energy resources (DERs), such as photovoltaic (PV) energy and wind power (WP), to optimize the system efficiency. However, the high cost for installation and maintenance of DER equipments require VPP operating in high long-term efficiency. This paper first aims to formulate an optimal VPP operating model based on the time of use (TOU) prices by including different kinds and numbers of plug-in electric vehicles (PEVs), dynamic loads able to perform DR scheme, specific capacities of PV and BESS in the distribution system, and then the problem is solved with a proposed methodology combined with a PSO-based optimization method and an iterative dynamic programming approach to determine the charging/discharging schedule for each BESS and the DR scheme to optimize the VPP operating efficiency. Under considering system load, PEV behaviors and TOU stochastically, this paper used a Monte Carlo Simulation (MCS) procedure to evaluate the VPP long-term operating efficiency and give suggestions for future reinforcement. Different distribution system cases were tested to validate the proposed method.