| 引用本文: | 白牧可,唐 巍,谭 煌,高 峰,闫 涛.基于虚拟分区调度和二层规划的城市配电网光伏-储能优化配置[J].电力自动化设备,2016,36(5): |
| BAI Muke,TANG Wei,TAN Huang,GAO Feng,YAN Tao.Optimal PV-generation & ES configuration based on virtual partition scheduling and bi-level programming for urban distribution network[J].Electric Power Automation Equipment,2016,36(5): |
|
| 摘要: |
| 针对含高渗透率分布式能源城市配电网的特点，提出基于虚拟分区和二层规划的光伏电源与储能规划模型和方法。依据历史数据，采用密度聚类提出光伏和负荷分时段多状态生成方法，该方法能考虑光伏与负荷的时序性、随机性及内在关联，且状态数大幅减少。以年费用最小为上层规划目标函数，光伏安装位置和容量为上层优化变量，等效负荷方差和最小为下层规划目标函数，虚拟分区和储能安装位置与容量为下层优化变量，同时考虑各区内储能的优化调度，建立基于虚拟分区的光伏和储能配置的二层规划模型。采用精英保留策略遗传算法和粒子群优化算法联合迭代求解模型。IEEE 33节点系统仿真结果验证了所提模型和方法的有效性，所获得的光伏-储能配置方案能够有效提升配电网的经济效益和减少网损。 |
| 关键词: 光伏发电 虚拟分区 储能 优化配置 二层规划 城市配电网 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金资助项目（51377162）；中央高校基本科研业务费专项资金资助项目（2013YJ008） |
|
| Optimal PV-generation & ES configuration based on virtual partition scheduling and bi-level programming for urban distribution network |
|
BAI Muke, TANG Wei, TAN Huang, GAO Feng, YAN Tao
|
|
College of Information and Electrical Engineering，China Agriculture University，Beijing 100083，China
|
| Abstract: |
| According to the features of urban distribution network with a high penetration of distributed generation，a planning model and method of PV（PhotoVoltaic） generation and ES（Energy Storage） are proposed based on the virtual partition and bi-level programming. Based on the historical data，the density clustering is adopted to generate the multiple states of PV-generation and load for different periods，which considers the time-sequence，randomness and their internal correlation of PV-generation and load to greatly reduce the number of states. A bi-level programming model of PV-generation and ES based on the virtual partition is established，which takes the minimum annual cost as the objective function of upper level programming，the installation position and capacity of PV-generation as the variables of upper level optimization，the minimum sum of equivalent load variance as the objective function of lower level programming，the installation position and capacity of virtual partition and ES as the variables of lower level optimization，as well as considers the optimal dispatch of ES for each partition. The elitist strategy genetic algorithm and particle swarm optimization algorithm are applied to solve the model. The simulative results of IEEE 33-bus system verify the effectiveness of the proposed model and method. The obtained PV-generation & ES configuration scheme can effectively enhance the network economy and reduce the network loss. |
| Key words: photovoltaic generation virtual partition energy storage optimal configuration bi-level programming urban distribution network |
