| 引用本文: | 郝雨辰,窦晓波,吴在军,胡敏强,孙纯军,李 桃,赵 波.微电网分层分布式能量优化管理[J].电力自动化设备,2014,34(1): |
| HAO Yuchen,DOU Xiaobo,WU Zaijun,HU Minqiang,SUN Chunjun,LI Tao,ZHAO Bo.Hierarchical and distributed optimization of energy management for microgrid[J].Electric Power Automation Equipment,2014,34(1): |
|
| 摘要: |
| 以系统经济性、环保性、供电可靠性为优化目标，提出基于多代理系统的微电网分层分布式能量优化管理策略，将传统集中式能量优化转变为包含细化变量定义域的微电源分布式计算、基于遗传算法的中心控制器优化、修正优化解的分布式并行调节以及生成最终解的全局协调这4个阶段的分层分布式能量优化，并将各微电源的运行特性与用户控制目标以约束条件的形式，分散到各阶段予以实现。在此基础上，分别针对微电网单目标与多目标优化问题，介绍了各单元代理在每个阶段的工作任务和协作关系，并利用MATLAB和JADE平台构建针对未来24 h的系统优化算例。计算结果表明所提分层分布式能量优化策略能够弱化对微电网中心控制器性能的要求，并且具有更快的运算速度和较好的优化效果。 |
| 关键词: 微电网 能量管理 分层分布式优化 多代理系统 模型 |
| DOI： |
| 分类号: |
| 基金项目:国家高技术研究发展计划（863计划）资助项目（2011AA05A107）；国家自然科学基金资助项目（51177015）；江苏省产学研前瞻性联合研究项目（BY2010142） |
|
| Hierarchical and distributed optimization of energy management for microgrid |
|
HAO Yuchen1, DOU Xiaobo1, WU Zaijun1, HU Minqiang1, SUN Chunjun2, LI Tao2, ZHAO Bo3
|
|
1.School of Electrical Engineering，Southeast University，Nanjing 210096，China;2.Jiangsu Electric Power Design Institute，Nanjing 211102，China;3.Zhejiang Electric Power Test & Research Institute，Hangzhou 310014，China
|
| Abstract: |
| A hierarchical and distributed optimization of energy management is proposed based on multi Agent system for microgrid，which takes the economy，environmental protection and power-supply reliability as its objectives. It converts the traditional centralized energy optimization into four sub-optimizations：the distributed computing of micro-sources to refine the domain of each variable，the GA-based optimization of central controller，the distributed parallel regulation to modify the optimization solution and the global coordination to generate the final solution，in which，the operational features and control objectives of different micro-sources are implemented as constraints. The special tasks of diverse Agents and the collaborative relationship among them are described respectively for the single-objective and multi-objective optimizations of microgrid. A microgrid optimization model for next 24 hours is built based on MATLAB and JADE. Numerical results demonstrate that，the proposed strategy of hierarchical and distributed optimization has higher optimization efficiency and better optimization results，weakening the dependence on the performance of central controller. |
| Key words: microgrid energy management hierarchical and distributed optimization multi Agent systems models |
