文章摘要
李昊飞,瞿凯平,余涛.低碳动态电-气最优能流的加速凸分散优化求解[J].电力系统自动化,2019,43(12):85-93. DOI: 10.7500/AEPS20180911008.
LI Haofei,QU Kaiping,YU Tao.Optimization Method of Accelerated Convex Decentralization for Low-carbon Dynamic Optimal Gas-Power Flow[J].Automation of Electric Power Systems,2019,43(12):85-93. DOI: 10.7500/AEPS20180911008.
低碳动态电-气最优能流的加速凸分散优化求解
Optimization Method of Accelerated Convex Decentralization for Low-carbon Dynamic Optimal Gas-Power Flow
DOI:10.7500/AEPS20180911008
关键词: 电气互联综合能源系统  天然气动态潮流  二阶锥规划  分散优化  碳交易
KeyWords: integrated electricity and natural gas energy system  dynamic optimal gas flow  second-order constraint programming  decentralized optimization  carbon trading
上网日期:2019-04-28
基金项目:国家自然科学基金资助项目(51777078)
作者单位E-mail
李昊飞 华南理工大学电力学院, 广东省广州市 510640  
瞿凯平 华南理工大学电力学院, 广东省广州市 510640 13560361373@163.com 
余涛 华南理工大学电力学院, 广东省广州市 510640  
摘要:
      在综合能源系统背景下,提出一种计及碳交易机制的低碳动态电-气最优能流模型。考虑到气流传输的延迟特性,采用时空偏微分方程精确描述气流动态。此外,该模型引入一种碳交易机制以充分发挥电-气互联系统的经济与低碳效益。为实现电-气互联网络的分散优化并得到一个可靠的高质量解,提出两种内外层相协调的凸分散优化方法--罚凸凹过程(PCCP)-Benders和PCCP-交替方向乘子算法(ADMM)。在此凸分散优化框架下,外层优化基于一种PCCP将非凸气流动态凸化成一个迭代的二阶锥规划,而内层优化分别基于Benders分解和ADMM求解外层PCCP迭代的每一步。最后,通过算例仿真对比分析了上述两种凸分散优化算法的计算性能,并详细分析了碳交易成本对系统运行的影响,所得结果验证了文中模型及算法的有效性。
Abstract:
      To adapt to the “Energy Internet”, this paper proposes a low-carbon dynamic optimal gas-power flow model with carbon trading. However, the dynamic characteristics of gas flow are remarkably slower than those of power flow, the space-time partial differential equations are used to accurately describe the dynamic gas flow. In addition, this model introduces a carbon trading mechanism for fully exploiting economic and low-carbon benefits of the integrated electricity and natural gas energy system. In order to achieve the decentralized optimization of gas-power interconnection network and obtain a high-quality solution, two kinds of inner and outer-coordinated convex decentralized optimization methods, named as penalty convex-concave procedure(PCCP)-Benders and PCCP-auternating direction method of multipliers(ADMM), are proposed. Under this framework, the outer layer based on a PCCP transforms the non-convex dynamic gas flow into a second-order constraint programming and the inner layer solves each step of the outer PCCP iteration with Benders and ADMM. Finally, a calculation example tests the optimization performance of PCCP-Benders and PCCP-ADMM, while the operation effect with different carbon trading prices is investigated in detail, the validity of model and algorithm is verified by results.
查看全文(Free!)   查看附录   查看/发表评论  下载PDF阅读器