文章摘要
刘佳,程浩忠,徐谦,等.安全距离理论下计及故障恢复的智能配电网随机规划[J].电力系统自动化,2018,42(5):64-71. DOI: 10.7500/AEPS20170412007.
LIU Jia,CHENG Haozhong,XU Qian, et al.Stochastic Planning of Smart Distribution Network Based on Security Distance Methodology Considering Fault Recovery[J].Automation of Electric Power Systems,2018,42(5):64-71. DOI: 10.7500/AEPS20170412007.
安全距离理论下计及故障恢复的智能配电网随机规划
Stochastic Planning of Smart Distribution Network Based on Security Distance Methodology Considering Fault Recovery
DOI:10.7500/AEPS20170412007
关键词: 智能配电网  随机规划  安全距离  N-1安全  混合求解算法
KeyWords: smart distribution network  stochastic planning  security distance  N-1 security  hybrid solution algorithm
上网日期:2017-08-25
基金项目:国家重点研发计划资助项目(2016YFB0900102);国家电网公司科技项目(5211JY160004)
作者单位E-mail
刘佳 电力传输与功率变换控制教育部重点实验室(上海交通大学), 上海市 200240 liujia911011@126.com 
程浩忠 电力传输与功率变换控制教育部重点实验室(上海交通大学), 上海市 200240  
徐谦 国网浙江省电力公司经济技术研究院, 浙江省杭州市 310008  
兰洲 国网浙江省电力公司经济技术研究院, 浙江省杭州市 310008  
马则良 国家电网公司华东分部, 上海市 200120  
张建平 国家电网公司华东分部, 上海市 200120  
摘要:
      在智能配电网主变站间充分互联背景下,为提高其供电效率、节省新建投资,提出了一种基于安全距离的配电网随机规划方法。首先,基于安全域理论,介绍了配电网N-1安全性分析方法和指标族。其次,考虑系统故障负荷的转供和分布式电源出力及时变负荷带来的不确定性,以总成本现值最小为目标函数,采用机会约束规划方法建立了配电网规划模型。然后,采用模糊C-means聚类法与动态小生境差分进化算法相结合的策略对模型加以求解。最后,通过算例对比所提规划方法与现有两种规划方法,结果表明,所述模型所得规划方案在保证N-1安全的前提下,实现了系统经济性和安全性的全面优化,充分发挥了配电网的供电潜力,更适用于负荷增长平缓和占地通道资源紧张的城市建成区配电网近期规划。
Abstract:
      Against the background of the interconnected structures among substations in smart distribution network, this paper proposes a stochastic planning method for distribution network based on security distance to improve power supply efficiency and save investment cost. Firstly, the method and indices for N-1 security evaluation of the distribution network are described based on security region methodology. Secondly, the fault load restoration as well as uncertainties of time-varying load and distributed generator(DG)output is considered for the chance-constrained distribution network planning model, which aims to minimize the present value of total cost. Then, the hybrid solving strategy in conjunction with fuzzy C-means clustering and dynamic niche differential evolution algorithm is utilized to solve the model. Finally, compared with two available planning methods, planning results show that the scheme proposed can improve power supply efficiency of the original network and optimize system economy and security at the same time on the premise of N-1 security. The methodology is more suitable for short-term planning of urban developed distribution networks, which have the characteristics of slow load growth and scarce line corridors.
查看全文(Free!)   查看附录   查看/发表评论  下载PDF阅读器