文章摘要
韩畅,林振智,杨莉,等.考虑经济性和网架性能的抗灾型骨干网架多目标规划[J].电力系统自动化,2019,43(2):34-41. DOI: 10.7500/AEPS20180313002.
HAN Chang,LIN Zhenzhi,YANG Li, et al.Multi-objective Planning for Anti-disaster Backbone Grid Considering Economics and Network Frame Performance[J].Automation of Electric Power Systems,2019,43(2):34-41. DOI: 10.7500/AEPS20180313002.
考虑经济性和网架性能的抗灾型骨干网架多目标规划
Multi-objective Planning for Anti-disaster Backbone Grid Considering Economics and Network Frame Performance
DOI:10.7500/AEPS20180313002
关键词: 抗灾型骨干网架  可恢复性  网络抗毁性  全面学习粒子群优化  纳什均衡
KeyWords: anti-disaster backbone grid  resilience  network survivability  comprehensive learning particle swarm optimization  Nash equilibrium
上网日期:2018-11-09
基金项目:国家重点研发计划资助项目(2016YFB0900100);国家自然科学基金资助项目(51377005)
作者单位E-mail
韩畅 浙江大学电气工程学院, 浙江省杭州市 310027  
林振智 浙江大学电气工程学院, 浙江省杭州市 310027  
杨莉 浙江大学电气工程学院, 浙江省杭州市 310027 eeyangli@zju.edu.cn 
蔡景东 广东电网有限责任公司惠州供电局, 广东省惠州市 516001  
吕云锋 广东电网有限责任公司惠州供电局, 广东省惠州市 516001  
张素明 广东电网有限责任公司惠州供电局, 广东省惠州市 516001  
摘要:
      为增强极端自然灾害下电力系统的供电能力和抗灾能力,提出一种抗灾型骨干网架的多目标规划方法。该方法在满足负荷保障率、电力系统安全运行和网络拓扑连通性约束的基础上,综合考虑差异化规划加固费用、灾害后恢复全网供电的效率和骨干网架抵御灾害的能力,构建以最大化经济性、系统可恢复性和网络抗毁性为目标的抗灾型骨干网架优化模型。采用嵌入图论修复策略的全面学习粒子群优化算法求解模型,增大了算法的可行解空间。引入混合策略纳什均衡来选取算法所求得的帕累托解集中具有最优联合均衡值的前沿解作为最优的骨干网架方案,从而能较好地兼顾各个目标函数的利益。广东某区域电网的仿真结果表明了所提方法的有效性。
Abstract:
      In order to enhance the power supply capability and disaster resistance ability of power systems under extreme natural disasters, a multi-objective planning method of anti-disaster backbone grid is proposed with the guarantee rate of the load requirements, security operation constraints of the power system and connectivity of the network topology satisfied. In the proposed strategy, a multi-objective planning model of anti-disaster backbone grid, in which the reinforce cost of differential planning, the efficiency of power resupply after a disaster and the ability of the backbone grid to resist the disaster are considered comprehensively, is constructed for maximizing economics, resilience and network survivability. The graph repair strategy is utilized in comprehensive learning particle swarm optimization algorithm, which increases the feasible solution space of the algorithm. Then, the mixed strategy Nash equilibrium, which can balance the benefit of each objective function, is adopted to extract the best compromise solution with the optimal equilibrium value from the Pareto fronts obtained by the algorithm. The feasibility of the proposed method is verified by the numerical results of a regional power grid in Guangdong Province.
查看全文(Free!)   查看附录   查看/发表评论  下载PDF阅读器