外部逼近的新型电力系统线损分析与优化方法OA
Line loss analysis and optimization method of novel power system of external approximation
随着电力系统规模与复杂性的不断扩大,传统线损分析方法逐渐显露出计算复杂度高、精度不足等局限,阻碍了其在实践中的有效应用.因此,文中提出了一种针对新型电力系统的线损分析与优化方法,采用线损外估模型,通过惰性逼近的方式初步估计线路损耗,并观察到在至少2 次迭代中线路损耗即可收敛,分析了该模型在直流最优潮流和安全约束经济调度中的应用算法,以提高电网安全性和稳定性,通过对新型电力系统的实验验证了所提出的线损外估模型具有显著的快速收敛特性.结果表明,与传统线损方法相比,所提出的模型在解决大规模试点时,速度快了 16 倍,并且绝对误差小于0.08%,能够在保证准确性的前提下大幅减少迭代时间和计算成本,为新型电力系统的优化调度提供了有效工具.
As the scale and complexity of power system continue to increase,traditional line loss analysis methods have gradually revealed limitations such as high computational complexity and insufficient accuracy,hindering their effective application in practice.Therefore,this paper proposes a line loss analysis and optimization method tailored for novel power systems.A line loss outer approximation model is employed,which initially estimates line losses through an inertial approximation,with convergence of the line losses observed within at least two iterations.The application algorithm of this model in DC optimal power flow and security-constrained economic dispatching is ana-lyzed to enhance power grid security and stability.Experiments conducted on novel power systems validate that the proposed line loss outer approximation model exhibits significant rapid convergence characteristics.The results indi-cate that,compared to traditional line loss methods,the proposed model achieves a computation speed 16 times fas-ter when addressing large-scale test cases,with an absolute error of less than 0.08%,substantially reducing itera-tion time and computational costs while maintaining accuracy,thus providing an effective tool for the optimal sched-uling of novel power systems.
辛翠红;钟丹丹;邹丽;李智鹏;赵鹏
国网新疆电力公司阿克苏供电公司,新疆 阿克苏 843000国网新疆电力公司阿克苏供电公司,新疆 阿克苏 843000国网新疆电力公司阿克苏供电公司,新疆 阿克苏 843000国网新疆电力公司阿克苏供电公司,新疆 阿克苏 843000重庆大学 输配电装备及系统安全与新技术国家重点实验室,重庆 400044
信息技术与安全科学
线损外估模型新型电力系统直流最优潮流电网安全
line loss outer approximation modelnovel power systemDC optimal power flowpower grid security
《电测与仪表》 2026 (6)
52-61,10
国家自然科学基金资助项目(52077015)
评论