|国家科技期刊平台
首页|期刊导航|电工技术学报|考虑5G基站储能参与配电网供电恢复研究

考虑5G基站储能参与配电网供电恢复研究OA北大核心CSTPCD

Distribution Network Restoration Supply Method Considers 5G Base Station Energy Storage Participation

中文摘要英文摘要

5G的快速发展使得 5G基站储能容量大幅增加,如何充分发挥时常处于休眠状态的基站储能资源,使其参与配电网供电恢复,降低配电网失电量问题亟待研究.该文通过协调基站储能参与配电网供电恢复,以发挥其储能价值.首先,建立风光联合出力场景集,通过赤池信息准则和平方欧式距离确定各时段最优Copula函数,进而对其进行抽样聚类,获得风光典型出力场景集;然后,建立基站备用储能模型,通过正弦函数叠加建立不同区域的通信量模型,利用泰尔熵和修正基尼系数建立综合脆弱度模型确定各基站储能备用时间及备用储能模型,进而获得不同基站储能的可调度容量;最后,以失负荷量损失最小为目标,建立考虑基站储能参与及风光出力波动的含可调参数的两阶段鲁棒优化模型.以改进的IEEE 33 节点模型为例进行仿真验证,分别对比了定基站备用时间与该文所提方法确定备用时间、鲁棒优化与确定性优化等方案下系统失负荷量的变化情况.仿真结果表明,5G基站储能可降低失负荷量损失成本,提高风光消纳率.

China has vigorously promoted the development of 5G communication in recent years.The number of 5G base stations is growing rapidly,greatly demanding energy storage capacity.At present,China's power supply reliability has improved.Base stations'energy storage is often in a dormant state,thus causing a large amount of energy storage resource waste.It is urgent to study how to use the base station's energy storage resources and play its energy storage value.This paper proposed a 5G base stations'energy storage scheduling model,which jointly participates in the power supply restoration of the distribution network by combining wind-solar output,thus reducing the power loss load of the distribution network. Firstly,the wind-solar combined output scenario set is established.The optimal Copula function in each period is determined by the Akaike information criterion and squared Euclidian distance,obtained the typical wind-solar output scenario set by sampling and clustering.Second,a backup energy storage model for 5G base stations is established.Different regions communication traffic volume models are established through sine function superposition.The synthetic vulnerability model is established by using the Theil's entropy and the modified Gini coefficient to determine the backup time and the backup storage model of each base station energy storage.Then the schedulable capacity of different base stations'energy storage is obtained.Finally,a two-stage robust optimization model considering the synergistic scheduling of base stations'energy storage and wind-solar output is established with the objective of minimizing the loss of load volume. Simulations are performed with the improved IEEE 33 node model.The results indicate that the reserve capacity of 5G base station'energy storage is directly proportional to the node vulnerability of the power grid and inversely proportional to the load level.Compared with the traditional fixed backup time,this paper's method can increase the base stations'callable capacity by 317 kW·h.When performing emergency power restoration to the distribution network,it reduced the lost load by 1 609 kW and the cost of lost power by ¥47 208.9 compared to the fixed backup time approach.At the same time,it can use the 5G base station's energy storage for wind power and photovoltaic absorption during the restoration of the power supply in the distribution network.Among them,the wind-light absorption rate increases by 0.007 9,0.009 3,and 0.181 8 at the moments of t=14 h,t=15 h,and t=16 h,respectively,which in turn reduces the wind and light abandonment rate of the system and improves the distribution network economics. The following conclusions can be drawn from the simulation analysis:(1)Make full use of base station energy storage,improve the utilization rate of base station energy storage,and reduce the amount of distribution network load loss.At the same time,it can enhance the wind-solar absorption rate and reduce the degree of wind and light abandonment.(2)Combining the vulnerability of the grid node where the base station is located and the communication volume,the backup energy storage is dynamically determined,which can make full use of the energy storage resources of the base station and bring its value into play.

王晓卫;康乾坤;梁振锋;郭亮;张帆

西安理工大学电气工程学院 西安 710048国网江西省电力有限公司电力科学研究院 南昌 330000

动力与电气工程

5G基站备用储能供电恢复修正基尼系数

5G base stationbackup energy storagepower supply restorationcorrected Gini coefficient

《电工技术学报》 2024 (011)

基于原子稀疏分解的配电网故障选线机理研究

3538-3555 / 18

国家自然科学基金(52177114,61403127)和国网江西省电力有限公司科技项目(521820210005,521820220016)资助.

10.19595/j.cnki.1000-6753.tces.230341

评论