PDF(2517 KB)
Post-fault Self-healing Reconfiguration Strategy for Low-Voltage Passive Station Areas Based on Fog Computing Load Prediction
LIU Yin, GUI Yuan, LIU Ruoxi, SONG Yifan, GAO Yang, YANG Wenqin, WANG Yue
Electric Power Construction ›› 2025, Vol. 46 ›› Issue (11) : 35-46.
PDF(2517 KB)
PDF(2517 KB)
Post-fault Self-healing Reconfiguration Strategy for Low-Voltage Passive Station Areas Based on Fog Computing Load Prediction
[Objective] To improve the intelligence level of fault recovery in low-voltage substations， a low-voltage passive substation post-fault self-healing strategy based on fog computing load prediction was proposed for the problem of low-voltage substation fault recovery.[Methods] First， to avoid equipment overload caused by network reconstruction， the load level of the network must be determined in advance. Combining the typical structure of low-voltage passive substations and the characteristics of fog computing communication architecture， a fog computing ultra-short-term load prediction method based on the dynamic aggregation of an incremental learning model was designed. This method embedded two ultrashort-term load prediction technologies with complementary characteristics. It used a real-time load for model incremental learning in a dynamic weighted manner and rapidly predicted low-voltage loads in a fault event-triggered manner. In addition， based on the proposed fog computing load prediction， a low-voltage passive substation switch reconstruction self-healing recovery model without line parameters was proposed and modeled as a mixed-integer quadratic programming problem.[Results] The simulation results showed that the average absolute scale-free error of the proposed fog computing load prediction was mainly affected by the load mutation that could be controlled between 5 and 40， and the relative error was between 1% and 8%.[Conclusions] The proposed post-fault self-healing strategy effectively completed the transfer of single-phase loads and maintained the inter-phase load balance as much as possible， while maintaining the radial network operation and avoiding equipment overload.
low-voltage power station / fault recovery / fog computing / ultra-short-term load forecasting / mixed-integer quadratic programming
| [1] |
王浩成, 宋厚营, 于佳. 新型电力系统终端通信网优化方案研究[J]. 山东电力技术, 2024, 51(2): 56-66.
|
| [2] |
张弘喆, 丛伟, 王瑞梅, 等. 基于Bhattacharyya系数和Pearson相关系数的交直流混联电网电压波形相似度纵联保护方法[J]. 山东电力技术, 2024, 51(12): 19-26.
|
| [3] |
袁林涛, 孙媛媛, 刘振, 等. 基于潮流变向点的配电网低压分布式光伏无功影响及优化[J]. 山东电力技术, 2023, 50(8): 11-19, 41.
|
| [4] |
范馨予, 黄媛, 吴疆, 等. 考虑源网荷储协同优化的配电网韧性提升策略[J]. 电力建设, 2023, 44(4): 63-73.
随着分布式电源的规模化接入,针对极端灾害引起的大规模停电事故难以采用传统的配电网故障恢复策略。首先,提出了极端灾害下考虑源网荷储协调优化提升配电网韧性的策略框架。其次,针对分布式能源出力的不可控性和时变性,建立了光储和风储系统模型(optical storage and wind storage system,OWS)。同时,考虑到负荷的价格需求弹性,建立了极端灾害下的负荷需求响应(load demand response,LDR)模型。再次,以负荷恢复的总价值最大为目标,考虑LDR补偿、故障抢修与网络重构过程中的网损成本,建立了考虑源网荷储协同优化的配电网韧性提升模型。最后,在改进的PG&E 69节点配电网系统算例中验证了所提策略的有效性,结果表明利用多能互补的特性进行源网荷储的协同优化有利于提高配电网的故障恢复能力。
With the large-scale access of distributed power generation, the traditional fault-recovery strategy for distribution network is difficult to be applied to large-scale blackouts caused by extreme disasters. Firstly, this paper proposes a strategic framework to optimize and improve the resilience of distribution network considering the coordination of source, network, load and storage under extreme disasters. Secondly, aiming at the uncontrollability and time variability of distributed energy output, the PV-storage and wind-storage (PWS) system models are established. At the same time, considering the relationship between electricity price demand response (DR) and load demand, a load demand response (LDR) model under extreme disasters is established. Thirdly, aiming at maximizing the total value of load recovery, considering the cost of network loss in the process of LDR compensation, fault repair and network reconstruction, a resilience promotion model for distribution network considering the source-network-load-storage collaborative optimization is established. Finally, the effectiveness of the proposed method is verified in an improved PG &E 69-node distribution network system. The results show that the source-network-load-storage cooperative optimization according to the characteristics of multi-energy complementation is beneficial to improve the fault recovery ability of the distribution network. |
| [5] |
张晓丹. 基于深度学习的低压台区故障隔离及自愈调度方法[J]. 电脑编程技巧与维护, 2024(2): 128-130.
|
| [6] |
王长伟, 招永锦, 吴晗, 等. 考虑新型配电元件多状态可靠性模型的配电网可靠性评估[J]. 供用电, 2023, 40(4): 74-82.
|
| [7] |
沈兴杰, 陈沛, 高群, 等. 考虑多种故障场景的柔性互联低压配电网故障恢复方法[J]. 机械与电子, 2024, 42(3): 17-25.
|
| [8] |
周燕, 刘卫民, 陈帆, 等. 不同光伏渗透率下考虑需求响应的配电网储能双层规划[J]. 高压电器, 2024, 60(10): 64-77.
|
| [9] |
王尚斌, 苑丽伟, 方莹, 等. 计及新能源波动的分布式储能分层控制策略[J]. 山东电力技术, 2024, 51(9): 74-84.
|
| [10] |
李军, 梁宵, 梁嘉城, 等. 基于ARIMA-GM改进算法的建筑负荷预测[J]. 电气自动化, 2023, 45(6): 59-61.
|
| [11] |
叶永盛, 徐燕龙, 李阳, 等. 基于相似周和模态分解的融合模型电动汽车负荷预测[J]. 电网技术, 2025, 49(5): 1910-1919.
|
| [12] |
李丽芬, 张近月, 曹旺斌, 等. 基于双重注意力时间卷积长短期记忆网络的短期负荷预测[J]. 系统仿真学报, 2025, 37(8): 2004-2015.
In order to improve the accuracy of load forecasting and fully extract the hidden relationships between load and other characteristic factors, a load forecasting method based on dual-attention temporal convolutional LSTM network (DA-TCLSNet) was proposed. Correlation analysis was conducted on the dataset using the maximum information coefficient method to perform feature screening to reduce the computational cost of the model. The model input was constructed using a sliding window. The DA-TCLSNet forecasting model was constructed. The temporal convolutional layer extracted dependencies at different time scales and captured the nonlinear characteristics among variables such as load and weather. The multi-head sparse self-attention layer focused on important information. The LSTM network layer explored the long-term dependency of time series. The temporal pattern attention layer realized adaptive learning of the relationships among different variables at the same time step and connected the above modules through a residual structure to improve the expressive power of the model. Experimental results show that the proposed algorithm offers better forecasting performance than other load forecasting methods. |
| [13] |
孟衡, 张涛, 王金, 等. 基于多尺度时空图卷积网络与Transformer融合的多节点短期电力负荷预测方法[J]. 电网技术, 2024, 48(10): 4297-4311.
|
| [14] |
|
| [15] |
|
| [16] |
|
| [17] |
|
| [18] |
|
| [19] |
张伟骏, 李智诚, 陈大玮, 等. 配网侧分布式储能系统的随机优化配置和选址方法[J]. 高压电器, 2023, 59(7): 125-135.
|
| [20] |
|
| [21] |
|
| [22] |
PJM Interconnection. Data miner 2-five minute load forecast[DB/OL]. (2024-07-19)[2024-12-19]. https://dataminer2.pjm.com/feed/very_short_load_frcst.
|
| [23] |
|
/
| 〈 |
|
〉 |
