深度学习辅助约束辨识的电力市场快速出清方法

doi:10.11930/j.issn.1004-9649.202005028

中国电力 ›› 2020, Vol. 53 ›› Issue (9): 90-97,207.DOI: 10.11930/j.issn.1004-9649.202005028

• 电力现货市场运行分析与机制设计专栏 • 上一篇下一篇

深度学习辅助约束辨识的电力市场快速出清方法

吴云亮¹, 张建新¹, 李豹¹, 李鹏¹, 李智勇¹, 周鑫¹, 杨燕², 赖晓文²

1. 中国南方电网电力调度控制中心，广东广州 510663;
2. 北京清能互联科技有限公司，北京 100084

收稿日期:2020-05-06 修回日期:2020-08-21 发布日期:2020-09-09
作者简介:吴云亮(1984—)，男，博士，高级工程师，从事电力系统运行与控制研究，E-mail：wuyunliang@csg.cn;杨燕(1993—)，女，通信作者，博士研究生，从事人工智能在电力系统中的应用研究，E-mail：2275881834@qq.com;赖晓文(1988—)，男，博士，从事电力调度优化和电力市场的研究，E-mail：laixw@tsintergy.com
基金资助:
中国南方电网有限责任公司科技项目(ZDKJXM20180070)

A Fast Power Market Clearing Method Based on Active Constraints Identification by Deep Learning

WU Yunliang¹, ZHANG Jianxin¹, LI Bao¹, LI Peng¹, LI Zhiyong¹, ZHOU Xin¹, YANG Yan², LAI Xiaowen²

1. CSG Power Dispatching Control Center, Guangzhou 510663, China;
2. Beijing Tsintergy Technology Co., Ltd., Beijing 100084, China

Received:2020-05-06 Revised:2020-08-21 Published:2020-09-09
Supported by:
This work is supported by Science and Technology Project of CSG (No.ZDKJXM20180070)

摘要/Abstract

摘要： 日前电力市场出清需要求解大规模安全约束经济调度问题，尽管实际采用线化处理方法，但需要考虑N－1场景下的大量安全约束，导致其规模庞大，难以快速求解。提出了一种深度学习辅助的日前市场快速出清方法，以满足快速出清计算场合的应用需求。首先，设计基于深度神经网络的安全约束经济调度模型计算框架，将深度学习技术应用于日前电力市场出清计算过程，兼顾安全约束经济调度模型的求解速度和求解精度；其次，提出面向起作用约束辨识的深度学习策略，从特征向量、深度神经网络结果处理2个方面，实现安全约束经济调度起作用约束集的辨识，并将其纳入日前市场出清模型，以简化模型的复杂度；最后，在接入新能源的IEEE 30标准测试系统中验证了所述方法的有效性。利用深度神经网络预辨识安全约束经济调度模型的起作用约束，有利于降低模型复杂度，提高日前市场出清的计算效率。

关键词: 安全约束经济调度, 约束辨识, 深度学习, 日前市场出清

Abstract: The day-ahead power market clearing needs to solve the security-constrained economic dispatch (SCED) problem. Although the SCED problem is a linear programming (LP) model, the model size is too large to be effectively solved because the massive security constraints in the N–1 scenarios need to be considered. Therefore, this paper proposes a fast clearing method for day-ahead power market based on the deep neural network. Firstly, a computation framework for SCED model based on deep neural network is designed, and embeds deep learning technology into the existing day-ahead power market clearing framework, which can effectively improve the solving speed of the SCED model without compromising precision. Secondly, a deep learning strategy is proposed for identification of active constraint sets, which can provide technical support for deep neural networks to effectively identify the active constraints of SCED from two aspects: feature vector design and efficient processing of the results of deep neural network. Finally, the effectiveness of the proposed method is verified in the IEEE 30 standard test system with renewable energy sources. The deep neural network is used to pre-identify the active constraints of the SCED model, which is beneficial to reduce the complexity of the model and improve the calculation efficiency of market clearing.

Key words: security-constrained economic dispatch, active constraints identification, deep learning technology, day-ahead power market clearing

吴云亮, 张建新, 李豹, 李鹏, 李智勇, 周鑫, 杨燕, 赖晓文. 深度学习辅助约束辨识的电力市场快速出清方法[J]. 中国电力, 2020, 53(9): 90-97,207.

WU Yunliang, ZHANG Jianxin, LI Bao, LI Peng, LI Zhiyong, ZHOU Xin, YANG Yan, LAI Xiaowen. A Fast Power Market Clearing Method Based on Active Constraints Identification by Deep Learning[J]. Electric Power, 2020, 53(9): 90-97,207.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202005028

https://www.electricpower.com.cn/CN/Y2020/V53/I9/90

参考文献

[1] 陈政, 张翔, 马子明, 等. 引导电力供需长期有效均衡的容量市场设计[J]. 中国电力, 2020, 53(8): 164–172.
CHEN Zheng, ZHANG Xiang, MA Ziming, et al. A capacity market design guiding the long-term effective balance of power supply and demand[J]. Electric Power, 2020, 53(8): 164–172.
[2] 房欣欣, 杨知方, 余娟, 等. 节点电价的理论剖析与拓展[J]. 中国电机工程学报, 2020, 40(2): 379–390
FANG Xinxin, YANG Zhifang, YU Juan, et al. Theoretical analysis and extension of locational marginal price[J]. Proceedings of the CSEE, 2020, 40(2): 379–390
[3] CAPITANESCU F, GLAVIC M, ERNST D, et al. Contingency filtering techniques for preventive security-constrained optimal power flow[J]. IEEE Transactions on Power Systems, 2007, 22(4): 1690–1697.
[4] ALLEN J W, BRUCE F W. Power generation operation and control[M]. 2nd edition. Beijing: Tsinghua University Press, 1996.
[5] BRANDWAJN V. Efficient bounding method for linear contingency analysis[J]. IEEE Transactions on Power Systems, 1988, 3(1): 38–43.
[6] MADANI R, LAVAEI J, BALDICK R. Constraint screening for security analysis of power networks[J]. IEEE Transactions on Power Systems, 2017, 32(3): 1828–1838.
[7] ZHAI Q Z, GUAN X H, CHENG J H, et al. Fast identification of inactive security constraints in SCUC problems[J]. IEEE Transactions on Power Systems, 2010, 25(4): 1946–1954.
[8] YANG Y F, GUAN X H, ZHAI Q Z. Fast grid security assessment with N-k contingencies[J]. IEEE Transactions on Power Systems, 2017, 32(3): 2193–2203.
[9] ARDAKANI A J, BOUFFARD F. Identification of umbrella constraints in DC-based security-constrained optimal power flow[J]. IEEE Transactions on Power Systems, 2013, 28(4): 3924–3934.
[10] HUA B W, BIE Z H, LIU C, et al. Eliminating redundant line flow constraints in composite system reliability evaluation[J]. IEEE Transactions on Power Systems, 2013, 28(3): 3490–3498.
[11] TEJADA-ARANGO D A, SANCHEZ-MARTIN P, RAMOS A. Security constrained unit commitment using line outage distribution factors[J]. IEEE Transactions on Power Systems, 2018, 33(1): 329–337.
[12] XAVIER Á S, QIU F, WANG F Y, et al. Transmission constraint filtering in large-scale security-constrained unit commitment[J]. IEEE Transactions on Power Systems, 2019, 34(3): 2457–2460.
[13] XAVIER Á S, QIU F, AHMED S. Learning to solve large-scale security-constrained unit commitment problems[EB/OL]. (2019-12-28) [2020-04-29]. https://arxiv.org/abs/1902.01697.
[14] NG Y, MISRA S, ROALD L A, et al. Statistical learning for DC optimal power flow[C]//2018 Power Systems Computation Conference (PSCC). Dublin, Ireland. IEEE, 2018: 1–7.
[15] DEKA D, MISRA S. Learning for DC -OPF: Classifying active sets using neural nets. [EB/OL]. (2019-02-14)[2020-04-29]. https://arxiv.org/abs/1902.05607.
[16] 杨晓楠, 孙博, 郎燕生. 基于深度学习的特高压直流闭锁故障智能调度决策[J]. 中国电力, 2020, 53(6): 8–17
YANG Xiaonan, SUN Bo, LANG Yansheng. Intelligent dispatch decision-making for UHVDC blocking fault based on deep learning[J]. Electric Power, 2020, 53(6): 8–17
[17] SHAHAM U, CLONINGER A, COIFMAN R R. Provable approximation properties for deep neural networks[J]. Applied and Computational Harmonic Analysis, 2018, 44(3): 537–557.
[18] 邓韦斯, 吴云亮, 孙宇军, 等. 面向现货市场出清的发电计划校正决策方法[J]. 电力系统自动化: 1–14[2020-08-20].http://kns.cnki.net/kcms/detail/32.1180.TP.20200807.1442.002.html.
DENG Weisi, WU Yunliang, SUN Yujun, et al. Decision-making method of generation scheduling correction oriented to electricity spot market clearing[J]. Automation of Electric Power Systems: 1–14[2020-08-20].http://kns.cnki.net/kcms/detail/32.1180.TP.20200807.1442.002.html.
[19] 吴云亮, 李豹, 罗会洪, 等. 面向现货市场出清的条件断面约束模型化处理方法[J]. 电网技术, 2020, 44(8): 2819–2831
WU Yunliang, LI Bao, LUO Huihong, et al. Approach for conditional section constraints modeling based on spot market clearing[J]. Power System Technology, 2020, 44(8): 2819–2831
[20] 余娟, 杨燕, 杨知方, 等. 基于深度学习的概率能量流快速计算方法[J]. 中国电机工程学报, 2019, 39(1): 22–30, 317
YU Juan, YANG Yan, YANG Zhifang, et al. Fast probabilistic energy flow analysis based on deep learning[J]. Proceedings of the CSEE, 2019, 39(1): 22–30, 317
[21] TIELEMAN T, HINTON G. Rmsprop: Divide the gradient by a running average of its recent magnitude[EB/OL]. (2012-10-11)[2020-04-29]. https://amara.org/en/videos/vrXNiLBHyW92/en/180511/.
[22] QIN Z L, LI W Y, XIONG X F. Incorporating multiple correlations among wind speeds, photovoltaic Powers and bus loads in composite system reliability evaluation[J]. Applied Energy, 2013, 110: 285–294.
[23] RASKUTTI G, WAINWRIGHT M J, YU B. Early stopping and non-parametric regression: an optimal data-dependent stopping rule[J]. Journal of Machine Learning Research, 2014, 15(1): 335–366.

深度学习辅助约束辨识的电力市场快速出清方法

A Fast Power Market Clearing Method Based on Active Constraints Identification by Deep Learning

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 14

编辑推荐

Metrics

[1]	姜通海, 王峰, 刘子琪, 单帅杰. 基于改进生成对抗网络的风光气象资源联合场景生成方法[J]. 中国电力, 2025, 58(3): 183-192.
[2]	刘东, 张越, 皮俊波, 单连飞, 刘赫, 宋鹏程, 姜涛. 面向电网调控信息智能检索的知识图谱构建及应用[J]. 中国电力, 2023, 56(7): 78-84.
[3]	张鑫, 叶俊杰, 崔瑶, 黄鑫, 仲林林. 基于语义信息距离解耦的变电运维多类别缺陷图像检测[J]. 中国电力, 2023, 56(6): 209-218.
[4]	周震震, 宋云海, 何宇浩, 王黎伟, 黄和燕, 何珏, 朱志航, 闫云凤. 基于分组查询注意力的可扩展电力人员行为分类方法[J]. 中国电力, 2023, 56(11): 77-85.
[5]	李丰君, 王磊, 赵健, 张建宾, 张世尧, 田杨阳. 基于天气融合和LSTM网络的分布式光伏短期功率预测方法[J]. 中国电力, 2022, 55(11): 149-154.
[6]	赵永良, 付鑫, 郭阳, 边迎迎, 王思宁. 基于深度学习和图像识别的电力配件智能出入库[J]. 中国电力, 2021, 54(3): 55-60.
[7]	马静怡, 崔昊杨, 张明达, 孙益辉, 许永鹏. 基于改进Faster RCNN的小尺度入侵目标识别及定位[J]. 中国电力, 2021, 54(3): 38-44.
[8]	周俊煌, 黄廷城, 谢小瑜, 范纹郡, 易婷婷, 张勇军. 视频图像智能识别技术在输变电系统中的应用研究综述[J]. 中国电力, 2021, 54(1): 124-134,166.
[9]	王志成, 王秀丽, 王帅. 考虑实时市场平衡费用的含风电日前市场电能-备用联合出清模型[J]. 中国电力, 2020, 53(9): 19-27.
[10]	赵会茹, 赵一航, 郭森. 基于互补集合经验模态分解和长短期记忆神经网络的短期电力负荷预测[J]. 中国电力, 2020, 53(6): 48-55.
[11]	杨晓楠, 孙博, 郎燕生. 基于深度学习的特高压直流闭锁故障智能调度决策[J]. 中国电力, 2020, 53(6): 8-17.
[12]	周艳真, 查显煜, 兰健, 郭庆来, 孙宏斌, 薛峰, 王胜明. 基于数据增强和深度残差网络的电力系统暂态稳定预测[J]. 中国电力, 2020, 53(1): 22-31.
[13]	赖秋频, 杨军, 谭本东, 王亮, 傅思遥, 韩立伟. 基于YOLOv2网络的绝缘子自动识别与缺陷诊断模型[J]. 中国电力, 2019, 52(7): 31-39.
[14]	郭敬东, 陈彬, 王仁书, 王佳宇, 仲林林. 基于YOLO的无人机电力线路杆塔巡检图像实时检测[J]. 中国电力, 2019, 52(7): 17-23.

模态框（Modal）标题

深度学习辅助约束辨识的电力市场快速出清方法

A Fast Power Market Clearing Method Based on Active Constraints Identification by Deep Learning

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 14

编辑推荐

Metrics