基于数据驱动和深度置信网络的配电网无功优化
|
摘要
随着分布式电源和随机负荷电动汽车等的大量接入,配电网的运行环境日益复杂,对在线无功优化及其快速性提出了更高的要求。该文将"深度学习"引入配电网无功优化,提出了基于深度置信网络的无功优化方法。通过构造高维随机矩阵,从配电网运行数据中提取统计特征作为输入,将历史控制策略进行编码作为输出,利用先无监督后有监督的方式训练深层架构,学习系统特征与无功优化策略之间的映射关系,建立基于数据驱动和深度置信网络的配电网无功优化模型。基于改造的IEEE-37节点主动配电网仿真模型,对比分析了历史数据量和分布式电源渗透率场景对传统优化方法,场景匹配方法和所提方法的无功优化效果的影响。结果表明,所提方法可明显降低网络损耗和节点电压偏移,它不依赖于系统的模型和参数,在线决策速度快,且对历史数据量要求较低,在高渗透率分布式发电等未知场景下仍能表现优良,验证了该方法的正确性、有效性和较强的鲁棒性。
With a large number of distributed generators(DG) and electrical vehicles integrated into distribution system, operation complexity of the distribution system increased, arising higher requirements to fast online reactive power optimization. In this paper, "deep learning" is introduced, and an online reactive power optimization method based on deep belief network(DBN) is proposed. Firstly, utilizing the electrical and ambient big data from the distribution system operation, high-dimension random matrixes are constructed, 57 statistic features are extracted and used as the inputs of DBN, and conventional reactive power control strategies are used as the outputs of DBN. Secondly, deep architecture is trained with unsupervised and supervised methods to learn the nonlinear mapping between the system features and the reactive power strategies, and an online reactive power optimization model is established based on data driven and DBN. Based on the modified simulation model of IEEE-37 active distribution network, the effects of historical data and scenarios of distributed generation penetration on reactive power optimization results of traditional method, scenario matching method and the proposed method are compared and analyzed. The results show that, the proposed method reduces network losses and node voltage deviations obviously, not depending on distribution system model and parameter, requires less historical data, and possesses excellent performance even under unknown high DG penetration scenes, verifying correctness, effectiveness and robustness of the proposed method.
With a large number of distributed generators(DG) and electrical vehicles integrated into distribution system, operation complexity of the distribution system increased, arising higher requirements to fast online reactive power optimization. In this paper, "deep learning" is introduced, and an online reactive power optimization method based on deep belief network(DBN) is proposed. Firstly, utilizing the electrical and ambient big data from the distribution system operation, high-dimension random matrixes are constructed, 57 statistic features are extracted and used as the inputs of DBN, and conventional reactive power control strategies are used as the outputs of DBN. Secondly, deep architecture is trained with unsupervised and supervised methods to learn the nonlinear mapping between the system features and the reactive power strategies, and an online reactive power optimization model is established based on data driven and DBN. Based on the modified simulation model of IEEE-37 active distribution network, the effects of historical data and scenarios of distributed generation penetration on reactive power optimization results of traditional method, scenario matching method and the proposed method are compared and analyzed. The results show that, the proposed method reduces network losses and node voltage deviations obviously, not depending on distribution system model and parameter, requires less historical data, and possesses excellent performance even under unknown high DG penetration scenes, verifying correctness, effectiveness and robustness of the proposed method.
引文
[1]董海龙.含分布式发电配电网无功优化研究[D].沈阳:沈阳农业大学,2017.
[2]程杉,陈民铀,黄薏宸.含分布式发电的配电网多目标无功优化策略研究[J].电力系统保护与控制,2013,41(10):45-50.Cheng Shan,Chen Minyou,Huang Yichen.Multi-objective reactive power optimization of distribution system penetrated with distributed generation[J].Power System Protection and Control,2013,41(10):45-50(in Chinese).
[3]李惠玲.配电网无功优化与控制研究[D].北京:中国电力科学研究院有限公司,2008.
[4]刘惠姣.基于混合智能算法的配电网无功优化的研究[D].保定:华北电力大学(河北),2009.
[5]王克文,张东岳.电力系统无功优化算法综述[J].电测与仪表,2016,53(10):73-79.Wang Kewen,Zhang Dongyue.A summary of reactive power optimization algorithm in power system[J].Electrical Measurement&Instrumentation,2016,53(10):73-79(in Chinese).
[6]范博文.基于场景分析的含分布式电源的配电网无功优化[D].济南:山东大学,2016.
[7]陈旭,张勇军.主动配电网背景下无功电压控制方法综述[J].电系统自动化,2016,40(1):143-151.Chen Xu,Zhang Yongjun.Review of reactive power and voltage control method in the background of active distribution network[J].Automation of Electric Power Systems,2016,40(1):143-151(in Chinese).
[8]罗欢,陈民铀,程庭莉.含风力发电的配电网自适应分时段无功优化[J].电网技术,2014,38(8):2207-2212.Luo Huan,Chen Minyou,Cheng Tingli.Adaptive time-intervalled reactive power optimization for distribution network containing wind power generation[J].Power System Technology,2014,38(8):2207-2212(in Chinese).
[9]徐俊俊,黄永红,王琪,等.基于自然选择粒子群算法的含DG接入的配电网无功优化[J].电测与仪表,2014,51(10):33-38.Xu Junjun,Huang Yonghong,Wang Qi,et al.Reactive power optimization in distribution network with DG based on natural selection particle swarm optimization[J].Electrical Measurement&Instrumentation,2014,51(10):33-38(in Chinese).
[10]段建东,杨杉.基于改进差分进化法的含双馈型风电场的配电网无功优化[J].电力自动化设备,2013,33(11):123-127.Duan Jiandong,Yang Shan.Reactive power optimization based on modified differential evolution algorithm for power distribution system with DFIG wind farms[J].Electric Power Automation Equipment,2013,33(11):123-127(in Chinese).
[11]Dedinec A,Filiposka S,Dedinec A,et al.Deep belief network based electricity load forecasting:an analysis of Macedonian case[J].Energy,2016,115(3):1688-1700.
[12]陈亮,王震,王刚.深度学习框架下LSTM网络在短期电力负荷预测中的应用[J].电力信息与通信技术,2017,15(5):8-11.Chen Liang,Wang Zhen,Wang Gang.Application of LSTM networks in short-term power load forecasting under the deep learning framework[J].Electric Power Information and Communication Technology,2017,15(5):8-11(in Chinese).
[13]Li Q,Zhang Y,Ji T,et al.Volt/var control for power grids with connections of large-scale wind farms:a review[J].IEEE Access,2018,6:26675-26692.
[14]Gong Jinxia,Xie Da,Zhang Yanchi,et al.Combination of data mining and ant colony algorithm for reactive power optimization[C]//Measuring Technology and Mechatronics Automation(ICMTMA),2011 Third International Conference on.Shangshai,China:IEEE Computer Society,2011.
[15]周后良.基于数据挖掘技术的电网无功优化和降损研究[D].长沙:湖南大学,2017.
[16]Chen X,Yi Y,Zhang Y,et al.Approach to setting gateway reactive power control band for distribution networks with wind power[J].IETGeneration,Transmission&Distribution,2017,11(3):596-604.
[17]Sheng Wanxing,Liu Keyan,Niu Huanna,et al.The anomalous data identification study of reactive power optimization system based on big data[C]//International Conference on Probabilistic Methods Applied to Power Systems.Beijing,China:IEEE,2016.
[18]刘科研,季玉琦,陆凌芝,等.基于负荷分布匹配与熵权法的配电网无功优化[J].电网技术,2017,41(12):3980-3988.Liu Keyan,Ji Yuqi,Lu Lingzhi,et al.Reactive power optimization in distribution network based on load distribution matching and entropy weight method[J].Power System Technology,2017,41(12):3980-3988(in Chinese).
[19]缪书唯,谢开贵,杨贺钧,等.基于数据驱动方法的风电机组功率优化[J].电力系统自动化,2016,40(22):7-14.Miao Shuwei,Xie Kaigui,Yang Hejun,et al.Power optimization of wind turbine generators based on data-driven approach[J].Automation of Electric Power Systems,2016,40(22):7-14(in Chinese).
[20]Schmidhuber.Deep learning in neural networks:an overview[J].Neural Networks,2015,61:85-117.
[21]朱乔木,党杰,陈金富,等.基于深度置信网络的电力系统暂态稳定评估方法[J].中国电机工程学报,2018,38(3):735-743.Zhu Qiaomu,Dang jie,Chen Jinfu,et al.A method for power system transient stability assessment based on deep belief networks[J].Proceedings of the CSEE,2018,38(3):735-743(in Chinese).
[22]胡伟,郑乐,闵勇,等.基于深度学习的电力系统故障后暂态稳定评估研究[J].电网技术,2017,41(10):3140-3146.Hu Wei,Zheng Le,Min Yong,et al.Research on power system transient stability assessment based on deep learning of big data technique[J].Power System Technology,2017,41(10):3140-3146(in Chinese).
[23]Hinton G E.A practical guide to training restricted boltzmann machines[J].Momentum,2012,9(1):599-619.
[24]竹瑞博,安毅,巫健.基于深度学习的电力业务通信带宽需求预测方法[J].电力信息与通信技术,2017,15(9):26-31.Zhu Ruibo,An Yi,Wu Jian.Bandwidth prediction for business requirement of electric power communication network with deep learning[J].Electric Power Information and Communication Technology,2017,15(9):26-31(in Chinese).
[25]刘雷涛.基于机器学习的电力系统暂态稳定评估[D].北京:华北电力大学,2017.
[26]Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets[J].Neural Computation,2014,18(7):1527-1554.
[27]贺兴,艾芊,邱才明,等.随机矩阵理论在电力系统认知中的应用初探[J].电网技术,2017,41(4):1165-1173.He Xing,Ai Qian,Qiu Caiming,et al.A primary study on the situation awareness of power systems using random matrix theory[J].Power System Technology,2017,41(4):1165-1173(in Chinese).
[28]张力,张子仲,顾建炜.基于随机矩阵理论的电网状态分析与扰动定位方法[J].电力系统自动化,2018,42(12):93-99.Zhang Li,Zhang Zizhong,Gu Jianwei.State analysis and disturbance positioning method of power grid based on random matrix theory[J].Automation of Electric Power Systems,2018,42(12):93-99(in Chinese).
[29]徐心怡,贺兴,艾芊,等.基于随机矩阵理论的配电网运行状态相关性分析方法[J].电网技术,2016,40(3):781-790.Xu Xinyi,He Xing,Ai Qian,et al.A correlation analysis method for operation status of distribution network based on random matrix theory[J].Power System Technology,2016,40(3):781-790(in Chinese).
[2]程杉,陈民铀,黄薏宸.含分布式发电的配电网多目标无功优化策略研究[J].电力系统保护与控制,2013,41(10):45-50.Cheng Shan,Chen Minyou,Huang Yichen.Multi-objective reactive power optimization of distribution system penetrated with distributed generation[J].Power System Protection and Control,2013,41(10):45-50(in Chinese).
[3]李惠玲.配电网无功优化与控制研究[D].北京:中国电力科学研究院有限公司,2008.
[4]刘惠姣.基于混合智能算法的配电网无功优化的研究[D].保定:华北电力大学(河北),2009.
[5]王克文,张东岳.电力系统无功优化算法综述[J].电测与仪表,2016,53(10):73-79.Wang Kewen,Zhang Dongyue.A summary of reactive power optimization algorithm in power system[J].Electrical Measurement&Instrumentation,2016,53(10):73-79(in Chinese).
[6]范博文.基于场景分析的含分布式电源的配电网无功优化[D].济南:山东大学,2016.
[7]陈旭,张勇军.主动配电网背景下无功电压控制方法综述[J].电系统自动化,2016,40(1):143-151.Chen Xu,Zhang Yongjun.Review of reactive power and voltage control method in the background of active distribution network[J].Automation of Electric Power Systems,2016,40(1):143-151(in Chinese).
[8]罗欢,陈民铀,程庭莉.含风力发电的配电网自适应分时段无功优化[J].电网技术,2014,38(8):2207-2212.Luo Huan,Chen Minyou,Cheng Tingli.Adaptive time-intervalled reactive power optimization for distribution network containing wind power generation[J].Power System Technology,2014,38(8):2207-2212(in Chinese).
[9]徐俊俊,黄永红,王琪,等.基于自然选择粒子群算法的含DG接入的配电网无功优化[J].电测与仪表,2014,51(10):33-38.Xu Junjun,Huang Yonghong,Wang Qi,et al.Reactive power optimization in distribution network with DG based on natural selection particle swarm optimization[J].Electrical Measurement&Instrumentation,2014,51(10):33-38(in Chinese).
[10]段建东,杨杉.基于改进差分进化法的含双馈型风电场的配电网无功优化[J].电力自动化设备,2013,33(11):123-127.Duan Jiandong,Yang Shan.Reactive power optimization based on modified differential evolution algorithm for power distribution system with DFIG wind farms[J].Electric Power Automation Equipment,2013,33(11):123-127(in Chinese).
[11]Dedinec A,Filiposka S,Dedinec A,et al.Deep belief network based electricity load forecasting:an analysis of Macedonian case[J].Energy,2016,115(3):1688-1700.
[12]陈亮,王震,王刚.深度学习框架下LSTM网络在短期电力负荷预测中的应用[J].电力信息与通信技术,2017,15(5):8-11.Chen Liang,Wang Zhen,Wang Gang.Application of LSTM networks in short-term power load forecasting under the deep learning framework[J].Electric Power Information and Communication Technology,2017,15(5):8-11(in Chinese).
[13]Li Q,Zhang Y,Ji T,et al.Volt/var control for power grids with connections of large-scale wind farms:a review[J].IEEE Access,2018,6:26675-26692.
[14]Gong Jinxia,Xie Da,Zhang Yanchi,et al.Combination of data mining and ant colony algorithm for reactive power optimization[C]//Measuring Technology and Mechatronics Automation(ICMTMA),2011 Third International Conference on.Shangshai,China:IEEE Computer Society,2011.
[15]周后良.基于数据挖掘技术的电网无功优化和降损研究[D].长沙:湖南大学,2017.
[16]Chen X,Yi Y,Zhang Y,et al.Approach to setting gateway reactive power control band for distribution networks with wind power[J].IETGeneration,Transmission&Distribution,2017,11(3):596-604.
[17]Sheng Wanxing,Liu Keyan,Niu Huanna,et al.The anomalous data identification study of reactive power optimization system based on big data[C]//International Conference on Probabilistic Methods Applied to Power Systems.Beijing,China:IEEE,2016.
[18]刘科研,季玉琦,陆凌芝,等.基于负荷分布匹配与熵权法的配电网无功优化[J].电网技术,2017,41(12):3980-3988.Liu Keyan,Ji Yuqi,Lu Lingzhi,et al.Reactive power optimization in distribution network based on load distribution matching and entropy weight method[J].Power System Technology,2017,41(12):3980-3988(in Chinese).
[19]缪书唯,谢开贵,杨贺钧,等.基于数据驱动方法的风电机组功率优化[J].电力系统自动化,2016,40(22):7-14.Miao Shuwei,Xie Kaigui,Yang Hejun,et al.Power optimization of wind turbine generators based on data-driven approach[J].Automation of Electric Power Systems,2016,40(22):7-14(in Chinese).
[20]Schmidhuber.Deep learning in neural networks:an overview[J].Neural Networks,2015,61:85-117.
[21]朱乔木,党杰,陈金富,等.基于深度置信网络的电力系统暂态稳定评估方法[J].中国电机工程学报,2018,38(3):735-743.Zhu Qiaomu,Dang jie,Chen Jinfu,et al.A method for power system transient stability assessment based on deep belief networks[J].Proceedings of the CSEE,2018,38(3):735-743(in Chinese).
[22]胡伟,郑乐,闵勇,等.基于深度学习的电力系统故障后暂态稳定评估研究[J].电网技术,2017,41(10):3140-3146.Hu Wei,Zheng Le,Min Yong,et al.Research on power system transient stability assessment based on deep learning of big data technique[J].Power System Technology,2017,41(10):3140-3146(in Chinese).
[23]Hinton G E.A practical guide to training restricted boltzmann machines[J].Momentum,2012,9(1):599-619.
[24]竹瑞博,安毅,巫健.基于深度学习的电力业务通信带宽需求预测方法[J].电力信息与通信技术,2017,15(9):26-31.Zhu Ruibo,An Yi,Wu Jian.Bandwidth prediction for business requirement of electric power communication network with deep learning[J].Electric Power Information and Communication Technology,2017,15(9):26-31(in Chinese).
[25]刘雷涛.基于机器学习的电力系统暂态稳定评估[D].北京:华北电力大学,2017.
[26]Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets[J].Neural Computation,2014,18(7):1527-1554.
[27]贺兴,艾芊,邱才明,等.随机矩阵理论在电力系统认知中的应用初探[J].电网技术,2017,41(4):1165-1173.He Xing,Ai Qian,Qiu Caiming,et al.A primary study on the situation awareness of power systems using random matrix theory[J].Power System Technology,2017,41(4):1165-1173(in Chinese).
[28]张力,张子仲,顾建炜.基于随机矩阵理论的电网状态分析与扰动定位方法[J].电力系统自动化,2018,42(12):93-99.Zhang Li,Zhang Zizhong,Gu Jianwei.State analysis and disturbance positioning method of power grid based on random matrix theory[J].Automation of Electric Power Systems,2018,42(12):93-99(in Chinese).
[29]徐心怡,贺兴,艾芊,等.基于随机矩阵理论的配电网运行状态相关性分析方法[J].电网技术,2016,40(3):781-790.Xu Xinyi,He Xing,Ai Qian,et al.A correlation analysis method for operation status of distribution network based on random matrix theory[J].Power System Technology,2016,40(3):781-790(in Chinese).