摘要
主动配电网通过市场机制来引导群体性的主动需求行为,电网负荷特性随之发生改变,降低了传统负荷预测技术的预测精度。考虑主动配电网在传统负荷的基础上引入响应负荷的计算分析,提出适用于主动配电网的组合预测方法。分析用户响应特性及规律,考虑影响用户响应度的因素并将其线性参数化,建立基于响应度评估模型的响应负荷预测方法;将传统负荷分解成季节性基荷和残差,基荷采用相似日负荷预测技术进行预测,残差利用集成神经网络模型进行预测,建立改进神经网络组合预测模型。利用时变响应模型模拟主动配电网负荷数据集来进行仿真验证,与其他负荷预测技术进行对比,实验结果证明了所提负荷预测方法的有效性。
The active distribution network guides the people's demand response behavior through the market mechanism, and the load characteristics of the power grid are changed, which in turn reduces the prediction accuracy of the traditional load forecasting technology. Considering the calculation and analysis of the response load based on the traditional load, a combined forecasting method for active distribution network is proposed. Analyzing the characteristics and patterns of user response, and considering the factors that affect user response and parameterizing them linearly, a response load forecasting method based on a responsiveness assessment model is established. The traditional load is decomposed into seasonal base load and the residual load. The base load is predicted by similar daily load forecasting technology, and the residuals are predicted by integrated neural network model. An improved neural network combination forecasting model is established. The time-varying response model is used to simulate the active distribution network load data set for simulation verification. Compared with other load forecasting technologies, the experimental results prove the validity of the proposed load forecasting method.
引文
[1] 范明天,张祖平,苏傲雪,等(Fan Mingtian, Zhang Zuping, Su Aoxue, et al.).主动配电系统可行技术的研究(Enabling technologies for active distribution systems)[J].中国电机工程学报(Proceedings of the CSEE),2013,33(22):12-18.
[2] 赵波,王财胜,周金辉,等(Zhao Bo, Wang Caisheng, Zhou Jinhui, et al.).主动配电网现状与未来发展(Present and future development trend of active distribution network)[J].电力系统自动化(Automation of Electric Power Systems),2014,38(18):125-135.
[3] 潘超,孟涛,蔡国伟,等(Pan Chao, Meng Tao, Cai Guowei, et al.).主动配电网广义电源多目标优化规划(Multi-objective optimization planning of generalized power in active distribution network)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2016,35(3):41-46.
[4] 张建华,曾博,张玉莹,等(Zhang Jianhua, Zeng Bo, Zhang Yuying, et al.).主动配电网规划关键问题与研究展望(Key issues and research prospects of active distribution network planning)[J].电工技术学报(Transactions of China Electrotechnical Society),2014,28(2):13-23.
[5] 张艺渊,江岳文(Zhang Yiyuan, Jiang Yuewen).多种需求响应和日前小时电价优化促进风电接纳研究(Research on promoting wind power accommodation with multi-type demand response and day-ahead hourly price optimization)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2018,37(1):57-65.
[6] 汤奕,鲁针针,伏祥运(Tang Yi, Lu Zhenzhen, Fu Xiangyun).居民主动负荷促进分布式电源消纳的需求响应策略(Demand response strategies for promoting consumption of distributed power generation with residential active loads)[J].电力系统自动化(Automation of Electric Power Systems),2015,39(24): 49-55.
[7] 杨丽君,曹玉洁,梁景志,等(Yang Lijun, Cao Yujie, Liang Jingzhi, et al.).基于博弈思想的需求响应视角下的主动配电网故障恢复(Active distribution network fault restoration based on game theory in demand response perspective)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2018,37(4):1-9.
[8] 苏小林,刘孝杰,阎晓霞,等(Su Xiaolin, Liu Xiaojie, Yan Xiaoxia, et al.).计及需求响应的主动配电网短期负荷预测(Short-term load forecast of active distribution network based on demand response)[J].电力系统自动化(Automation of Electric Power Systems),2018,42(10):60-66.
[9] 雷涛,吕勇,马淑慧(Lei Tao, Lv Yong, Ma Shuhui). 基于改进灰色理论的主动配电网中长期负荷预测(Mid-long term load forecasting of active distribution network based on improved grey theory)[J].电网与清洁能源(Power System and Clean Energy),2016,32(9):22-28.
[10] 于道林,张智晟,韩少晓,等(Yu Daolin, Zhang Zhisheng, Han Shaoxiao, et al.).计及需求响应的Elman-NN短期负荷预测模型研究(Study of short-term load forecasting model based on Elman-NN considering demand response)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2017,36(4):59-65.
[11] 刘会家,管鑫,陈波,等(Liu Huijia, Guan Xin, Chen Bo, et al.).考虑主动需求的主动配电网负荷预测(Load forecasting for active distribution network in the presence of active demand)[J].电力系统保护与控制(Power System Protection and Control),2018,46(10):68-74.
[12] 钟清,孙闻,余南华,等(Zhong Qing, Sun Wen, Yu Nanhua, et al.).主动配电网规划中的负荷预测与发电预测(Load and power forecasting in active distribution network planning)[J].中国电机工程学报(Proceedings of the CSEE),2014,34(19):3050-3056.
[13] Garulli A, Paoletti S, Vicino A. Models and techniques for electric load forecasting in the presence of demand response[J]. IEEE Transactions on Control Systems Technology, 2015, 23(3): 1087-1097.
[14] 刘孝杰,苏小林,阎晓霞,等(Liu Xiaojie, Su Xiaolin, Yan Xiaoxia, et al.).面向主动响应和售电市场的主动配电系统负荷预测(Load forecast of active distribution system based on active response and electricity market)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2017,29(2):121-128.
[15] 钟清,张文峰,周佳威,等(Zhong Qing, Zhang Wenfeng, Zhou Jiawei, et al.).主动配电网分层分布控制策略及实现(Hierarchical and distributed control strategy for active distribution network & its implementation)[J].电网技术(Power System Technology),2015,39(6):1511-1517.
[16] 尤毅,刘东,钟清,等(You Yi, Liu Dong, Zhong Qing, et al.).主动配电网优化调度策略研究(Research on optimal schedule strategy for active distribution network)[J].电力系统自动化(Automation of Electric Power Systems),2014,38(9):177-183.
[17] Paoletti S, Casini M, Giannitrapani A, et al. Load forecasting for active distribution networks[A]. 2011 2nd IEEE PES International Conference and Exhibition on Innovative Smart Grid Technologies[C]. 2011. 1-6.
[18] 李滨,黄佳,吴茵,等(Li Bin, Huang Jia, Wu Yin, et al.).基于分形特性修正气象相似日的节假日短期负荷预测方法(Holiday short-term load forecasting based on fractal characteristic modified meteorological similar day)[J].电网技术(Power System Technology),2017,41(6):1949-1955.
[19] 安晨帆,杜志叶,李慧慧,等(An Chenfan, Du Zhiye, Li Huihui, et al.).基于组合赋权和BP神经网络的500kV交流输电线路电磁环境评估方法研究(Study of 500kV AC transmission line electromagnetic environment evaluation method based on combination empowerment and BP neural network)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2016,35(3):62-68.
[20] 马天男,牛东晓,黄雅莉,等(Ma Tiannan, Niu Dongxiao, Huang Yali, et al.).基于Spark平台和多变量L2-Boosting回归模型的分布式能源系统短期负荷预测(Short-term load forecasting for distributed energy system based on Spark platform and multi-variable L2-Boosting regression model)[J].电网技术(Power System Technology),2016,40(6):1642-1649.