基于Bagging-组合核函数相关向量机的短期负荷预测模型研究*

doi:10.11985/2023.02.013

摘要/Abstract

摘要：

为充分发挥组合核函数在相关向量机预测模型中的优势，有效提高负荷预测的精度，提出基于Bagging-组合核函数相关向量机的短期负荷预测模型。首先构造了高斯核函数与Morlet小波核函数加权组合的组合核函数相关向量机的预测模型，然后采用粒子群算法对两个核函数的最优权值进行优选。为提高模型的泛化能力，采用Bagging算法对原始数据多次抽样构造训练样本集。通过实际算例仿真，与多种相关向量机预测模型对比分析，验证了该模型具有较好的预测精度。

Abstract:

In order to give full play to the advantages of the combined kernel function in the relevance vector machine forecasting model and effectively improve the accuracy of load forecasting, a short-term load forecasting model based on the Bagging-combined kernel function correlation vector machine is proposed. Firstly, the forecasting model of combined kernel function relevance vector machine is constructed by weighted combination of Gaussian kernel function and Morlet wavelet kernel function, and then the particle swarm optimization algorithm is used to optimize the optimal weights of the two kernel functions. In order to improve the generalization ability of the model, the Bagging algorithm is used to sample the original data multiple times to construct a training sample set. Through the simulation of actual example, compared with a variety of relevance vector machine forecasting models, it is verified that the proposed model has good prediction accuracy.

Key words: Short-term load forecasting, relevance vector machine, combinatorial kernel function, Bagging algorithm, Morlet wavelet kernel function

中图分类号:

TM715

邱山, 龚文杰, 张智晟. 基于Bagging-组合核函数相关向量机的短期负荷预测模型研究^*[J]. 电气工程学报, 2023, 18(2): 142-148.

QIU Shan, GONG Wenjie, ZHANG Zhisheng. Research on Short-term Load Forecasting Model Based on Bagging-combined Kernel Function Relevance Vector Machine[J]. Journal of Electrical Engineering, 2023, 18(2): 142-148.

图/表 7

图1

图2

图3

表1

表2

表3

表4

参考文献 23

[1]	LIU Yahui, LUO Huan, ZHAO Bing, et al. Short-term power load forecasting based on clustering and XGBoost method[C]// 2018 IEEE 9th International Conference on Software Engineering and Service Science (ICSESS), November 23-25,2018, Beijing,China. Piscataway:IEEE, 2018:536-539.
[2]	李丹, 张远航, 杨保华, 等. 基于约束并行LSTM分位数回归的短期电力负荷概率预测方法[J]. 电网技术, 2021, 45(4):1356-1364.
	LI Dan, ZHANG Yuanhang, YANG Baohua, et al. Short time power load probabilistic forecasting based on constrained parallel-LSTM neural network quantile regression mode[J]. Power System Technology, 2021, 45(4):1356-1364.
[3]	程宇也. 基于人工神经网络的短期电力负荷预测研究[D]. 杭州: 浙江大学, 2017.
	CHENG Yuye. Short-term load forecasting based on neural network[D]. Hangzhou: Zhejiang University, 2017.
[4]	MARTENS D, BAESENS B, VAN G T. Decompositional rule extraction from support vector machines by active learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2009, 21(2):178-191. doi: 10.1109/TKDE.69
[5]	李元诚, 方廷健, 于尔铿. 短期负荷预测的支持向量机方法研究[J]. 中国电机工程学报, 2003, 23(6):55-59.
	LI Yuancheng, FANG Tingjian, YU Erkeng. Study of support vector machines for short-term load forecasting[J]. Proceedings of the CSEE, 2003, 23(6):55-59.
[6]	YU Weimiao, DU Tiehua, LIM K. Comparison of the support vector machine and relevant vector machine in regression and classification problems[C]// 8th International Conference on Control,Automation,Robotics and Vision, December 6-9,2004,Kunming,China. IEEE, 2004, 2:1309-1314.
[7]	何晔, 邹晓松, 李卓, 等. 一种面向运行可靠性的短期负荷预测方法研究[J]. 电测与仪表, 2019, 56(10):93-98.
	HE Ye, ZOU Xiaosong, LI Zhuo, et al. A short-term load forecasting method oriented for operational reliability[J]. Electrical Measurement & Instrumentation, 2019, 56(10):93-98.
[8]	段青, 赵建国, 马艳. 优化组合核函数相关向量机电力负荷预测模型[J]. 电机与控制学报, 2010, 14(6):33-38.
	DUAN Qing, ZHAO Jianguo, MA Yan. Relevance vector machine based on particle swarm optimization of compounding kernels in electricity load forecasting[J]. Electric Machines and Control, 2010, 14(6):33-38.
[9]	谭吉玉, 刘高常. 基于小波支持向量机模型的矿区生态安全评价方法研究[J]. 黄金科学技术, 2020, 28(6):902-909.
	TAN Jiyu, LIU Gaochang. Ecological security evaluation of mining area based on WSVM[J]. Gold Science and Technology, 2020, 28(6):902-909.
[10]	TAO Dacheng, TANG Xiaoou, LI Xuelong, et al. Asymmetric bagging and random subspace for support vector machines-based relevance feedback in image retrieval[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(7):1088-1099. pmid: 16792098
[11]	杨树仁, 沈洪远. 基于相关向量机的机器学习算法研究与应用[J]. 计算技术与自动化, 2010, 29(1):43-47.
	YANG Shuren, SHEN Hongyuan. Research and application of machine learning algorithm based on relevance vector machine[J]. Computing Technology and Automation, 2010, 29(1):43-47.
[12]	DEMIR B, ERTURK S. Hyperspectral image classification using relevance vector machines[J]. IEEE Geoscience and Remote Sensing Letters, 2007, 4(4):586-590. doi: 10.1109/LGRS.2007.903069
[13]	黄帅栋, 卫志农, 高宗和, 等. 基于非负矩阵分解的相关向量机短期负荷预测模型[J]. 电力系统自动化, 2012, 36(11):62-66.
	HUANG Shuaidong, WEI Zhinong, GAO Zonghe, et al. A short-term load forecasting model based on relevance vector machine with nonnegative matrix factorization[J]. Automation of Electric Power Systems, 2012, 36(11):62-66.
[14]	杨柳, 张磊, 张少勋, 等. 单核和多核相关向量机的比较研究[J]. 计算机工程, 2010, 36(12):195-197. doi: 10.3969/j.issn.1000-3428.2010.12.067
	YANG Liu, ZHANG Lei, ZHANG Shaoxun, et al. Comparison research of single kernel and multi-kernel relevance vector machine[J]. Computer Engineering, 2010, 36(12):195-197. doi: 10.3969/j.issn.1000-3428.2010.12.067
[15]	张朝龙, 何怡刚, 杜博伦, 等. 一种基于多核相关向量机的模拟电路故障预测方法[J]. 电子测量与仪器学报, 2019, 33(11):95-101.
	ZHANG Chaolong, HE Yigang, DU Bolun, et al. Analog circuit fault prognostic approach based on MKRVM[J]. Journal of Electronic Measurement and Instrument, 2019, 33(11):95-101.
[16]	雷亚国, 陈吴, 李乃鹏, 等. 自适应多核组合相关向量机预测方法及其在机械设备剩余寿命预测中的应用[J]. 机械工程学报, 2016, 52(1):87-93. doi: 10.3901/JME.2016.01.087
	LEI Yaguo, CHEN Wu, LI Naipeng, et al. A relevance vector machine prediction method based on adaptive multi-kernel combination and its application to remaining useful life prediction of machinery[J]. Journal of Mechanical Engineering, 2016, 52(1):87-93. doi: 10.3901/JME.2016.01.087
[17]	郑永康, 陈维荣, 戴朝华, 等. Gaussian小波SVM及其混沌时间序列预测[J]. 控制工程, 2009, 16(4):468-471.
	ZHENG Yongkang, CHEN Weirong, DAI Chaohua, et al. Gaussian wavelet SVM and its application to chaotic time series forecasting[J]. Control Engineering of China, 2009, 16(4):468-471.
[18]	TOLAMBIYA A, KALRA P. WSVM with morlet wavelet kernel for image compression[C]// 2007 IEEE International Conference on System of Systems Engineering, April 16-18,2007,San Antonio,TX,USA. Piscataway:IEEE, 2007:1-5.
[19]	武方方, 赵银亮. 一种基于Morlet小波核的约简支持向量机[J]. 控制与决策, 2006(8):848-852,856.
	WU Fangfang, ZHAO Yinliang. Novel reduced support vector machine on Morlet wavelet kernel function[J]. Control and Decision, 2006(8):848-852,856.
[20]	毕云帆, 撖奥洋, 张智晟, 等. 基于模糊Bagging-GBDT的短期负荷预测模型研究[J]. 电力系统及其自动化学报, 2019, 31(7):51-56.
	BI Yunfan, HAN Aoyang, ZHANG Zhisheng, et al. Study on short-term load forecasting model based on fuzzy Bagging-GBDT[J]. Proceedings of the CSU-EPSA, 2019, 31(7):51-56.
[21]	孔祥玉, 郑锋, 鄂志君, 等. 基于深度信念网络的短期负荷预测方法[J]. 电力系统自动化, 2018, 42(5):133-139.
	KONG Xiangyu, ZHENG Feng, E Zhijun, et al. Short-term load forecasting based on deep belief network[J]. Automation of Electric Power Systems, 2018, 42(5):133-139.
[22]	孔祥玉, 李闯, 郑锋, 等. 基于经验模态分解与特征相关分析的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(5):46-52.
	KONG Xiangyu, LI Chuang, ZHENG Feng, et al. Short-term load forecasting method based on empirical mode decomposition and feature correlation analysis[J]. Automation of Electric Power Systems, 2019, 43(5):46-52.
[23]	王继东, 宋智林, 冉冉. 基于改进支持向量机算法的光伏发电短期功率滚动预测[J]. 电力系统及其自动化学报, 2016, 28(11):9-13.
	WANG Jidong, SONG Zhilin, RAN Ran. Short-term photovoltaic power generation rolling forecast based on optimized SVM[J]. Proceedings of the CSU-EPSA, 2016, 28(11):9-13.

预测模型	E_MAPE(%)	E_max(%)	RMSE/MW
模型1	1.52	4.27	55.440 5
模型2	1.67	7.48	59.368 0
模型3	1.18	3.96	34.293 1
模型4	1.41	4.45	38.190 7
模型5	1.01	3.52	32.089 1

时间	模型1		模型2		模型3		模型4		模型5
时间	E_MAPE(%)	E_max(%)	E_MAPE(%)	E_max(%)	E_MAPE(%)	E_max(%)	E_MAPE(%)	E_max(%)	E_MAPE(%)	E_max(%)
周一	3.19	7.76	2.32	6.80	2.12	7.52	2.24	5.00	2.07	6.15
周二	2.84	5.16	2.89	7.71	2.36	7.16	2.51	4.45	1.92	3.55
周三	1.49	5.15	2.69	6.51	2.01	7.08	2.61	6.51	1.68	6.12
周四	1.52	4.27	1.67	7.48	1.18	3.96	1.45	4.89	1.01	3.52
周五	1.65	4.06	3.82	7.19	1.10	3.92	2.08	4.22	1.00	3.89
周六	2.73	7.56	3.62	8.15	2.73	7.79	2.32	5.90	1.98	6.87
周日	2.97	8.59	3.05	8.20	2.71	7.88	2.87	6.45	2.50	5.75
平均值	2.34	6.07	2.86	7.42	2.03	6.47	2.29	5.34	1.73	5.12

时间	模型5
时间	E_MAPE(%)	E_max(%)	RMSE/MW
周一	2.17	4.22	63.815 9
周二	2.89	5.39	75.512 2
周三	2.44	5.95	63.000 1
周四	2.47	4.37	65.576 4
周五	2.56	5.37	59.214 2
周六	2.89	6.69	71.786 1
周日	2.76	6.87	70.422 7
平均值	2.59	5.55	67.046 5

时间	模型5
时间	E_MAPE(%)	E_max(%)	RMSE/MW
周一	1.34	5.89	47.625 6
周二	1.63	5.33	47.779 1
周三	1.79	4.93	52.546 1
周四	1.45	5.17	41.663 8
周五	1.83	4.24	50.637 6
周六	1.89	6.48	52.998 0
周日	2.22	6.87	55.945 1
平均值	1.73	5.54	49.885 0