基于多模型融合的CNN-LSTM-XGBoost短期电力负荷预测方法

doi:10.11930/j.issn.1004-9649.202004026

中国电力 ›› 2021, Vol. 54 ›› Issue (5): 46-55.DOI: 10.11930/j.issn.1004-9649.202004026

• 国家“十三五”智能电网重大专项专栏：（七）电力信息通信新技术在能源互联网中的研究与应用专栏 • 上一篇下一篇

基于多模型融合的CNN-LSTM-XGBoost短期电力负荷预测方法

庄家懿¹, 杨国华^1,2, 郑豪丰¹, 张鸿皓¹

1. 宁夏大学物理与电子电气工程学院，宁夏银川 750021;
2. 宁夏电力能源安全重点实验室，宁夏银川 750004

收稿日期:2020-04-05 修回日期:2020-10-30 发布日期:2021-05-05
作者简介:庄家懿(1996-),男,硕士研究生,从事电力系统负荷预测研究,E-,mail:zhjy_ha@126.com;杨国华(1972-),男,通信作者,硕士,教授,从事新能源电力系统自动化技术研究,Email:ygh@nxu.edu.cn
基金资助:
国家自然科学基金资助项目(61763040，71263043)

Short-Term Load Forecasting Method Based on Multi-model Fusion Using CNN-LSTM-XGBoost Framework

ZHUANG Jiayi¹, YANG Guohua^1,2, ZHENG Haofeng¹, ZHANG Honghao¹

1. School of Physics and Electronic-Electrical Engineering, Ningxia University, Yinchuan 750021, China;
2. Ningxia Key Laboratory of Electrical Energy Security, Yinchuan 750004, China

Received:2020-04-05 Revised:2020-10-30 Published:2021-05-05
Supported by:
This work is supported by National Natural Science Foundation of China (No.61763040, No.71263043)

摘要/Abstract

摘要： 短期电力负荷的精准预测可以有效指导机组组合调度、经济调度与电力市场运营。针对输入数据特征量受限时负荷预测的低精度问题，提出一种基于多模型融合的CNN-LSTM-XGBoost短期电力负荷预测方法。通过建立融合局部特征预提取模块的LSTM（long short term memory）网络结构，并将其与XGBoost（eXtreme boosting system）预测模型并行结合，之后结合MAPE-RW（mean absolute percentage error-reciprocal weight）算法进行模型融合初始权重设置，对最佳权重进行搜索，构建最佳融合模型。通过运用电力负荷数据对所提方法进行预测实验，结果表明CNN-LSTM- XGBoost模型的MAPE（mean absolute percentage error）与RMSE（root mean square error）分别为0.377%与148.419 MW，相比于单一网络模型与融合模型结构实现了误差指标的显著降低，验证了基于多模型融合的CNN-LSTM-XGBoost短期电力负荷预测方法具有较快的模型训练速度、较高的预测准确度与较低的预测误差。

关键词: 短期负荷预测, 局部特征预提取, LSTM, XGBoost, 多模型融合

Abstract: Accurate short-term load forecasting can provide effective guidance for unit scheduling, economic dispatch and power market operations. Concerning the low accuracy problem of load forecasting brought by the limited features of input data, a method based on multi-model fusion using CNN-LSTM-XGBoost framework is proposed. The Long Short-Term Memory network structure fused with local feature pre-extraction module is first established and then integrated with the XGBoost prediction model in parallel. Afterwards by using mean absolute percentage error-reciprocal weight algorithm to set initial model fusion weights and start searching for optimal weight, the optimal fusion model is built. From the prediction experiment of load data by virtual of the proposed method, it is discovered that the mean average percentage error and the root mean squared error of CNN-LSTM-XGBoost are 0.337% and 148.419 MW respectively, which indicates significant decrease of the error metrics compared with the outcome using single network model and multi-model structure. Therefore, it is verified that the method based on multi-model fusion using CNN-LSTM-XGBoost framework has faster training speed, higher accuracy and lower error of prediction.

Key words: short-term load forecasting, local feature pre-extraction, long short-term memory, XGBoost, multi-model fusion

庄家懿, 杨国华, 郑豪丰, 张鸿皓. 基于多模型融合的CNN-LSTM-XGBoost短期电力负荷预测方法[J]. 中国电力, 2021, 54(5): 46-55.

ZHUANG Jiayi, YANG Guohua, ZHENG Haofeng, ZHANG Honghao. Short-Term Load Forecasting Method Based on Multi-model Fusion Using CNN-LSTM-XGBoost Framework[J]. Electric Power, 2021, 54(5): 46-55.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202004026

https://www.electricpower.com.cn/CN/Y2021/V54/I5/46

参考文献

[1] 郭艳飞, 程林, 李洪涛, 等. 基于支持向量机和互联网信息修正的空间负荷预测方法[J]. 中国电力, 2019, 52(4): 80-88
GUO Yanfei, CHENG Lin, LI Hongtao, et al. Spatial load forecasting method based on support vector machine and internet information correction[J]. Electric Power, 2019, 52(4): 80-88
[2] DUDEK G. Pattern-based local linear regression models for short-term load forecasting[J]. Electric Power Systems Research, 2016: 139-147.
[3] GOIA A, MAY C, FUSAI G. Functional clustering and linear regression for peak load forecasting[J]. International Journal of Forecasting, 2010, 26(4): 700-711.
[4] 艾欣, 周志宇, 魏妍萍, 等. 基于自回归积分滑动平均模型的可转移负荷竞价策略[J]. 电力系统自动化, 2017, 41(20): 26-31
AI Xin, ZHOU Zhiyu, WEI Yanping, et al. Bidding strategy for time-shiftable loads based on autoregressive integrated moving average model[J]. Automation of Electric Power Systems, 2017, 41(20): 26-31
[5] 于群, 张铮, 屈玉清, 等. 基于ARMA-GABP组合模型的电网大停电事故损失负荷预测[J]. 中国电力, 2018, 51(11): 38-44
YU Qun, ZHANG Zheng, QU Yuqing, et al. Power loss prediction of large blackouts in power grid based on ARMA-GABP combined model[J]. Electric Power, 2018, 51(11): 38-44
[6] ABBAS F, FENG D, HABIB S, et al. Short term residential load forecasting: an improved optimal nonlinear auto regressive (NARX) method with exponential weight decay function[J]. Electronics, 2018, 7(12): 432.
[7] 刘雨竹, 徐楠. 基于混沌时间序列的IGA- WLSSVR短期负荷预测模型[J/OL]. 控制工程: 1-6[2020-02-07]. https://doi.org/10.14107/j.cnki.kzgc.20190495.
LIU Yuzhu, XU Nan. Short-term load forecasting model using IGA-WLSSVR based on chaotic time series[J/OL]. Control Engineering of China: 1-6[2020-02-07]. https://doi.org/10.14107/j.cnki.kzgc.20190495.
[8] BALIYAN A, GAURAV K, MISHRA S K. A review of short term load forecasting using artificial neural network models[J]. Procedia Comouter Science, 2015, 48: 121-125.
[9] BARBULESCU C, KILYENI S, DEACU A, et al. Artificial neural network based monthly load curves forecasting[C]//2016 IEEE 11th International Symposium on Applied Computational Intelligence and Informatics (SACI). Timisoara, Romania. IEEE, 2016: 237-242.
[10] 林芳, 林焱, 吕宪龙, 等. 基于均衡KNN算法的电力负荷短期并行预测[J]. 中国电力, 2018, 51(10): 88-94, 102
LIN Fang, LIN Yan, LV Xianlong, et al. Short-term parallel power load forecasting based on balanced KNN[J]. Electric Power, 2018, 51(10): 88-94, 102
[11] BIANCHI F M, MAIORINO E, KAMPFFMEYER M C, et al. Recurrent neural networks for short-term load forecasting[M]. Cham: Springer International Publishing, 2017.
[12] QING X Y, NIU Y G. Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM[J]. Energy, 2018, 148: 461-468.
[13] 张宇帆, 艾芊, 林琳, 等. 基于深度长短时记忆网络的区域级超短期负荷预测方法[J]. 电网技术, 2019, 43(6): 29-37
ZHANG Yufan, AI Qian, LIN Lin, et al. A very short-term load forecasting method based on Deep LSTM RNN at zone level[J]. Power System Technology, 2019, 43(6): 29-37
[14] 王增平, 赵兵, 纪维佳, 等. 基于GRU-NN模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(5): 86-95
WANG Zengping, ZHAO Bing, JI Weijia, et al. Short-term load forecasting method based on GRU-NN model[J]. Automation of Electric Power Systems, 2019, 43(5): 86-95
[15] 庄家懿, 杨国华, 郑豪丰, 等. 并行多模型融合的混合神经网络超短期负荷预测[J]. 电力建设, 2020, 41(10): 1-8
ZHUANG Jiayi, YANG Guohua, ZHENG Haofeng, et al. Ultra-short-term load forecasting using hybrid neural network based on parallel multi-model combination[J]. Electric Power Construction, 2020, 41(10): 1-8
[16] 陈振宇, 刘金波, 李晨, 等. 基于LSTM与XGBoost组合模型的超短期电力负荷预测[J]. 电网技术, 2020, 44(2): 614-620
CHEN Zhenyu, LIU Jinbo, LI Chen, et al. Ultra short-term power load forecasting based on combined LSTM-XGBoost model[J]. Power System Technology, 2020, 44(2): 614-620
[17] CHEN T, GUESTRIN C. Xgboost: a scalable tree boosting system[C]//Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016: 785-794.
[18] CHEN T, HE T, BENESTY M, et al. Xgboost: extreme gradient boosting[J]. R package version 0.4-2, 2015: 1-4.
[19] LLOYD J R. GEFCom2012 hierarchical load forecasting: Gradient boosting machines and Gaussian processes[J]. International Journal of Forecasting, 2014, 30(2): 369-374.
[20] ALMALAQ A, EDWARDS G. A review of deep learning methods applied on load forecasting[C]//2017 16th IEEE international conference on machine learning and applications (ICMLA). IEEE, 2017: 511-516.
[21] ABADI Martín, BARHAM P, CHEN J M, et al. TensorFlow: a system for large-scale machine learning[C]// Proceedings of the 12th USENIX conference on Operating Systems Design and Implementation (OSDI'16). USENIX Association, Berkeley, CA, USA, 2016: 265-283.
[22] GOLDBERG Y. Neural network methods for natural language processing[J]. Synthesis Lectures on Human Language Technologies, 2017, 10(1): 1-309.
[23] HUANG C J, KUO P H. A deep cnn-lstm model for particulate matter (PM_2.5) forecasting in smart cities[J]. Sensors, 2018, 18(7): 2220.
[24] GRAVES, A. MOHAMED A R, HINTON G. Speech recognition with deep recurrent neural networks[C]//2013 IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver, BC, Canada. IEEE, 2013: 6645-6649.
[25] 魏延, 曾绍华, 王勇, 等. 基于大训练样本集的ε-SVR改进算法研究[J]. 计算机仿真, 2010, 27(10): 352-356
WEI Yan, ZENG Shaohua, WANG Yong, et al. An improved algorithm of ε-SVR based on large-scale training sample set[J]. Computer Simulation, 2010, 27(10): 352-356
[26] LEE C W, LIN B Y. Applications of the chaotic quantum genetic algorithm with support vector regression in load forecasting[J]. Energies, 2017, 10(11): 1832.

基于多模型融合的CNN-LSTM-XGBoost短期电力负荷预测方法

Short-Term Load Forecasting Method Based on Multi-model Fusion Using CNN-LSTM-XGBoost Framework

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 12

编辑推荐

Metrics

[1]	徐宇颂, 邹山花, 卢先领. 基于特征选择和组合模型的短期电力负荷预测[J]. 中国电力, 2022, 55(7): 121-127.
[2]	贾睿, 杨国华, 郑豪丰, 张鸿皓, 柳萱, 郁航. 基于自适应权重的CNN-LSTM&GRU组合风电功率预测方法[J]. 中国电力, 2022, 55(5): 47-56,110.
[3]	樊江川, 于昊正, 刘慧婷, 杨丽君, 安佳坤. 基于多分支门控残差卷积神经网络的短期电力负荷预测[J]. 中国电力, 2022, 55(11): 155-162,174.
[4]	杨胡萍, 余阳, 汪超, 李向军, 胡奕涛, 饶楚楚. 基于VMD-CNN-BIGRU的电力系统短期负荷预测[J]. 中国电力, 2022, 55(10): 71-76.
[5]	曾囿钧, 肖先勇, 徐方维, 郑林. 基于CNN-BiGRU-NN模型的短期负荷预测方法[J]. 中国电力, 2021, 54(9): 17-23.
[6]	滕伟, 黄乙珂, 吴仕明, 柳亦兵. 基于XGBoost与LSTM的风力发电机绕组温度预测[J]. 中国电力, 2021, 54(6): 95-103.
[7]	吴琼, 余文铖, 洪海生, 喻蕾, 段炼, 尚明远, 刘哲. 基于XGBoost算法的配网台区低压跳闸概率预测[J]. 中国电力, 2020, 53(4): 105-113.
[8]	钱志. 基于改进型SVR的电网短期负荷预测[J]. 中国电力, 2016, 49(8): 54-58.
[9]	朱斌，姜宁，霍雪松，王勇，吴海伟，孙凯祺，胡爽. 南京城市电网核心区短期负荷特性分析及预测[J]. 中国电力, 2016, 49(2): 67-72.
[10]	张玲玲，杨明玉，梁武. 微网用户短期负荷预测相似日选择算法[J]. 中国电力, 2015, 48(4): 156-160.
[11]	赵慧材，陈跃辉，陈瑞先，彭子扬. 结合模糊粗糙集和支持向量机的电力负荷短期预测方法[J]. 中国电力, 2015, 48(2): 45-48.
[12]	陈超, 黄国勇, 邵宗凯, 王晓东, 范玉刚. 基于日特征量相似日的PSO-SVM短期负荷预测[J]. 中国电力, 2013, 46(7): 91-94.

模态框（Modal）标题

基于多模型融合的CNN-LSTM-XGBoost短期电力负荷预测方法

Short-Term Load Forecasting Method Based on Multi-model Fusion Using CNN-LSTM-XGBoost Framework

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 12

编辑推荐

Metrics