Research on Autoformer-based electricity load forecasting and analysis

doi:10.3969/j.issn.1000-5641.2023.05.012

Abstract

Abstract:

Next-generation power grids is the main direction of future smart grid development, and the accurate prediction of power loads is an important basic task of smart grids. To improve the accuracy of load prediction in smart power systems, this work characterized the load dataset based on an Autoformer, a prediction model with an autocorrelation mechanism; adds a feature extraction layer to the original model; optimized the model parameters in terms of the number of coding layers, decoding layers, learning rate, and batch size; and achieved cycle-flexible load prediction. The experimental results show that the model performs better in prediction, with an MAE, MSE, and coefficient of determination of 0.2512, 0.1915, and 0.9832, respectively. Compared with other methods, this method has better load prediction results.

Key words: Autoformer, power load forecasting, autocorrelation mechanism, feature extraction, load characteristics

CLC Number:

TP391

Litao TANG, Zhiyong ZHANG, Jun CHEN, Linna XU, Jiachen ZHONG, Peisen YUAN. Research on Autoformer-based electricity load forecasting and analysis[J]. Journal of East China Normal University(Natural Science), 2023, 2023(5): 135-146.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Fig.12

Table 1

References 23

1	刘洲红.. “双碳”背景下新型电力系统规划新问题及主要技术. 科技创新导报, 2022, 19 (8): 24- 26.
2	方嘉祥.. 智能电网信息安全及新技术研究综述. 科技与创新, 2022, (4): 21- 25.
3	马得银, 孙波, 刘澈.. 基于天气信息的短期冷热电负荷联合预测方法. 电网技术, 2021, 45 (3): 1015- 1022.
4	杨书强, 王涛, 檀晓林, 等.. 基于长短期记忆的图像化短期电力负荷预测方法. 全球能源互联网, 2023, 6 (3): 282- 288.
5	王峰.. 基于IPSO算法的短期电力负荷预测模型研究. 自动化仪表, 2023, 44 (4): 22- 26.
6	李明华, 汪长智, 李会娟.. 基于随机森林的电网基建工程数据自适应整合系统设计. 电子设计工程, 2023, 31 (8): 147- 151.
7	林佳亮, 李暖群, 黄庆键.. 基于支持向量机方法的短期负荷预测研究. 自动化应用, 2016, (12): 150- 152.
8	姜山, 周秋鹏, 董弘川, 等.. 考虑数据周期性及趋势性特征的长期电力负荷组合预测方法. 电测与仪表, 2022, 59 (6): 98- 104.
9	宋海明.. 负荷特性对电力系统电压稳定性的影响. 自动化应用, 2023, 64 (5): 212- 214.
10	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need [C]// 31st Conference on Neural Information Processing Systems (NIPS 2017). Long Beach, CA, USA, 2017.
11	ZHOU H, ZHANG S, PENG J, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting [C]// Proceedings of the AAAI conference on artificial intelligence. 2021, 35(12): 11106-11115.
12	常月, 侯元波, 谭奕舟, 等.. 基于自注意力机制的多模态场景分类. 复旦学报(自然科学版), 2023, 62 (1): 46- 52.
13	WU H, XU J, WANG J, et al.. Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. Advances in Neural Information Processing Systems, 2021, 34, 22419- 22430.
14	张涛. 基于区域动态电碳因子的企业低碳用电研究 [D]. 杭州: 浙江大学, 2023.
15	李烈. 机器学习方法在淮河水位预测中的应用 [D]. 安徽蚌埠: 安徽财经大学, 2022.
16	庞传军, 余建明, 冯长有, 等.. 基于LSTM自动编码器的电力负荷聚类建模及特性分析. 电力系统自动化, 2020, 44 (23): 57- 63.
17	邱子川, 罗丽, 彭润海, 等.. 基于负荷特性的用电分析系统设计. 电子技术, 2023, 52 (2): 216- 217.
18	黄薇, 温蜜, 张照贝. 考虑用户用电行为聚类的电力负荷预测方法[J]. 计算机仿真. 2022, 39(12): 148-153.
19	樊倩男, 刘树勇, 蔡云帆, 等.. 基于Transformer的稳健电力负荷预测. 电力大数据, 2022, 25 (5): 19- 27.
20	朱嘉奕, 宋自根.. 单调激活函数惯性项神经耦合系统的混沌共存. 力学季刊, 2023, 44 (1): 38- 44.
21	殷家伟, 姜宗梁, 徐凯宏.. 数据拟合算法在受电弓检测误差补偿中的应用. 传感器与微系统, 2023, 42 (2): 157- 160.
22	潘锦业, 王苗苗, 阚威, 等.. 基于Adam优化算法和长短期记忆神经网络的锂离子电池荷电状态估计方法. 电气技术, 2022, 23 (4): 25- 30.
23	罗萍, 王涛, 彭云奉.. 基于多层特征融合的多光谱行人检测方法. 计算机工程与设计, 2023, 44 (5): 1579- 1585.

模型名	${R^2}$	MAE	MSE
本文模型	0.9823	0.2512	0.1915
Informer	0.9613	0.2849	0.2387
Transformer	0.9249	0.3196	0.2561

[1]	XU Yang;ZHANG Xuedong . Method on StructureBased Fingerprint Feature Extraction (Chinese) [J]. Journal of East China Normal University(Natural Sc, 2007, 2007(1): 84-90.
[2]	XU Jian-qiang;YUAN Zhen-dong;LI Gao-ping;LIN Jing-yu. Feature Extraction Method for 12-Lead ECG Based on Wavelet Transform(Chinese) [J]. Journal of East China Normal University(Natural Sc, 2005, 2005(2): 27-32.