使用快速傅里叶变换优化周期参数的EMD-FFT-SARIMA光伏发电预测模型

熊川羽; 廖晓红; 何诗英; 陈然; 王巍; 臧楠; 王瀛; 肖梦涵

doi:10.11884/HPLPB202436.230349

使用快速傅里叶变换优化周期参数的EMD-FFT-SARIMA光伏发电预测模型

doi: 10.11884/HPLPB202436.230349

熊川羽^1,,
廖晓红¹,
何诗英^2, ,,
陈然¹,
王巍¹,
臧楠^{2, 3},
王瀛^{2, 4},
肖梦涵^{5, 6}

1.
国网湖北省电力有限公司经济技术研究院，武汉 430000
2.
中国科学院合肥物质科学研究院等离子体物理研究所，合肥 230031
3.
安徽建筑大学机械与电气工程学院，合肥 230601
4.
安徽大学物质科学与信息技术研究院，合肥 230601
5.
华中科技大学，电气与电子工程学院，武汉 430074
6.
国网湖北省电力有限公司电力科学研究院，武汉 430000

基金项目: 北省电力有限公司科技研究项目（521538220003）

详细信息

作者简介:
熊川羽，xcy1989528@126.com

通讯作者:
何诗英，shyinghe@ipp.ac.cn。

中图分类号: TM341
计量
- 文章访问数: 167
- HTML全文浏览量: 99
- PDF下载量: 22
- 被引次数: 0
出版历程
- 收稿日期: 2023-10-11
- 修回日期: 2024-03-11
- 录用日期: 2024-03-11
- 网络出版日期: 2024-05-27
- 刊出日期: 2024-07-04

EMD-FFT-SARIMA photovoltaic power generation prediction model using fast fourier transform optimization cycle parameters

Xiong Chuanyu^1
,,
Liao Xiaohong¹,
He Shiying^{2
, ,},
Chen Ran¹,
Wang Wei¹,
Zang Nan^{2, 3},
Wang Ying^{2, 4},
Xiao Menghan^{5, 6}

1.
Economic and Technological Research Institute, State Grid Hubei Electric Power Co., Ltd. Wuhan 430000, China
2.
Institute of Plasma Physics, Chinese Academy of Sciences, Hefei 230031, China
3.
School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei 230601, China
4.
Institutes of Physical Science and Information Technology, Anhui University, Hefei 230601, China
5.
School of Electrical Engineering and Electronics, Huazhong University of Science and Technology, Wuhan 430074, China
6.
Electric Power Research Institute, State Grid Hubei Electric Power Co., Ltd. Wuhan 430000, China

摘要

摘要: 根据分布式能源工业园区的光伏电力单元特点，对园区光伏发电功率预测模型进行优化，为后续的调度策略提供数据支持。针对经验模式分解（EMD）与季节性差分自回归移动平均模型（SARIMA）相组合的EMD-SARIMA预测模型中，原始数据经过EMD分解得到的各固有本征模态函数（IMF）分量周期计算问题，提出加入快速傅里叶变换（FFT）的周期计算方法，建立EMD-FFT-SARIMA光伏发电功率预测模型。再将每个IMF对应的预测结果进行叠加重构得到最终的预测结果。通过预测结果的误差计算可以发现，加入FFT环节后均方根误差（RMSE）从120.6 MW下降到19.3 MW，平均绝对误差（MAE）从52.87 MW下降到12.3 MW。
- 经验模式分解 /
- 季节性差分自回归移动平均模型 /
- 周期计算 /
- 固有本征模态函数信号分量 /
- 快速傅里叶变换 /
- 光伏发电预测
Abstract: In this paper, the photovoltaic (PV) power prediction model is optimized according to the characteristics of PV output units in distributed energy industrial parks to provide data support for the subsequent dispatching strategy. The EMD-SARIMA forecasting model is a combination of Empirical Mode Decomposition (EMD) and Seasonal Autoregressive Integrated Moving Average (SARIMA). In the model, the problem of determining the period of each IMF component of the signal component is proposed, the period T calculation method incorporating fast Fourier transform (FFT) is proposed, and the obtained period is fed into SARIMA as an input parameter together with the IMF sequence for prediction, which constitutes the EMD-FFT-SARIMA prediction model. Then, the prediction results corresponding to each IMF are superimposed and reconstructed to obtain the final prediction results. The error calculation of the prediction results reveals that the root mean square error (RMSE) decreases from 120.6 MW to 19.3 MW, and the mean absolute error (MAE) decreases from 52.87 MW to 12.3 MW.
- empirical mode decomposition /
- seasonal autoregressive integrated moving average /
- cycle calculation /
- signal component intrinsic mode function /
- fast Fourier transform /
- photovoltaic power generation forecast

HTML全文

图 1 预测模型流程图

Figure 1. Predictive model flowchart

下载: 全尺寸图片幻灯片

图 2 预测值与原始值对比图

Figure 2. Chart of predicted values versus raw values

下载: 全尺寸图片幻灯片

图 3 FFT分解后的分量周期和功率图

Figure 3. Component period and power diagram after FFT decomposition

下载: 全尺寸图片幻灯片

图 4 IMF-SARIMA 模型的预测误差

Figure 4. Prediction error of the IMF-SARIMA model

下载: 全尺寸图片幻灯片

图 5 两个模型预测值的比较

Figure 5. Comparison of the predicted values of the two models

下载: 全尺寸图片幻灯片

图 6 EMD-FFT-SARIMA 预测值与原始值的比较

Figure 6. Comparison between EMD-FT-SARIMA predicted values and original values

下载: 全尺寸图片幻灯片

表 1 p，q参数选择表

Table 1. p, q parameters selection table

mode	A_CF	P_ACF
AR(p)	trailing	truncated after p-order
MA(q)	truncated after q-order	trailing
ARMA(p，q)	trailing	trailing

下载: 导出CSV

表 2 p，q参数选择

Table 2. p, q parameters selection

p	q	B_AIC
0	1	−1311
1	0	−1369
1	1	−1052

下载: 导出CSV

表 3 P，Q参数选择

Table 3. P, Q parameter selection

P	Q	B_AIC
1	1	−1471
2	1	−1699
3	3	−1162

下载: 导出CSV

表 4 EMD-SARIMA 模型预测误差指标

Table 4. EMD-SARIMA model prediction error indicators

predictive models	δ_RMSE	δ_MAE
EMD-SARIMA	120.6	52.87

下载: 导出CSV

表 5 IMF 部件和周期值

Table 5. IMF components and cycle values

portion size	T_period
IMF1	36
IMF2	24
IMF3	48
IMF4	48
IMF5	48
IMF6	11001
IMF7	97
IMF8	5004
IMF9	13291
IMF10	13078
IMF11	1764
IMF12	3333
IMF13	7500
IMF14	15000
IMF15	15000

下载: 导出CSV

表 6 EMD-FFT-SARIMA的预测结果指标

Table 6. EMD-FFT-SARIMA model prediction error indicators

predictive models	δ_RMSE	δ_MAE
EMD-FFT-SARIMA	19.3	12.3
EMD-SARIMA	120.6	52.87

下载: 导出CSV

参考文献(14)

[1]	张晓燕, 林鸿才, 黄波, 等. 基于最优交集相似日的EMD-SVR短期负荷预测[J]. 海峡科学, 2023(7):30-35,41 doi: 10.3969/j.issn.1673-8683.2023.07.007 Zhang Xiaoyan, Lin Hongcai, Huang Bo, et al. Short-term load prediction of EMD-SVR based on optimal intersection similar day[J]. Straits Science, 2023(7): 30-35,41 doi: 10.3969/j.issn.1673-8683.2023.07.007
[2]	何坚, 王晓芳. 基于ARIMA和LS-SVM组合模型的短期风速预测[J]. 机电工程技术, 2023, 52(8):30-34 He Jian, Wang Xiaofang. Short-term wind speed prediction based on ARIMA and LS-SVM composite model[J]. Mechanical & Electrical Engineering Technology, 2023, 52(8): 30-34
[3]	邵必林, 程婉荣. 基于SARIMA模型的短期天然气负荷区间预测[J]. 计算机与现代化, 2023(8):54-59 doi: 10.3969/j.issn.1006-2475.2023.08.009 Shao Bilin, Cheng Wanrong. Short-term natural gas load forecasting based on SARIMA model[J]. Computer and Modernization, 2023(8): 54-59 doi: 10.3969/j.issn.1006-2475.2023.08.009
[4]	娄泽生, 张高博, 贾相宇, 等. 基于FFT和小波变换的台风天气GNSS ZTD周期特征分析[J]. 大地测量与地球动力学, 2023, 43(10):1032-1038 Lou Zesheng, Zhang Gaobo, Jia Xiangyu, et al. Typhoon weather GNSS ZTD cycle characteristics analysis based on FFT and wavelet transform[J]. Journal of Geodesy and Geodynamics, 2023, 43(10): 1032-1038
[5]	王堃, 郑晨, 张立中, 等. 一种基于SARIMA-LSTM模型的电网主机负载预测方法[J]. 计算机工程与科学, 2022, 44(11):2064-2070 Wang Kun, Zheng Chen, Zhang Lizhong, et al. A load forecasting method for power grid host based on SARIMA-LSTM model[J]. Computer Engineering and Science, 2022, 44(11): 2064-2070
[6]	李军, 梁宵, 梁嘉城, 等. 基于ARIMA-GM改进算法的建筑负荷预测[J]. 电气自动化, 2023, 45(6):59-61 Li Jun, Liang Xiao, Liang Jiacheng, et al. Building load forecasting based on improved ARIMA-GM algorithm[J]. Electrical Automation, 2023, 45(6): 59-61
[7]	周亚中, 何怡刚, 邢致恺, 等. 基于IDBO-ARIMA的电力变压器振动信号预测[J]. 电子测量与仪器学报, 2023, 37(8):11-20 Zhou Yazhong, He Yigang, Xing Zhikai, et al. Power transformer vibration signal prediction based on IDBO-ARIMA[J]. Journal of Electronic Measurement and Instrumentation, 2023, 37(8): 11-20
[8]	郭忠臣, 邹慧. 基于快速傅里叶变换法的地球自转参数周期性研究[J]. 宿州学院学报, 2018, 33(1):114-117 Guo Zhongchen, Zou Hui. Research on earth rotation parameters periodicity based on fast Fourier transform method[J]. Journal of Suzhou University, 2018, 33(1): 114-117
[9]	曹庆皇, 陈晓霞, 刘怀彦. 基于SARIMA-LSTM组合模型的网络流量预测方法[J]. 江苏通信, 2023, 39(2):87-91 doi: 10.3969/j.issn.1007-9513.2023.02.019 Cao Qinghuang, Chen Xiaoxia, Liu Huaiyan. A network traffic prediction method based on SARIMA-LSTM combined model[J]. Jiangsu Communication, 2023, 39(2): 87-91 doi: 10.3969/j.issn.1007-9513.2023.02.019
[10]	田密, 熊自民. 基于MARS与AIC准则的泥石流冲出距离数据驱动预测方法[J/OL]. 武汉大学学报(工学版): 1-11[2023-12-29]. http://kns.cnki.net/kcms/detail/42.1675.T.20230828.0924.002.html Tian Mi, Xiong Zimin. A data-driven prediction method of debris flow runout distance based on MARS and AIC criteria[J/OL]. Engineering Journal of Wuhan University, 1-11[2023-12-29]. http://kns.cnki.net/kcms/detail/42.1675.T.20230828.0924.002.html.
[11]	李宏玉, 彭康, 宋来鑫, 等. 基于EMD-BiLSTM-ANFIS的负荷区间预测[J]. 吉林大学学报(信息科学版), 2024, 42(1):176-185 Li Hongyu, Peng Kang, Song Laixin, et al. Load interval forecast based on EMD-BiLSTM-ANFIS[J]. Journal of Jilin University (Information Science Edition), 2024, 42(1): 176-185
[12]	梁基重, 晋涛, 牛曙, 等. 基于EMD-FFT特征提取的GIS机械缺陷诊断方法研究[J]. 电力科学与技术学报, 2023, 38(3):216-223 Liang Jichong, Jin Tao, Niu Shu, et al. Research on GIS mechanical defect diagnosis method based on EMD-FFT feature extraction[J]. Journal of Electric Power Science and Technology, 2023, 38(3): 216-223
[13]	胡慧江, 李利平, 靳昊, 等. 基于高阶统计量偏斜度和赤池信息准则的突涌水微振信号初至拾取方法[J]. 工业建筑, 2023, 53(5):132-136,195 Hu Huijiang, Li Liping, Jin Hao, et al. An initial arrival-moment pickup method of microvibration signals from water inrush based on PAI-S and AIC[J]. Industrial Construction, 2023, 53(5): 132-136,195
[14]	张旭宁. 基于EMD-SARIMA模型的铁路商品汽车运量预测[J]. 物流技术, 2022, 41(7):87-91 Zhang Xuning. Prediction of railway transportation volume of commodity vehicles based on EMD-SARIMA[J]. Logistics Technology, 2022, 41(7): 87-91