基于TCN-BiLSTM-Attention-ESN的光伏功率预测

时培明; 郭轩宇; 杜清灿; 许学方; 贺长波; 李瑞雄

doi:10.19912/j.0254-0096.tynxb.2023-0737

PDF(13501 KB)

太阳能学报 ›› 2024, Vol. 45 ›› Issue (9) : 304-316. DOI: 10.19912/j.0254-0096.tynxb.2023-0737

基于TCN-BiLSTM-Attention-ESN的光伏功率预测

时培明¹, 郭轩宇¹, 杜清灿², 许学方¹, 贺长波³, 李瑞雄⁴

作者信息 +

PHOTOVOLTAIC POWER PREDICTION BASED ON TCN-BILSTM-ATTENTION-ESN

Shi Peiming¹, Guo Xuanyu¹, Du Qingcan², Xu Xuefang¹, He Changbo³, Li Ruixiong⁴

Author information +

文章历史 +

摘要

针对光伏发电功率随机性强、难以准确预测的问题,提出一种基于时间卷积网络（TCN）、双向长短期记忆网络（BiLSTM）和回声状态网络（ESN）的组合预测方法。首先,使用自适应噪声完备集合经验模态分解（CEEMDAN）将功率数据分解为一系列相对平稳、不同波动模式的子功率序列;再将分解重构后的功率序列和其他特征序列输入到TCN-BiLSTM-Attention-ESN组合模型中,其中TCN-BiLSTM-Attention用于提取光伏序列波动特征并构建时空特征向量;最后,将所提取的时空特征向量输入ESN获得预测结果。采用新疆某光伏电站的光伏功率数据进行验证,结果表明与时下先进的预测方法相比,所提方法具有更高的预测精度,有助于提升光伏发电占比,保障电力系统平衡和运行安全。

Abstract

Aiming at the problem of strong randomness of photovoltaic power data and difficulty in accurate prediction, a combined prediction method based on temporal convolutional network （TCN）, bidirectional long short-term memory network （BiLSTM） and echo state network （ESN） is proposed. Firstly, the complete ensemble empirical mode decomposition with adaptive noise analysis （CEEMDAN） is used to decompose the power data into a series of relatively stable sub power subsequences. Then, the decomposed and reconstructed power sequence and other feature sequences are input into the TCN-BiLSTM Attention-ESN. TCN-BiLSTM Attention-ESN is applied to extract features and then spatiotemporal feature vectors are constructed. Finally, the extracted spatiotemporal feature vectors are input into ESN to obtain the prediction results. The proposed method is validated using photovoltaic power data from photovoltaic power stations in Xinjiang, China. The results showe that compared with current advanced prediction methods, the proposed method has higher prediction accuracy, which helps to increase the proportion of photovoltaic power generation and ensure the balance and operation safety of the power system.

导出引用

时培明, 郭轩宇, 杜清灿, 许学方, 贺长波, 李瑞雄. 基于TCN-BiLSTM-Attention-ESN的光伏功率预测[J]. 太阳能学报. 2024, 45(9): 304-316 https://doi.org/10.19912/j.0254-0096.tynxb.2023-0737

Shi Peiming, Guo Xuanyu, Du Qingcan, Xu Xuefang, He Changbo, Li Ruixiong. PHOTOVOLTAIC POWER PREDICTION BASED ON TCN-BILSTM-ATTENTION-ESN[J]. Acta Energiae Solaris Sinica. 2024, 45(9): 304-316 https://doi.org/10.19912/j.0254-0096.tynxb.2023-0737

中图分类号： TM615

参考文献

[1] FENG Y, GONG D Z, ZHANG Q W, et al.Evaluation of temperature-based machine learning and empirical models for predicting daily global solar radiation[J]. Energy conversion and management, 2019, 198: 111780.
[2] 王献志, 曾四鸣, 周雪青, 等. 基于XGBoost联合模型的光伏发电功率预测[J]. 太阳能学报, 2022, 43(4): 236-242.
WANG X Z, ZENG S M, ZHOU X Q, et al.Power forecast of photovoltaic generation based on XGBoost combined model[J]. Acta energiae solaris sinica, 2022, 43(4): 236-242.
[3] 邢晨, 张照贝. 基于改进时间卷积网络的短期光伏出力概率预测方法[J]. 太阳能学报, 2023, 44(2): 373-380.
XING C, ZHANG Z B.Short-term probabilistic forecasting method of photovoltaic output power based on improved temporal convolutional network[J]. Acta energiae solaris sinica, 2023, 44(2): 373-380.
[4] YANG L.A review of the application of artificial intelligence algorithms in photovoltaic power generation prediction[J]. Solar energy, 2020, 29(8): 30-35.
[5] 吴忠强, 谢宗奎, 刘重阳, 等. 基于混沌搜索的改进狮群算法及其在光伏电池参数辨识中的应用[J]. 计量学报, 2021, 42(4): 415-423.
WU Z Q, XIE Z K, LIU C Y, et al.Lion swarm optimization based on chaotic search strategy and application in parameters identification of photovoltaic cell models[J]. Acta metrologica sinica, 2021, 42(4): 415-423.
[6] 胡坚, 刘超. NTLBO算法优化ELM的SOC预测方法[J]. 计量学报, 2022, 43(1): 92-96.
HU J, LIU C.The SOC prediction method of ELM optimized by NTLBO algorithm[J]. Acta metrologica sinica, 2022, 43(1): 92-96.
[7] AKINYELE D O, RAYUDU R K, NAIR N K C. Life cycle impact assessment of photovoltaic power generation from crystalline silicon-based solar modules in Nigeria[J]. Renewable energy, 2017, 101: 537-549.
[8] GANDOMAN F H, ABDEL ALEEM S H E, OMAR N, et al. Short-term solar power forecasting considering cloud coverage and ambient temperature variation effects[J]. Renewable energy, 2018, 123: 793-805.
[9] 刘帅, 朱永利, 张科, 等. 基于误差修正ARMA-GARCH模型的短期风电功率预测[J]. 太阳能学报, 2020, 41(10): 268-275.
LIU S, ZHU Y L, ZHANG K, et al.Short-term wind power forecasting based on error correction ARMA-GARCH model[J]. Acta energiae solaris sinica, 2020, 41(10): 268-275.
[10] LI Y T, WU K, LIU J.Self-paced ARIMA for robust time series prediction[J]. Knowledge-based systems, 2023, 269: 110489.
[11] ZHANG W Q, LIN Z, LIU X L.Short-term offshore wind power forecasting-a hybrid model based on discrete wavelet transform (DWT), seasonal autoregressive integrated moving average (SARIMA), and deep-learning-based long short-term memory (LSTM)[J]. Renewable energy, 2022, 185: 611-628.
[12] 张淑清, 任爽, 姜安琦, 等. PCA-GRNN在综合气象短期负荷预测中的应用[J]. 计量学报, 2017, 38(3): 340-344.
ZHANG S Q, REN S, JIANG A Q, et al.Application of PCA-GRNN in integrated meteorological short-term load forecasting[J]. Acta metrologica sinica, 2017, 38(3): 340-344.
[13] 张世辉, 王红蕾, 陈宇翔, 等. 基于深度学习利用特征图加权融合的目标检测方法[J]. 计量学报, 2020, 41(11): 1344-1351.
ZHANG S H, WANG H L, CHEN Y X, et al.An object detection method based on deep learning using feature map weighted fusion[J]. Acta metrologica sinica, 2020, 41(11): 1344-1351.
[14] ETXEGARAI G, LÓPEZ A, AGINAKO N, et al. An analysis of different deep learning neural networks for intra-hour solar irradiation forecasting to compute solar photovoltaic generators'energy production[J]. Energy for sustainable development, 2022, 68: 1-17.
[15] HE Y T, GAO Q Z, JIN Y Y, et al.Short-term photovoltaic power forecasting method based on convolutional neural network[J]. Energy reports, 2022, 8: 54-62.
[16] DING K, CHEN X, WENG S, et al.Health status evaluation of photovoltaic array based on deep belief network and Hausdorff distance[J]. Energy, 2023, 262: 125539.
[17] LIU R H, WEI J C, SUN G P, et al.A short-term probabilistic photovoltaic power prediction method based on feature selection and improved LSTM neural network[J]. Electric power systems research, 2022, 210: 108069.
[18] 倪超, 王聪, 朱婷婷, 等. 基于CNN-Bi-LSTM的太阳辐照度超短期预测[J]. 太阳能学报, 2022, 43(3): 197-202.
NI C, WANG C, ZHU T T, et al.Super-short-term forecast of solar irradiance based on CNN-Bi-LSTM[J]. Acta energiae solaris sinica, 2022, 43(3): 197-202.
[19] YAO X S, WANG Z S, ZHANG H G.A novel photovoltaic power forecasting model based on echo state network[J]. Neurocomputing, 2019, 325: 182-189.
[20] LI G, GUO S D, LI X F, et al.Short-term forecasting approach based on bidirectional long short-term memory and convolutional neural network for regional photovoltaic power plants[J]. Sustainable energy, grids and networks, 2023, 34: 101019.
[21] MUSTAQEEM, ISHAQ M, KWON S.A CNN-assisted deep echo state network using multiple time-scale dynamic learning reservoirs for generating short-term solar energy forecasting[J]. Sustainable energy technologies and assessments, 2022, 52: 102275.
[22] PENG S B, CHEN R L, YU B, et al.Daily natural gas load forecasting based on the combination of long short term memory, local mean decomposition, and wavelet threshold denoising algorithm[J]. Journal of natural gas science and engineering, 2021, 95: 104175.
[23] MONJOLY S, ANDRÉ M, CALIF R, et al.Hourly forecasting of global solar radiation based on multiscale decomposition methods: a hybrid approach[J]. Energy, 2017, 119: 288-298.
[24] GAO B X, HUANG X Q, SHI J S, et al.Hourly forecasting of solar irradiance based on CEEMDAN and multi-strategy CNN-LSTM neural networks[J]. Renewable energy, 2020, 162: 1665-1683.
[25] 许学方, 胡诗婷, 时培明, 等. 基于高维时空张量CP分解的风速监测缺失数据恢复[J]. 振动与冲击, 2023, 42(11): 163-169.
XU X F, HU S T, SHI P M, et al.Wind speed data recovery based on CP decomposition of a higher-dimensional spatial-temporal tensor[J]. Journal of vibration and shock, 2023, 42(11): 163-169.
[26] 雷亚国, 许学方, 蔡潇, 等. 面向机械装备健康监测的数据质量保障方法研究[J]. 机械工程学报, 2021, 57(4): 1-9.
LEI Y G, XU X F, CAI X, et al.Research on data quality assurance for health condition monitoring of machinery[J]. Journal of mechanical engineering, 2021, 57(4): 1-9.