GVMD-TSNE-TCN-LSTMre SHORT TERM FORECAST METHOD FOR PHOTOVOLTAIC POWER GENERATION BASED ON INCREMENTAL LEARNING AND GORILLA OPTIMIZATION

Zhang Yiming; Zhang Yijing; Yang Ziyang; Li Jia; Qian Jing

doi:10.19912/j.0254-0096.tynxb.2024-0446

PDF(4042 KB)

Acta Energiae Solaris Sinica ›› 2025, Vol. 46 ›› Issue (7) : 690-700. DOI: 10.19912/j.0254-0096.tynxb.2024-0446

Special Topics of Academic Papers at the 101th Annual Meeting of the China Association for Science and Technology

GVMD-TSNE-TCN-LSTM^re SHORT TERM FORECAST METHOD FOR PHOTOVOLTAIC POWER GENERATION BASED ON INCREMENTAL LEARNING AND GORILLA OPTIMIZATION

Zhang Yiming¹, Zhang Yijing¹, Yang Ziyang¹, Li Jia^2,3, Qian Jing³

Author information +

History +

Abstract

The data of PV short-term power generation has high dimension and complex features. The key factors affecting the forecasting performance include the decomposition , extraction of data features and the construction of prediction model. In this paper, an improved PV short-term prediction method embedded gorilla parameter optimization algorithm is proposed. In the first layer, variational mode decomposition （VMD） and T-distributed stochastic neighbor（TSNE） embedding are adopted for the feature extraction, which are combined to obtain effective features of the photovoltaic data. VMD involves the selection of two key parameters, namely penalty factor and decomposition mode number. In this paper, the gorilla optimization algorithm is used to optimize these parameters, denoted as GVMD.The second layer is to construct the prediction model. The TCN-LSTM prediction model is built by combining the temporal convolutional neural network （TCN） and the long short-term memory network （LSTM）, and the learning, superposition and reconstruction of various features are completed.On this basis, the incremental learning method, denoted as GVMD-TSNE-TCN-LSTM^reis used to continuously modify the prediction model by the design of parameter freezing and full-connection layer update. Finally, the power data of a photovoltaic field in Gansu Province was simulated to verify the necessity of GVMD-TNSE data processing, the timeliness of GTO parameter optimization algorithm, and the effectiveness of the overall model.

Key words

photovoltaic power generation / short-term power prediction / incremental learning method / gorilla troops optimizer / GVMD-TSNE feature decomposition extraction / TCN-LSTM prediction model

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Zhang Yiming, Zhang Yijing, Yang Ziyang, Li Jia, Qian Jing. GVMD-TSNE-TCN-LSTM^re SHORT TERM FORECAST METHOD FOR PHOTOVOLTAIC POWER GENERATION BASED ON INCREMENTAL LEARNING AND GORILLA OPTIMIZATION[J]. Acta Energiae Solaris Sinica. 2025, 46(7): 690-700 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0446

References

[1] 政府工作报告:2023年3月5日第十四届全国人民代表大会第一次会议 [EB/OL]. (2023-03-14)[2023-08-07]. https://www.gov.cn/premier/2023-03/14/content_5746704. htm.
Government Work Report delivered at the first session of the 14th National People’s Congress on March 5, 2023.[EB/OL].(2023-03-14)[2023-08-07]. https://www.gov.cn/premier/2023-03/14/content_5746704. htm.
[2] 2023年全球电力评论[EB/OL].(2023-04-12)[2023-08-07].https://ember-climate.org/insights/research/global-electricity-review-2023/.
Global Power Review2023[EB/OL].(2023-04-12)[2023-08-07]. https://ember-climate.org/insights/research/global-electricity-review-2023/.
[3] 《2020—2025年可再生能源新增技术,主要案例和加速案例》[EB/OL]. (2020-11-10)[2023-08-07]. https://www.iea.org/data-and-statistics/charts/renewable-capacity-additi ons-by-technology-2020-2025-main-and-accelerated-cases.
Renewable capacity additions by technologies2020—2025:main and accelerated cases[EB/OL]. (2020-11-10)[2023-08-07].https://www.iea.org/data-and-statistics/charts/renewable-capacity-additions-by-technology-2020-2025-main-and-accelerated-cases.
[4] 推动光伏产业高质量发展. 经济日报[N].2023-08-22(010).
Promote the high-quality development of the photovoltaic industry. Economic Daily[N].2023-08-22(010).
[5] 蔡源, 吴浩, 唐丹. 光伏发电功率预测方法综述[J]. 四川电力技术, 2024, 47(2): 25-31.
CAI Y, WU H, TANG D.Reviews of photovoltaic power prediction methods[J]. Sichuan electric power technology, 2024, 47(2): 25-31.
[6] SHIVAM K, TZOU J C, WU S C.Multi-step short-term wind speed prediction using a residual dilated causal convolutional network with nonlinear attention[J]. Energies, 2020, 13(7): 1772.
[7] CHEN Y R, DONG Z K, WANG Y, et al.Short-term wind speed predicting framework based on EEMD-GA-LSTM method under large scaled wind history[J]. Energy conversion and management, 2021, 227: 113559.
[8] SHIVANI, SANDHU K S, NAIR A R. A comparative study of ARIMA and RNN for short term wind speed forecasting[C]//10th International Conference on Computing, Communication and Networking Technologies (ICCCNT). Kanpur, India, 2019.
[9] CHUNG J, GULCEHRE C, CHO K H, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[EB/OL].2014, https://doi.org/10.48550/arXiv.1412.3555.
[10] GERS F A.Learning to forget: continual prediction with LSTM[C]//9th International Conference on Artificial Neural Networks(ICANN). Edinburgh, UK, 1999.
[11] 朱润泽, 王德军. 基于LSTM神经网络的光伏系统功率预测[J]. 电力科技与环保, 2023, 39(3): 201-206.
ZHU R Z, WANG D J.Power prediction of photovoltaic system based on LSTM neural network[J]. Electric power technology and environmental protection, 2023, 39(3): 201-206.
[12] 闫文杰, 徐志杰, 薛桂香, 等. 基于EMD-TCN的多能供热系统太阳辐照度预测模型[J]. 太阳能学报, 2023, 44(11): 182-188.
YAN W J, XU Z J, XUE G X, et al.Solar irradiance forecasting model based on EMD-TCN for multi-energy heating systems[J]. Acta energiae solaris sinica, 2023, 44(11): 182-188.
[13] 董坤, 冉鹏, 刘旭, 等. 基于权参数优化的并行深度学习光伏功率预测[J]. 动力工程学报, 2024, 44(1): 91-98.
DONG K, RAN P, LIU X, et al.Photovoltaic power prediction by weight parameter optimization-based parallel deep learning[J]. Journal of Chinese Society of Power Engineering, 2024, 44(1): 91-98.
[14] 薛阳, 李金星, 杨江天, 等. 基于相似日分析和改进鲸鱼算法优化LSTM网络模型的光伏功率短期预测[J].南方电网技术,2024,18(11):97-105.
XUE Y, LI J X, YANG J T, et.al. Short-Term prediction of photovoltaic power based on similar day analysis and improved whale algorithm to optimize LSTM network model[J]. Southern power system technology,2024,18(11):97-105.
[15] 蔡志豪, 史洪岩. 麻雀搜索算法优化的RF-BILSTM短期电力负荷预测[J]. 黑龙江工程学院学报, 2024, 38(1): 15-20.
CAI Z H, SHI H Y.Short term load forecasting of RF-BILSTM optimized by sparrow search algorithm[J]. Journal of Heilongjiang Institute of Technology, 2024, 38(1): 15-20.
[16] 李建林, 孙浩元, 赵文鼎, 等.基于变分模态分解-多模糊控制的风电混储系统功率分配策略[J/OL]. 上海交通大学学报 , 2024 , https://doi.org/10.16183/j.cnki.jsjtu.2023.572.
LI J L, SUN H Y,ZHAO W D, et.al. Power allocation strategy for wind power hybrid storage system based on variational modal decomposition-multi-fuzzy control[J/OL]. Journal of Shanghai Jiaotong university, 2024, https://doi.org/10.16183/j.cnki.jsjtu.2023.572.
[17] 罗鹏, 刘美俊, 俞辉, 等. 基于MIC-EEMD-改进Informer的含高比例清洁能源与储能的电力市场短期电价多步预测[J]. 电网技术, 48(3):949-957.
LUO P, LIU M J, YU H, et. al. Short-term multi-step price prediction for the electricity market with a high proportion of clean energy and energy storage based on MIC-EEMD-improved Informer[J]. Power system technology, 48(3):949-957.
[18] 李争, 张杰, 徐若思, 等. 基于相似日聚类和PCC-VMD-SSA-KELM模型的短期光伏功率预测[J]. 太阳能学报, 2024, 45(2): 460-468.
LI Z, ZHANG J, XU R S, et al.Short term photovoltaic power prediction based on similar day clustering and PCC-VMD-SSA-KELM model[J]. Acta energiae solaris sinica, 2024, 45(2): 460-468.
[19] 刘志超, 袁三男, 唐万成. 基于BLSTM-随机森林的短期光伏发电输出功率预测[J]. 电源技术, 2021, 45(4): 495-498.
LIU Z C, YUAN S N, TANG W C.Output power prediction of short-term photovoltaic power generation based on BLSTM-random forest[J]. Chinese journal of power sources, 2021, 45(4): 495-498.
[20] 周浩杰, 杨建卫, 王尊, 等. 基于LSTM光伏发电功率超短期预测模型研究[J]. 电源技术, 2023, 47(6): 785-789.
ZHOU H J, YANG J W, WANG Z, et al.Research on ultra-short term prediction model of photovoltaic power generation based on LSTM[J]. Chinese journal of power sources, 2023, 47(6): 785-789.
[21] 付文龙, 章轩瑞, 张海荣,等. 基于INGO-SWGMN混合模型的超短期风速预测研究[J]. 太阳能学报, 45(5):133-143.
FU W L, ZHANG X R, ZHANG H R, et.al. Ultra-short-term wind speed prediction based on INGO-SWGMN hybrid model[J]. Acta energiae solaris sinica, 45(5):133-143.
[22] 汤德清, 朱武, 侯林超. 基于CNN-LSTM-XGBoost模型的超短期光伏功率预测[J]. 电源技术, 2022, 46(9): 1048-1052.
TANG D Q, ZHU W, HOU L C.Ultra-short-term photovoltaic power prediction based on CNN-LSTM-XGBoost model[J]. Chinese journal of power sources, 2022, 46(9): 1048-1052.
[23] 李文斌, 熊亚锟, 范祉辰, 等. 持续学习的研究进展与趋势[J]. 计算机研究与发展, 2024, 61(6): 1476-1496.
LI W B, XIONG Y K, FAN Z C, et al.Advances and trends of continual learning[J]. Journal of computer research and development, 2024, 61(6): 1476-1496.
[24] HOCHREITER S, SCHMIDHUBER J.Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.
[25] 孙玉玺, 刘寅韬, 耿光超, 等. 基于多模式增量更新的短期光伏功率预测方法[J]. 太阳能学报, 2024, 45(9): 386-393.
SUN Y X, LIU Y T, GENG G C, et al.Short-term photovoltaic power forecasting method based on multi-mode incremental update[J]. Acta energiae solaris sinica, 2024, 45(9): 386-393.