| 11 | 0 | 7 |
| 下载次数 | 被引频次 | 阅读次数 |
太阳能光伏发电受太阳辐射强度变化影响存在随机性和波动性，准确预测光伏发电量十分重要且具有挑战性。利用基于残差网络（ResNet）的LSTM模型，以两个气象变量为输入，用残差连接将LSTM层的输入和输出相加进行预测运算，并以澳大利亚爱丽丝泉光伏电站的实测数据验证预测准确性，结果表明，ResNet-LSTM模型的各项误差度量均小于LSTM模型，说明ResNet-LSTM模型更加精确，对保障电网稳定性，提高光电消纳能力具有重要意义和推广应用价值。
Abstract：Solar photovoltaic power generation is subject to randomness and fluctuations due to variations in solar radiation intensity, making accurate prediction of photovoltaic power output both important and challenging.A residual network(ResNet)-based long short-term memory(LSTM) model was employed, using two meteorological variables as inputs. By adding residual connections that combine the inputs and outputs of the LSTM layers, the prediction calculation is enhanced. The model's accuracy is validated using actual data from the Alice Springs photovoltaic power plant in Australia. Results indicate that the ResNet-LSTM model's error metrics are all lower than those of the LSTM model, demonstrating the superior accuracy of the ResNet-LSTM model.This prediction method holds significant importance and application value for ensuring grid stability and enhancing photovoltaic power integration.
[1]郝倩男.基于改进神经网络的光伏发电功率预测研究[D].南京:南京邮电大学,2021.
[2]王剑斌,傅金波,陈博.基于强化学习的多模型融合光伏发电功率预测方法[J].太阳能学报,2024,45(6):382-388.
[3]Agga A,Abbou A,Labbadi M,et al.CNN-LSTM:An efficient hybrid deep learning architecture for predicting short-term photovoltaic power production[J].Electric Power Systems Research,2022,208:107908.
[4]何志鹏.基于改进CNN-LSTM的光伏发电功率预测及仿真研究[D].天津:天津理工大学,2022.
[5]李秉晨.基于人工智能的短期光伏发电功率预测研究[D].株洲:湖南工业大学,2022.
[6]周伊,肖先勇,赵清华,等.基于组合数据清洗与改进注意力机制的光伏发电功率预测[J].供用电,2024,41(10):31-37,49.
[7]Ahmed R,Sreeram V,Mishra Y,et al.A review and evaluation of the state-of-the-art in PV solar power forecasting:Techniques and optimization[J].Renewable and Sustainable Energy Reviews,2020,124:109792.
[8]Hossain M S,Mahmood H.Short-term photovoltaic power forecasting using an LSTM neural network and synthetic weather forecast[J].IEEE Access,2020,8:172524-172533.
[9]Huang C J,Kuo P H.Multiple-input deep convolutional neural network model for short-term photovoltaic power forecasting[J].IEEE Access,2019,7:74822-74834.
[10]Yang B,Zhang Z,Li J,et al.Efficient multi-objective rolling strategy of photovoltaic/hydrogen system via short-term photovoltaic power forecasting[J].International Journal of Hydrogen Energy,2024,80:1339-1355.
[11]Qin C,Chen L M,Cai Z T,et al.Long short-term memory with activation on gradient[J].Neural networks:the official journal of the International Neural Network Society,2023,164:135-145.
[12]Lu Z,Jiang X,Kot A.Deep coupled Res Net for lowresolution face recognition[J].IEEE Signal Processing Letters,2018,25(4):526-530.
[13]Wen L,Li X,Gao L.A transfer convolutional neural network for fault diagnosis based on Res Net-50[J].Neural Computing and Applications,2020,32(10):6111-6124.
[14]Wang K J,Qi X X,Liu H D.Photovoltaic power forecasting based LSTM-Convolutional Network[J].Energy,2019,189:116225.
[15]Sulaiman M H,Jadin M S,Mustaffa Z,et al.Short-term forecasting of rooftop retrofitted photovoltaic power generation using machine learning[J].Journal of Building Engineering,2024,94:109948.
[16]Liu Y J,Zhang Z K,Ma L,et al.Short-term prediction of photovoltaic power based on DBSCAN-SVM data cleaning and PSO-LSTM Model[J].Energy Engineering,2024,121(10):3019-3035.
[17]Liu B C,Huo X K.Prediction of photovoltaic power generation and analyzing of carbon emission reduction capacity in China[J].Renewable Energy,2024,222:119967.
[18]Ghenai C,Ahmad F F,Rejeb O.Artificial neural networkbased models for short term forecasting of solar PVpower output and battery state of charge of solar electric vehicle charging station[J].Case Studies in Thermal Engineering,2024,61:105152.
基本信息:
DOI：10.20203/j.cnki.2095-8919.2025.02.012
中图分类号:TM615;TP183
引用信息:
[1]李高鹏,白莉,于克成.一种新型光伏发电功率短期预测模型[J].吉林建筑大学学报,2025,42(02):82-88.DOI:10.20203/j.cnki.2095-8919.2025.02.012.
基金信息:
国家“十四五”重点研发课题（2023YFC3804702）; 中国工程院院地合作项目（2023-DFZD-12）