光伏电站数据为时间序列数据,会受到通信传输、逆变器采集等因素的影响而包含大量异常数据,故该文研究一种基于深度学习的光伏电站数据预处理算法,进行数据清洗等预处理。一方面,根据组串逆变器的工作特性,对光伏电站数据的常见异常类型进行分析标记,结合滑动窗口法划分数据,构建用于深度学习训练的光伏电站数据集。另一方面,从激活函数、损失函数以及隐藏层等方面优化GRU神经网络模型,并利用自建数据集对该模型进行训练和测试。测试结果表明：该模型在实际光伏电站数据上的处理准确率达99.84%。
Abstract
Running data from the photovoltaic power system is time indexed, which may be incomplete caused by low quality communication, and always contain the amount of abnormal data from the inverter. This paper studies the algorithm to preprocess photovoltaic data before being used to evaluate the whole system performance. The preprocessing includes labeling abnormal data and cleaning noise data. One optimized GRU neural network is used to do that, which is trained on our lab-built dataset. The GRU network is optimized to process photovoltaic data more efficiently with the activation function, loss function, and hidden layer. The best accuracy is as good as 99.84% from the test dataset consisting of actual photovoltaic data which is not used in training.
关键词
光伏发电 /
逆变器 /
神经网络 /
数据处理 /
网络性能
Key words
photovoltaic power generation /
electric inverters /
neural networks /
data processing /
network performance
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 中国可再生能源学会.广州举行国际太阳能光伏展太阳能应用趋向大众化[EB/OL]. http://www.cres.org.cn/art/2017/8/23/art_7815_307303.html.
[2] International Renewable Energy Agency.Renewable energy statistics 2022 [EB/OL]. https://www.irena.org/Publications/2022/Jul/Renewable-Energy-Statistics-2022.
[3] JORDAN D C, SILVERMAN T J, WOHLGEMUTH J H, et al.Photovoltaic failure and degradation modes[J]. Progress in photovoltaics, 2017, 25(4): 318-326.
[4] MEYERS B E, APOSTOLAKI-IOSIFIDOU E, SCHELHAS L T.Solar data tools: automatic solar data processing pipeline[C]//IEEE 47th Photovoltaic Specialists Conference (PVSC), Calgary, Canada, 2020.
[5] MATAM M, WALTERS J.Data-integrity checks and balances in monitoring of a solar PV system[C]//2019 IEEE 46th Photovoltaic Specialists Conference (PVSC), Chicago, IL, USA, IEEE, 2019.
[6] LIVERA A, THERISTIS M, KOUMPLI E, et al.Data processing and quality verification for improved photovoltaic performance and reliability analytics[J]. Progress in photovoltaics: research and applications, 2020, 29(2): 143-158.
[7] PULINAKA S, KUMAR P, KAUSHAL R, et al.Performance modelling of PV generation with inverter level data through Internet of photovoltaics(IoPV) using artificial neural networks(ANN)[C]//2018 2nd International Conference on Power, Energy and Environment: Towards Smart Technology(ICEPE), Shillong, Meghalaya, India, IEEE, 2019.
[8] REINDERS A, BOGATINOSKA D C, BRAUN C, et al.Development of a big data bank for PV monitoring data, analysis and simulation in COST Action 'PEARL PV[C]//2019 IEEE 46th Photovoltaic Specialists Conference (PVSC), Chicago, IL, USA, 2019.
[9] SUBRAMANIYAN A B, RONG P, KUITCHE J, et al.Quantification of environmental effects on PV module degradation: a physics-based data-driven modeling method[J]. IEEE journal of photovoltaics, 2018, 8(5):1289-1296.
[10] ZHANG P, LI W Y, LI S W, et al.Reliability assessment of photovoltaic power systems: review of current status and future perspectives[J]. Applied energy, 2013, 104:822-833.
[11] 余操, 许盛之, 姚建曦, 等. 灰尘导致的光伏电站发电损失的对比实验[J]. 太阳能学报, 2022, 43(4): 242-247.
YU C, XU S Z,YAO J X, et al.Experiment to study loss of photovoltaic plant from dusts[J]. Acta energiae solaris sinica, 2022, 43(4): 242-247.
[12] CHO K, MERRIENBOER B, GULCEHRE C, et al.Learning phrase representations using RNN encoder-decoder for statistical machine translation[C]//2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, Qatar, 2014.
[13] YANG S D, YU X Y, ZHOU Y.LSTM and GRU neural network performance comparison study: taking yelp review dataset as an example[C]//2020 International Workshop on Electronic Communication and Artificial Intelligence(IWECAI), Qingdao, China, 2020.
[14] CHOI Y J, LIM H, CHOI H, et al.GAN-based anomaly detection and localization of multivariate time series data for power plant[C]//2020 IEEE International Conference on Big Data and Smart Computing (BigComp),Busan, Korea, IEEE, 2020.
[15] 张有健, 陈晨, 王再见. 深度学习算法的激活函数研究<inline-graphic xlink:href="-43-11-78.xml/img_1.tif"/>[J]. 无线电通信技术, 2021, 47(1): 115-120.
ZHANG Y J, CHEN C, WANG Z J.Research on activation function of deep learnimg algorithm[J]. Radio communications technology, 2021, 47(1) :115-120.
[16] ŞEN S Y, ÖZKURT N.Convolutional neural network hyperparameter tuning with adam optimizer for ECG classification[C]//Innovations in Intelligent Systems and Applications Conference(ASYU), Istanbul, Turkey 2020: 1-6.
[17] MAKAROVA A, SHEN H, PERRONE V, et al.Overfitting in Bayesian optimization: an empirical study and early-stopping solution[J]. arXiv: 2104.08166, 2021.
[18] THAMBAWITA V, JHA D, HAMMER H L, et al.An extensive study on cross-dataset bias and evaluation metrics interpretation for machine learning applied to gastrointestinal tract abnormality classification[J]. ACM transactions on computing for healthcare, 2020, 1(3): 1-29.
PDF(1633 KB)
