基于Tri-SE-CNN的风电机组叶片结冰检测研究

孙坚; 杨宇兵

doi:10.19912/j.0254-0096.tynxb.2023-1337

PDF(2064 KB)

太阳能学报 ›› 2024, Vol. 45 ›› Issue (12) : 360-369. DOI: 10.19912/j.0254-0096.tynxb.2023-1337

基于Tri-SE-CNN的风电机组叶片结冰检测研究

孙坚^1,2, 杨宇兵¹

作者信息 +

RESEARCH ON ICING DETECTION OF WIND TURBINE BLADES BASED ON Tri-SE-CNN

Sun Jian^1,2, Yang Yubing¹

Author information +

文章历史 +

摘要

针对现有风力机叶片结冰检测方法未能充分利用无标签数据,且分类性能差的问题,提出一种基于改进的三重训练和卷积神经网络（Tri-SE-CNN）的结冰检测方法。首先建立基于最优加权策略的三重训练（Tri-training）模型,对无标签样本的状态进行判别,用以扩充训练集;接着将压缩与激励（SE）模块嵌入到卷积神经网络（CNN）中,并用SE-CNN分类器学习扩充后的样本集。结合提取的叶片结冰主控特征,以2017年工业大数据创新竞赛平台中15号和21号风力机数据为例进行仿真,并用云南某风场历史数据进行验证。实验结果表明,所提方法的准确度优于CNN、支持向量机等方法,在15号风力机上达到99.96%,可为风力机叶片结冰预警提供有益参考。

Abstract

Aiming to address the issue of existing icing detection methods for wind turbine blades, which fail to fully utilize unlabeled data and have poor classification performance, we propose an approach called Tri-SE-CNN based on improved Tri-training and convolutional neural networks. Firstly, establish a Tri-training model based on an optimal weighted strategy to distinguish the state of unlabeled samples and expand the training set. The expanded sample set is learned by the SE-CNN model that embeds the Squeeze-and-Excite （SE） module into Convolutional Neural Networks （CNN）. Combined with the strong correlation characteristics of blade icing, this paper utilizes the data from wind turbines No.15 and No.21, which were provided by the Industrial Big Data Innovation Competition Platform in China in 2017, for simulation. Additionally, data from a wind farm in Yunnan, China, are used for verification. The experimental results show that the proposed method achieves higher accuracy than CNN, support vector machine and other methods. Specifically, it achieves an accuracy of 99.96% on the No.15 wind turbine, which can provide valuable references for early warning systems regarding wind turbine blade icing.

导出引用

孙坚, 杨宇兵. 基于Tri-SE-CNN的风电机组叶片结冰检测研究[J]. 太阳能学报. 2024, 45(12): 360-369 https://doi.org/10.19912/j.0254-0096.tynxb.2023-1337

Sun Jian, Yang Yubing. RESEARCH ON ICING DETECTION OF WIND TURBINE BLADES BASED ON Tri-SE-CNN[J]. Acta Energiae Solaris Sinica. 2024, 45(12): 360-369 https://doi.org/10.19912/j.0254-0096.tynxb.2023-1337

中图分类号： TM315

参考文献

[1] ALI Q S, KIM M H.Design and performance analysis of an airborne wind turbine for high-altitude energy harvesting[J]. Energy, 2021, 230: 120829.
[2] 何玉林, 李俊, 董明洪, 等. 冰载对风力机性能影响的研究[J]. 太阳能学报, 2012, 33(9): 1490-1496.
HE Y L, LI J, DONG M H, et al.Research on the effect of wind turbine performance under icing conditions[J]. Acta energiae solaris sinica, 2012, 33(9): 1490-1496.
[3] 朱津成, 丁云飞. 基于机器学习的风机叶片结冰预测方法综述[J]. 中国工程机械学报, 2022, 20(2): 129-133.
ZHU J C, DING Y F.A review of wind turbine blade icing prediction methods based on machine learning[J]. Chinese journal of construction machinery, 2022, 20(2): 129-133.
[4] SHU L C, LI H T, HU Q, et al.Study of ice accretion feature and power characteristics of wind turbines at natural icing environment[J]. Cold regions science and technology, 2018, 147: 45-54.
[5] 谭海辉, 李录平, 靳攀科, 等. 风力机叶片超声波除冰理论与方法[J]. 中国电机工程学报, 2010, 30(35): 112-117.
TAN H H, LI L P, JIN P K, et al.Ultrasonic de-icing theory and method for wind turbine blades[J]. Proceedings of the CSEE, 2010, 30(35): 112-117.
[6] GÓMEZ MUÑOZ C Q, GARCÍA MÁRQUEZ F P, SÁNCHEZ TOMÁS J M, et al. Ice detection using thermal infrared radiometry on wind turbine blades[J]. Measurement, 2016, 93: 157-163.
[7] 刘庆超, 郭鹏, 张伟, 等. 多参数模型风电机组叶片结冰监测与预警研究[J]. 太阳能学报, 2022, 43(2): 402-407.
LIU Q C, GUO P, ZHANG W, et al.Study on muti-parameter model of wind turbine blade icing detection and warning[J]. Acta energiae solaris sinica, 2022, 43(2): 402-407.
[8] TAO T, LIU Y Q, QIAO Y H, et al.Wind turbine blade icing diagnosis using hybrid features and Stacked-XGBoost algorithm[J]. Renewable energy, 2021, 180: 1004-1013.
[9] DONG X H, GAO D, LI J, et al.Blades icing identification model of wind turbines based on SCADA data[J]. Renewable energy, 2020, 162: 575-586.
[10] CAO H Q, BAI X, MA X D, et al.Numerical simulation of icing on NREL 5-MW reference offshore wind turbine blades under different icing conditions[J]. China ocean engineering, 2022, 36(5): 767-780.
[11] 刘杰, 杨娜, 谭玉涛, 等. 基于WD-LSTM的风电机组叶片结冰状态评测[J]. 太阳能学报, 2022, 43(8): 399-408.
LIU J, YANG N, TAN Y T, et al.Assessment of icing state of wind turbine blades based on WD-LSTM[J]. Acta energiae solaris sinica, 2022, 43(8): 399-408.
[12] 海涛, 范恒, 王楷杰, 等. 基于PSO-SVM算法的风电机组结冰故障诊断[J]. 智慧电力, 2021, 49(4): 1-6, 74.
HAI T, FAN H, WANG K J, et al.Icing fault diagnosis of wind turbines based on PSO-SVM algorithm[J]. Smart power, 2021, 49(4): 1-6, 74.
[13] YANG X Y, HUANG X X, GAO X X, et al.Icing diagnosis model for wind turbine blade based on feature optimization and 1D-convolutional neural network[J]. Journal of renewable and sustainable energy, 2022, 14(3): DOI:10.1063/5.0078364.
[14] DING S Y, WANG Z J, ZHANG J E, et al.A PCC-Ensemble-TCN model for wind turbine icing detection using class-imbalanced and label-missing SCADA data[J]. International journal of distributed sensor networks, 2021, 17(11): DOI: 10.1177/15501477211057737.
[15] CHENG X, SHI F, LIU X F, et al.A novel deep class-imbalanced semisupervised model for wind turbine blade icing detection[J]. IEEE transactions on neural networks and learning systems, 2022, 33(6): 2558-2570.
[16] ZHOU Z H, LI M.Tri-training: exploiting unlabeled data using three classifiers[J]. IEEE transactions on knowledge and data engineering, 2005, 17(11): 1529-1541.
[17] TSEN C M, HUANG T W, LIU T J.Data labeling with novel decision module of tri-training[C]//2020 2nd International Conference on Computer Communication and the Internet. Nagoya, Japan, 2020.
[18] XU H L, LI L Y, GUO P S.Semi-supervised active learning algorithm for SVMs based on QBC and tri-training[J]. Journal of ambient intelligence and humanized computing, 2021, 12(9): 8809-8822.
[19] 唐承, 郭书祥, 莫延彧, 等. 应用粒子群-序列二次规划算法的结构可靠性优化[J]. 空军工程大学学报(自然科学版), 2016, 17(2): 107-111.
TANG C, GUO S X, MO Y Y, et al.An optimal design of structural reliability based on particle swarm optimization-sequential quadratic programming algorithm[J]. Journal of Air Force Engineering University(natural science edition), 2016, 17(2): 107-111.
[20] 夏睿, 高云鹏, 朱彦卿, 等. 基于SE-CNN模型的窃电检测方法研究[J]. 电力系统保护与控制, 2022, 50(20): 117-126.
XIA R, GAO Y P, ZHU Y Q, et al.A detection method of electricity theft behavior based on an SE-CNN model[J]. Power system protection and control, 2022, 50(20): 117-126.
[21] 石重托, 姚伟, 黄彦浩, 等. 基于SE-CNN和仿真数据的电力系统主导失稳模式智能识别[J]. 中国电机工程学报, 2022, 42(21): 7719-7731.
SHI Z T, YAO W, HUANG Y H, et al.Power system dominant instability mode identification based on convolutional neural networks with squeeze and excitation block and simulation data[J]. Proceedings of the CSEE, 2022, 42(21): 7719-7731.
[22] JIN X H, ZHANG X Y, CHENG X, et al.A physics-based and data-driven feature extraction model for blades icing detection of wind turbines[J]. IEEE sensors journal, 2023, 23(4): 3944-3954.
[23] 白旭, 杜越, 曹慧清, 等. 明冰对寒区海上风力机叶片气动性能影响的分析[J]. 中国造船, 2023, 64(1): 34-46.
BAI X, DU Y, CAO H Q, et al.Study on aerodynamic performance of offshore wind turbine blades under glaze ice condition in cold region[J]. Shipbuilding of China, 2023, 64(1): 34-46.
[24] 周玲, 任永. 大偏航角下基于IPC的风力机变速率停机控制研究[J]. 太阳能学报, 2023, 44(3): 178-184.
ZHOU L, REN Y.Research on IPC-based variable-speed shutdown control of wind turbines under large yaw angles[J]. Acta energiae solaris sinica, 2023, 44(3): 178-184.