考虑波浪作用和时序特性的海上漂浮式光伏阵列故障诊断方法

doi:10.19595/j.cnki.1000-6753.tces.250113

Abstract
Figure/Table
References (0)
Related Citation (15)

Download: PDF (1346 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract

With the global transition towards decarbonization, photovoltaic (PV) power generation has become a key component in achieving carbon neutrality. Offshore floating photovoltaic systems, in particular, have gained attention due to their land-saving nature and the cooling effect of water, which improves power generation efficiency. However, these systems face significant operational challenges due to the harsh marine environment, including the effects of waves, salt spray, and contamination, which can lead to frequent system faults. This paper proposes a fault diagnosis method for offshore floating PV arrays that integrates wave effects and the temporal characteristics of fault progression. It aims at improving system reliability and efficiency in such dynamic environments.
The study first developed an integrated floating PV array model that incorporates both the dynamic behavior of ocean waves and the electrical characteristics of the PV system. By combining wave motion dynamics with the equivalent electrical circuit model of the floating PV array, a multi-model fusion approach was constructed to simulate the power generation characteristics under both normal and fault conditions. These simulations accounted for a variety of faults that could occur in offshore floating PV systems, such as open-circuit faults, short-circuit faults, contamination, aging, and hotspot issues. For each fault type, the system’s performance was evaluated through simulation to identify key fault indicators and characteristic behaviors, focusing on the changes observed in the I-V (current-voltage) curves under different conditions.
A novel fault feature extraction algorithm was introduced, which accounted for the dynamic effects of waves on tilt angles and the temporal characteristics of fault development. The algorithm extracted critical features, including open-circuit voltage, short-circuit current, maximum power point voltage, maximum power current, inflection points on the I-V curves, and environmental factors such as irradiance and temperature. These features were used to construct a fault feature vector that captured both instantaneous fault characteristics and the evolving nature of faults over time.
The primary contribution of this work was the development of a fault diagnosis model that combined Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM) networks, and a Spatial Attention Mechanism (SAM). This hybrid model enhanced the ability to capture both local fault features and temporal dependencies, which were crucial for accurately diagnosing faults in the dynamic marine environment. The CNN component extracted local features from the fault data, while the BiLSTM network captured the temporal dynamics of fault progression. The SAM module helped improve the model’s focus on critical regions of the input data, further enhancing diagnostic performance.
The proposed model was rigorously tested using simulation data and experimental data obtained from a wave simulation setup. The diagnostic performance of the CNN-BiLSTM-SAM model was compared with other approaches, including BP neural networks, CNN-SAM networks, and SE-TCN (Squeeze-and-Excitation Temporal Convolutional Networks). The results showed that the CNN-BiLSTM-SAM model significantly outperformed the other methods in fault detection accuracy. Specifically, the proposed model achieved an average improvement of 11.6% in the F1-score for faults such as contamination and aging, and up to a 50.1% improvement for hotspot faults, which exhibited significant temporal development.
In addition to simulation validation, the model’s robustness was further confirmed through experimental testing in a wave environment. The model maintained high diagnostic accuracy even when subjected to real-world wave dynamics, demonstrating its potential for real-time fault detection in offshore PV systems. The comparison of the confusion matrix and ROC (Receiver Operating Characteristic) curves confirmed that the CNN-BiLSTM-SAM model achieved the highest accuracy, with an AUC (Area Under Curve) value of 0.988, indicating superior fault classification performance compared to other models.

Key words： Offshore floating photovoltaics photovoltaic generation model fault diagnosis of photovoltaic arrays I-V curve time-sequence characteristic extraction of faults

Received: 16 January 2025

PACS:

TM615

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Gao Shuang
	Li Chenhao
	Li Zeyu
	Kong Xiangyu
	Liu Mengmeng

Cite this article:

Gao Shuang,Li Chenhao,Li Zeyu等. Fault Detection of Offshore Floating Photovoltaic Power System Considering Wave Motion and Time-Sequence Characteristics[J]. Transactions of China Electrotechnical Society, 0, (): 20250113-20250113.

URL:

https://dgjsxb.ces-transaction.com/EN/10.19595/j.cnki.1000-6753.tces.250113 OR https://dgjsxb.ces-transaction.com/EN/Y0/V/I/20250113

[1] 肖佳, 梅琦, 黄晓琪, 等. “双碳” 目标下我国光伏发电技术现状与发展趋势[J]. 天然气技术与经济, 2022, 16(5): 64-69.
Xiao Jia, Mei Qi, Huang Xiaoqi, et al.Status quo and development trend of photovoltaic power-generating technology under the dual-carbon goal[J]. Natural Gas Technology and Economy, 2022, 16(5): 64-69.
[2] Liu Haohui, Krishna V, Leung J L, et al.Field experience and performance analysis of floating PV technologies in the tropics[J]. Progress in Photovoltaics: Research and Applications, 2018, 26(12): 957-967.
[3] Sahu A, Yadav N, Sudhakar K.Floating photovoltaic power plant: a review[J]. Renewable and Sustainable Energy Reviews, 2016, 66: 815-824.
[4] 孔祥玉, 邵阳苹, 付强, 等. 海上漂浮式光伏多浮体间电缆跨接方法研究[J]. 工程科学学报, 2025, 47(2): 389-400.
Kong Xiangyu, Shao Yangping, Fu Qiang, et al.Study on a cable jumper method for offshore floating photovoltaic systems with multiple floating bodies[J]. Chinese Journal of Engineering, 2025, 47(2): 389-400.
[5] Bhang B G, Hyun J H, Ahn S H, et al.Optimal design of bifacial floating photovoltaic system with different installation azimuths[J]. IEEE Access, 2022, 11: 1456-1466.
[6] Tina G M, Bontempo Scavo F, Merlo L, et al.Analysis of water environment on the performances of floating photovoltaic plants[J]. Renewable Energy, 2021, 175: 281-295.
[7] Mannino G, Tina G M, Cacciato M, et al.Photovoltaic module degradation forecast models for onshore and offshore floating systems[J]. Energies, 2023, 16(5): 2117.
[8] Lim B Y, Ahn S H, Park M S, et al.Prediction of fault for floating photovoltaics via mechanical stress evaluation of wind speed and wave height[J]. IEEE Access, 2024, 12: 70105-70116.
[9] Kaymak M K, Ahmet Duran Ş.Problems encountered with floating photovoltaic systems under real conditions: a new FPV concept and novel solutions[J]. Sustainable Energy Technologies and Assessments, 2021, 47: 101504.
[10] 王小宇, 刘波, 孙凯, 等. 光伏阵列故障诊断技术综述[J]. 电工技术学报, 2024, 39(20): 6526-6543.
Wang Xiaoyu, Liu Bo, Sun Kai, et al.A review of photovoltaic array fault diagnosis technology[J]. Transactions of China Electrotechnical Society, 2024, 39(20): 6526-6543.
[11] Yu C, Wang H, Yao J, et al.A dynamic alarm threshold setting method for photovoltaic array and its application[J]. Renewable Energy, 2020, 158: 13-22.
[12] 马铭遥, 王泽澳, 马文婷, 等. 基于时空特征的光伏电站草木阴影遮挡故障诊断[J]. 电工技术学报, 2025, 40(3): 832-841.
Ma Mingyao, Wang Zeao, Ma Wenting, et al.Vegetation shading fault diagnosis in photovoltaic power stations based on temporal-spatial characteristics[J]. Transactions of China Electrotechnical Society, 2025, 40(3): 832-841.
[13] 冯锴, 林培杰, 俞金玲, 等. 结合电压阈值和最右端功率峰值点的光伏阵列故障检测与定位[J]. 电气技术, 2021, 22(9): 95-102.
Feng Kai, Lin Peijie, Yu Jinling, et al.Photovoltaic array fault detection and location based on voltage threshold and rightmost power peak point[J]. Electrical Engineering, 2021, 22(9): 95-102.
[14] 吴春华, 易苑, 李智华, 等. 考虑参数权重与分层映射的光伏组件健康程度检测[J]. 电工技术学报, 2024, 39(15): 4856-4867.
Wu Chunhua, Yi Yuan, Li Zhihua, et al.Health detection of photovoltaic modules considering parameter weights and hierarchical mapping[J]. Transactions of China Electrotechnical Society, 2024, 39(15): 4856-4867.
[15] 陈伟, 陈克松, 纪青春, 等. 基于1D-CNN+GRU的光伏阵列故障诊断方法研究[J]. 自动化仪表, 2022, 43(6): 13-17.
Chen Wei, Chen Kesong, Ji Qingchun, et al.Research on fault diagnosis method of photovoltaic array based on 1D-CNN+GRU[J]. Process Automation Instrumentation, 2022, 43(6): 13-17.
[16] 刘开石, 李田泽, 刘东, 等. 基于ABC-SVM算法的光伏阵列故障诊断[J]. 电源技术, 2021, 45(9): 1171-1174.
Liu Kaishi, Li Tianze, Liu Dong, et al.Fault diagnosis of PV array based on ABC-SVM algorithm[J]. Chinese Journal of Power Sources, 2021, 45(9): 1171-1174.
[17] 顾崇寅, 徐潇源, 王梦圆, 等. 基于CatBoost算法的光伏阵列故障诊断方法[J]. 电力系统自动化, 2023, 47(2): 105-114.
Gu Chongyin, Xu Xiaoyuan, Wang Mengyuan, et al.CatBoost algorithm based fault diagnosis method for photovoltaic arrays[J]. Automation of Electric Power Systems, 2023, 47(2): 105-114.
[18] 刘卫亮, 姜锴越, 许之胜, 等. 基于数字孪生与融合神经网络的光伏阵列故障诊断[J]. 太阳能学报, 2024, 45(11): 303-312.
Liu Weiliang, Jiang Kaiyue, Xu Zhisheng, et al.Fault diagnosis of photovoltaic array based on digital twin and fusion neural network[J]. Acta Energiae Solaris Sinica, 2024, 45(11): 303-312.
[19] Zhang Jingwei, Liu Yongjie, Li Yuanliang, et al.A reinforcement learning based approach for on-line adaptive parameter extraction of photovoltaic array models[J]. Energy Conversion and Management, 2020, 214: 112875.
[20] 邱国全, 夏艳君, 杨鸿毅. 晴天太阳辐射模型的优化计算[J]. 太阳能学报, 2001, 22(4): 456-460.
Qiu Guoquan, Xia Yanjun, Yang Hongyi.An optimized clear-day solar radiation model[J]. Acta Energiae Solaris Sinica, 2001, 22(4): 456-460.
[21] 韩斐, 潘玉良, 苏忠贤. 固定式太阳能光伏板最佳倾角设计方法研究[J]. 工程设计学报, 2009, 16(5): 348-353.
Han Fei, Pan Yuliang, Su Zhongxian.Research on optimal tilt angle of fixed PV panel[J]. Journal of Engineering Design, 2009, 16(5): 348-353.
[22] Padovan A, Del Col D.Measurement and modeling of solar irradiance components on horizontal and tilted planes[J]. Solar Energy, 2010, 84(12): 2068-2084.
[23] 李红岩, 王磊, 安平娟, 等. 基于改进黏菌算法的局部遮阴下光伏MPPT研究[J]. 太阳能学报, 2023, 44(10): 129-134.
Li Hongyan, Wang Lei, An Pingjuan, et al.Study on photovoltaic mppt under local shade based on improved slime mold algorithm[J]. Acta Energiae Solaris Sinica, 2023, 44(10): 129-134.
[24] 刘恒, 马铭遥, 张志祥, 等. 热斑晶硅光伏组件I-V曲线分析及特征模拟研究[J]. 太阳能学报, 2021, 42(4): 239-246.
Liu Heng, Ma Mingyao, Zhang Zhixiang, et al.Study on i-v curve analysis and characteristic simulation of silicon photovoltaic module with hot spot[J]. Acta Energiae Solaris Sinica, 2021, 42(4): 239-246.
[25] 车曦. 基于红外图像识别的光伏组件热斑故障检测方法研究[D]. 重庆: 重庆大学, 2015.
Che Xi.Research on hot spot fault detection method of photovoltaic module based on infrared image recognition[D]. Chongqing: Chongqing University, 2015.
[26] Fu Xueqian, Zhang Chunyu, Zhang Xiurong, et al.A novel GAN architecture reconstructed using Bi-LSTM and style transfer for PV temporal dynamics simulation[J]. IEEE Transactions on Sustainable Energy, 2024, 15(4): 2826-2829.