电源学报

基于改进BiLSTM模型的光伏阵列故障诊断

PDF下载

罗凯元中国电源学会学生会员 , 刘光宇 , 毛嘉铭

电源学报 | 人工智能与电力电子 2026,24(5): 239-247

收起

电源学报 |人工智能与电力电子 2026 , 24 (5) : 239 -247

基于改进BiLSTM模型的光伏阵列故障诊断

全屏

[Author(id=1322531767408513592, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1427554310@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322531767484011066, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, authorId=1322531767408513592, language=EN, stringName=Kaiyuan LUO, firstName=Kaiyuan, middleName=null, lastName=LUO, prefix=null, suffix=Student Member, CPSS, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322531767551119931, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, authorId=1322531767408513592, language=CN, stringName=罗凯元中国电源学会学生会员, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=杭州电子科技大学自动化学院，杭州　310018, bio={"content":"

罗凯元（1999—　），男，中国电源学会学生会员，硕士研究生。研究方向：光伏阵列故障诊断，人工智能。E-mail：1427554310@qq.com。

"}, bioImg=null, bioContent=

罗凯元（1999—　），男，中国电源学会学生会员，硕士研究生。研究方向：光伏阵列故障诊断，人工智能。E-mail：1427554310@qq.com。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322531767324627508, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, xref=null, ext=[AuthorCompanyExt(id=1322531767337210421, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, companyId=1322531767324627508, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China), AuthorCompanyExt(id=1322531767345599030, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, companyId=1322531767324627508, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=杭州电子科技大学自动化学院，杭州　310018)])]), Author(id=1322531767630811709, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=g.liu@hdu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1322531767702114879, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, authorId=1322531767630811709, language=EN, stringName=Guangyu LIU, firstName=Guangyu, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322531767786000960, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, authorId=1322531767630811709, language=CN, stringName=刘光宇, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=杭州电子科技大学自动化学院，杭州　310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322531767324627508, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, xref=null, ext=[AuthorCompanyExt(id=1322531767337210421, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, companyId=1322531767324627508, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China), AuthorCompanyExt(id=1322531767345599030, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, companyId=1322531767324627508, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=杭州电子科技大学自动化学院，杭州　310018)])]), Author(id=1322531767848915522, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=8275943552@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322531767924412996, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, authorId=1322531767848915522, language=EN, stringName=Jiaming MAO, firstName=Jiaming, middleName=null, lastName=MAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1322531767995716165, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1322531766489961007, authorId=1322531767848915522, language=CN, stringName=毛嘉铭, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=杭州电子科技大学自动化学院，杭州　310018, bio={"content":"

毛嘉铭（2000—　），男，硕士研究生。研究方向：光伏功率预测。E-mail：8275943552@qq.com。

"}, bioImg=null, bioContent=

毛嘉铭（2000—　），男，硕士研究生。研究方向：光伏功率预测。E-mail：8275943552@qq.com。

罗凯元中国电源学会学生会员, 刘光宇, 毛嘉铭

作者信息

杭州电子科技大学自动化学院，杭州　310018

通讯作者:

刘光宇（1975—　），男，通信作者，博士。研究方向：非线性控制理论，系统辨识，模式识别。E-mail：g.liu@hdu.edu.cn。

作者简介:

罗凯元（1999—　），男，中国电源学会学生会员，硕士研究生。研究方向：光伏阵列故障诊断，人工智能。E-mail：1427554310@qq.com。

毛嘉铭（2000—　），男，硕士研究生。研究方向：光伏功率预测。E-mail：8275943552@qq.com。

Photovoltaic Array Fault Diagnosis Based on Improved BiLSTM Model

Kaiyuan LUO Student Member, CPSS, Guangyu LIU, Jiaming MAO

Affiliations

School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China

出版时间: 2026-05-30 doi: 10.13234/j.issn.2095-2805.2026.5.239

文章导航

摘要

收起

为了正确辨识光伏阵列运行状态并降低故障漏诊率，提出一种改进双向长短期记忆网络BiLSTM（bidirectional long short-term memory network）的光伏阵列故障诊断方法。首先，结合注意力（Attention）机制优化调节故障标签的注意力权重分配，降低光伏阵列故障诊断的漏诊率；其次，通过融合佳点集、分段自适应逃逸能量机制及黄金正弦3种策略对传统的哈里斯鹰优化HHO（Harris hawk optimization）算法进一步优化，并通过改进的哈里斯鹰优化IHHO（improved HHO）算法对BiLSTM模型超参数进行寻优。对比实验表明，IHHO-Attention-BiLSTM模型对光伏阵列故障诊断的准确率达到99.26%，漏诊率低至0.13%。

关键词

光伏阵列 / 故障诊断 / 漏诊 / 双向长短期记忆网络 / 注意力机制

Abstract

收起

In order to correctly identify the operating status of photovoltaic arrays and reduce the missed diagnosis rate of faults, an improved bidirectional long short-term memory network (BiLSTM) fault diagnosis method for photovoltaic arrays is proposed. Firstly, by combining Attention mechanisms to optimize and adjust the attention weight allocation of fault labels, the missed diagnosis rate of photovoltaic array fault diagnosis can be reduced; Secondly, the traditional Harris hawks optimization (HHO) algorithm is further optimized by integrating the best point set, segmented adaptive escape energy mechanism, and golden sine strategy, and the optimized Harris hawks optimization (IHHO) algorithm is used to optimize the hyperparameters of the BiLSTM model. Comparative experiments show that the IHHO-Attention-BiLSTM model achieves a fault diagnosis accuracy of 99.26% for photovoltaic arrays, with a missed diagnosis rate as low as 0.13%.

Key words

Photovoltaic array / fault diagnosis / missed diagnosis / bidirectional long short-term memory network / Attention mechanism

引用本文

罗凯元中国电源学会学生会员, 刘光宇, 毛嘉铭. 基于改进BiLSTM模型的光伏阵列故障诊断. 电源学报, 2026 , 24 (5) : 239 -247 . DOI: 10.13234/j.issn.2095-2805.2026.5.239

Kaiyuan LUO, Guangyu LIU, Jiaming MAO. Photovoltaic Array Fault Diagnosis Based on Improved BiLSTM Model[J]. Journal of Power Supply, 2026 , 24 (5) : 239 -247 . DOI: 10.13234/j.issn.2095-2805.2026.5.239

基金

收起

国家自然科学基金资助项目(62273124)
杭州电子科技大学研究生科研创新基金资助项目(CXJJ2023202)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

[1]

Sun Tianyi, Xing Huishuang, Cao Shengxian, et al. A novel detection method for hot spots of photovoltaic (PV) panels using improved anchors and prediction heads of YOLOv5 network[J]. Energy Reports, 2022, 8: 1219-1229.

[2]

Waqar Akram M, Li Guiqiang, Jin Yi, et al. Improved outdoor thermography and processing of infrared images for defect detection in PV modules[J]. Solar Energy, 2019, 190: 549-560.

[3]

吕玉坤, 周庆文, 魏子安, 等. 积灰对光伏组件性能影响的试验研究及数值模拟[J]. 电源学报, 2024, 22(2): 175-182. Lü Yukun, Zhou Qingwen, Wei Zian, et al. Experimental study and numerical simulation of influence of ash deposition on performance of photovoltaic modules[J]. Journal of Power Supply, 2024, 22(2): 175-182. (in Chinese)

[4]

俞玮捷, 刘光宇. 基于BP神经网络的光伏系统故障诊断方法[J]. 杭州电子科技大学学报, 2018, 38(4): 52-57, 89. Yu Weijie, Liu Guangyu. A fault diagnosis technique for photovoltaic system based on BP neural networks[J]. Journal of Hangzhou Dianzi University, 2018, 38(4): 52-57, 89. (in Chinese)

[5]

戴森柏, 陈志聪, 吴丽君, 等. 利用LSTM和稳态时间序列的光伏阵列故障诊断方法[J]. 福州大学学报(自然科学版), 2022, 50(1): 54-60. Dai Senbo, Chen Zhicong, Wu Lijun, et al. A photovoltaic array fault diagnosis method using LSTM and steady-state time series[J]. Journal of Fuzhou University (Natural Science Edition), 2022, 50(1): 54-60. (in Chinese)

[6]

程起泽, 陈泽华, 张雲钦, 等. 基于CNN-LSTM的太阳能光伏组件故障诊断研究[J]. 电子技术应用, 2020, 46(4): 66-70. Cheng Qize, Chen Zehua, Zhang Yunqin, et al. Research on fault diagnosis of solar photovoltaic module based on CNN-LSTM[J]. Application of Electronic Technique, 2020, 46(4): 66-70. (in Chinese)

[7]

李斌, 郭自强, 高鹏. 改进北方苍鹰算法在光伏阵列中应用研究[J]. 电子测量与仪器学报, 2023, 37(7): 131-139. Li Bin, Guo Ziqiang, Gao Peng. Application of improved northern goshawk optimization algorithm in photovoltaic array[J]. Journal of Electronic Measurement and Instrumentation, 2023, 37(7): 131-139. (in Chinese)

[8]

Gong Bin, An Aimin, Shi Yaoke, et al. Fast fault detection method for photovoltaic arrays with adaptive deep multiscale feature enhancement[J]. Applied Energy, 2024, 353: 122071.

[9]

Lu Xiaoyang, Lin Yaohai, Lin Peijie, et al. Efficient fault diagnosis approach for solar photovoltaic array using a convolutional neural network in combination of generative adversarial network under small dataset[J]. Solar Energy, 2023, 253: 360-374.

[10]

Chao K H, Liao B J, Hung C P. Applying a cerebellar model articulation controller neural network to a photovoltaic power generation system fault diagnosis[J]. International Journal of Photoenergy, 2013, 2013: 839621.

[11]

孙培胜, 陈堂贤, 程陈, 等. 基于SOA-SVM模型的光伏阵列故障诊断研究[J]. 电源学报, 2025, 23(1): 143-150. Sun Peisheng, Chen Tangxian, Cheng Chen, et al. Research on fault diagnosis of photovoltaic array based on SOA-SVM model[J]. Journal of Power Supply, 2025, 23(1): 143-150. (in Chinese)

[12]

张逸豪, 王振雷. 基于最大信息系数的分组支持向量数据描述故障检测[J]. 化工学报, 2023, 74(9): 3865-3878. Zhang Yihao, Wang Zhenlei. Fault detection using grouped support vector data description based on maximum information coefficient[J]. CIESC Journal, 2023, 74(9): 3865-3878. (in Chinese)

[13]

Fang Mingkun, Zhang Fangfang, Yang Yang, et al. The influence of optimization algorithm on the signal predicttion accuracy of VMD-LSTM for the pumped storage hydropower unit[J]. Journal of Energy Storage, 2024, 78: 110187.

[14]

刘骏鹏. 哈里斯鹰算法的改进及应用研究[D]. 杭州: 浙江大学, 2021. Liu Junpeng. Research on improved Harris hawks optimization and its applications[D]. Hangzhou: Zhejiang University, 2021. (in Chinese)

[15]

Ji Bai, Lu Xiaozheng, Sun Geng, et al. Bio-inspired feature selection: An improved binary particle swarm optimization approach[J]. IEEE Access, 2020, 8: 85989- 86002.

[16]

于建芳, 刘升, 王俊杰, 等. 融合莱维飞行与黄金正弦的蚁狮优化算法[J]. 计算机应用研究, 2020, 37(8): 2349-2353. Yu Jianfang, Liu Sheng, Wang Junjie, et al. Ant lion optimization algorithm integrating with Lévy flight and golden sine[J]. Application Research of Computers, 2020, 37(8): 2349-2353. (in Chinese)

2026年第24卷第5期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.13234/j.issn.2095-2805.2026.5.239

接收时间：2024-03-05
首发时间：2026-10-29
出版时间：2026-05-30

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2024-03-05
修回日期：2024-04-03
录用日期：2024-04-23

基金

National Natural Science Foundation of China(62273124)

国家自然科学基金资助项目(62273124)

Hangzhou Dianzi University Graduate Research Innovation Fund(CXJJ2023202)

杭州电子科技大学研究生科研创新基金资助项目(CXJJ2023202)

作者信息

杭州电子科技大学自动化学院，杭州　310018

通讯作者:

刘光宇（1975—　），男，通信作者，博士。研究方向：非线性控制理论，系统辨识，模式识别。E-mail：g.liu@hdu.edu.cn。

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/dyxb/CN/10.13234/j.issn.2095-2805.2026.5.239

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT