Fault Diagnosis of Photovoltaic Array Based on Deep Belief Network Optimized by Genetic Algorithm*

doi:10.23919/CJEE.2020.000024

中国电气工程学报（英文） ›› 2020, Vol. 6 ›› Issue (3): 106-114.doi: 10.23919/CJEE.2020.000024

• • 上一篇

收稿日期:2019-10-20 修回日期:2020-04-18 接受日期:2020-05-21 发布日期:2020-10-14

Fault Diagnosis of Photovoltaic Array Based on Deep Belief Network Optimized by Genetic Algorithm^*

Caixia Tao¹, Xu Wang^1,*, Fengyang Gao¹, Min Wang²

1. Department of Automation and Electrical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China;
2. Ninghe Power Supply Co., Ltd., Tianjin 301500, China

Received:2019-10-20 Revised:2020-04-18 Accepted:2020-05-21 Published:2020-10-14
Contact: * E-mail: haizylxi@163.com
About author:Caixia Tao was born in Gansu Province of China, in January, 1972. She graduated from the Hunan University, Changsha, China, in 1994. She received a master's degree in electrical engineering from Lanzhou Jiaotong University in China. Presently, she is a professor at Lanzhou Jiaotong University in China. Her main research interests include motor and its control, and health management of PV systems.
Xu Wang was born in Gansu Province of China, in November, 1992. He is a Master's candidate in electrical engineering from the Lanzhou Jiaotong University in China. His main research interests include intelligent detection and health management of PV systems.
Fengyang Gao was born in Gansu Province of China, in 1970. He received a master's degree in electrical engineering from Southwest Jiaotong University, Chengdu, China, in 2005. Presently, he is a professorate senior engineer at Lanzhou Jiaotong University in China. His main research interests include safe and reliable operation of power system, and health assessment of distribution network.
Min Wang was born in Gansu Province of China, in September, 1995. He graduated from the North China Electric Power University, Baoding, China, in 2018. Presently, he is a construction project manager at State Grid Tianjin Ninghe Power Supply Company Limited in China. His main research interests include healthy and reliable operation of distribution network.
Supported by:
*National Key Research and Development Program of China (2017YFB1201003-020) and the Science and Technology Project of Gansu Province (18YF1FA058).

摘要/Abstract

Abstract: When using deep belief networks (DBN) to establish a fault diagnosis model, the objective function easily falls into a local optimum during the learning and training process due to random initialization of the DBN network bias and weights, thereby affecting the computational efficiency. To address the problem, a fault diagnosis method based on a deep belief network optimized by genetic algorithm (GA-DBN) is proposed. The method uses the restricted Boltzmann machine reconstruction error to structure the fitness function, and uses the genetic algorithm to optimize the network bias and weight, thus improving the network accuracy and convergence speed. In the experiment, the performance of the model is analyzed from the aspects of reconstruction error, classification accuracy, and time-consuming size. The results are compared with those of back propagation optimized by the genetic algorithm, support vector machines, and DBN. It shows that the proposed method improves the generalization ability of traditional DBN, and has higher recognition accuracy of photovoltaic array faults.

Key words: Deep belief network (DBN), fault diagnosis, genetic algorithm, PV array, recognition accuracy

. [J]. 中国电气工程学报（英文）, 2020, 6(3): 106-114.

Caixia Tao, Xu Wang, Fengyang Gao, Min Wang. Fault Diagnosis of Photovoltaic Array Based on Deep Belief Network Optimized by Genetic Algorithm^*[J]. Chinese Journal of Electrical Engineering, 2020, 6(3): 106-114.

参考文献

[1] Y Z Wang, Z H Li, C H Wu.Fault diagnosis technologies for photovoltaic system.Chinese Journal of Power Sourses, 2013, 37(9): 1700-1705.
[2] Y Z Wang, C H Wu, D Q Zhou, et al.A survey of fault diagnosis for PV array based on BP neural network.Power System Protection and Control, 2013, 41(16): 108-114.
[3] W Han, H H Wang, C L Wang, et al.Parameter identification based fault diagnosis model of References photovoltaic modules.Power System Technology, 2015, 39(5): 1198-1204.
[4] R Bi, M Ding, Z C Xu, et al.PV array fault diagnosis based on FCM.Acta Energiae Solaries Sinica, 2016, 37(3): 730-736.
[5] L Chen, W Han, J W Zhang.PV module fault diagnosis based on data fusion.Power System Technology, 2017, 41(6): 1864-1873.
[6] A Etemadi, Z Yi.Fault detection for photovoltaic systems based on multi-resolution signal decomposition and fuzzy inference systems. IEEE Transactions on Smart Grid, 2017, 99: 1-1.
[7] H Ren, J F Qu, Y Chai, et al.Deep learning for fault diagnosis: The state of the art and challenge.Control and Decision, 2017, 32(8): 1345-1358.
[8] W H Li, W P Shan, X Q Zeng.Bearing fault identification based on deep belief network.Journal of Vibration Engineering, 2016, 29(3): 340-347.
[9] X Shi, Y L Zhu, C Sa, et al.Power transformer fault classifying model based on deep belief network.Power System Protection and Control, 2016, 44(1): 71-76.
[10] C C Che, H W Wang, W Liu.Maintenance level decision for aero-engine based on deep belief network.Journal of Aerospace Power, 2018, 33(6): 1528-1536.
[11] Q M Zhu, J Dang, J F Chen, et al.A method for power system transient stability assessment based on deep belief networks.China Electrical Engineering, 2018, 38(2): 44-49.
[12] M S Li, D Yu, Z M Chen, et al.Fault diagnosis and isolation method for wind turbines based on deep belief network.Electric Machines and Control, 2019, 23(2): 114-122.
[13] H K Jiang, H D Shao, X Q Li.Deep learning theory with application in intelligent fault diagnosis of aircraft.Journal of Mechanical Engineering, 2019, 55(7): 27-34.
[14] J Yang, T Zhou, L F Guo, et al.Lung tumor image recognition algorithm based on cuckoo search and deep belief network. Journal of Computer Applications, 2018, 38(11): 3225-3230.
[15] Z Liang, G Q Sun, H C Li, et al.Short-term load forecasting based on VMD and PSO optimized deep belief network.Power System Technology, 2018, 42(2): 598-606.
[16] T Zhou, H L Lu, B Q Huo.Research progress of deep belief network.Computer Engineering and Applications, 2020, 56(9): 24-32.
[17] Y R Sun, Y X Li, L Chen, et al.Deactivation forecasting of SCR catalyst based on GABP neural networks.Proceedings of the CSEE, 2016, 36(S1): 112-120.
[18] W Fan, Y Y Lin, Z S Li.Prediction model of the creep of piezoceramic based on BP neural network optimized by genetic algorithm.Electric Machines and Control, 2018, 22(7): 91-96.
[19] R R Salakhutdinov, G Hinton.An efficient learning procedure for deep Boltzmann machines.Neural Computation, 2012, 24(8): 1967-2006.
[20] G Hinton.Training products of experts by minimizing contrastive divergence.Neural Computation, 2002, 14(8): 1771-1800.
[21] E L Meyer, E E V Dyk. Assessing the reliability and degradation of photovoltaic module performance parameters.IEEE Trans. on Reliability, 2004, 53(1): 83-93.
[22] D S Pillai, N Rajasekar.A comprehensive review on protection challenges and fault diagnosis in PV systems.Renewable and Sustainable Energy Reviews, 2018, 91: 18-40.

Fault Diagnosis of Photovoltaic Array Based on Deep Belief Network Optimized by Genetic Algorithm^*

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价