Abstract
Due to the difficulty in low diagnostic accuracy under the three-level inverter with multiple faults, a novel fault detection approach based on the improved stacked auto-encoder (SAE) is proposed in this paper. By analyzing the circuit characteristics of the three-level Neutral Point Clamped (NPC) inverter, the inverter bridge arm phase voltage is selected as the fault characteristic signal, and the feature vector formed by the amplitude spectrum is used as the input of SAE network. Since the feature extraction performance of the SAE network has a great correlation with the number of hidden layer nodes, the gravitational search algorithm is chosen to obtain the optimal neuron number of hidden layers, so that the deep learning network could extract features automatically and classify the multiple faults accurately. Finally, the simulation and comparison analysis results show that the proposed method has higher classification accuracy compared with traditional BP neural network and support vector machine. The classification accuracy rate of the three-level NPC inverter faults can reach 100%, which further validates the effectiveness of the developed method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Teymour, H.R., Sutanto, D., Muttaqi, K.M., et al.: Solar PV and battery storage integration using a new configuration of a three-level NPC inverter with advanced control strategy. IEEE Trans. Energy Convers. 29(2), 354–365 (2014)
Livera, A., Theristis, M., Makrides, G., et al.: Recent advances in failure diagnosis techniques based on performance data analysis for grid-connected photovoltaic systems. Renewable Energy 133, 126–143 (2019)
Wang, B., Cai, J., Du, X., et al.: Review of power semiconductor device reliability for power converters. CPSS Trans. Power Electron. Appl. 2(2), 101–117 (2017)
Song, Y., Wang, B.: Survey on reliability of power electronic systems. IEEE Trans. Power Electron. 28(1), 591–604 (2013)
Choi, U., Lee, J., Blaabjerg, F., et al.: Open-circuit fault diagnosis and fault-tolerant control for a grid-connected NPC inverter. IEEE Trans. Power Electron. 31(10), 7234–7247 (2016)
Shen, Y., Miao, B.: Open circuit fault diagnosis strategy for switch of three level inverter. J. Syst. Simul. 30(8), 3058–3065 (2018)
Dhumale, R.B., Lokhande, S.D.: Neural network fault diagnosis of voltage source inverter under variable load conditions at different frequencies. Measurement 91, 565–575 (2016)
Abari, I., Lahouar, A., Hamouda, M., et al.: Fault detection methods for three-level NPC inverter based on DC-bus electromagnetic signatures. IEEE Trans. Ind. Electron. 65(7), 5224–5236 (2018)
Sun, Q., Wang, Y., Jiang, Y.: A novel fault diagnostic approach for DC-DC converters based on CSA-DBN. IEEE Access 6, 6273–6285 (2018)
Qi, Y., Shen, C., Wang, D., et al.: Stacked sparse autoencoder-based deep network for fault diagnosis of rotating machinery. IEEE Access 5, 15066–15079 (2017)
Acknowledgement
This work is supported by Shanghai Science and Technology Innovation Plan (17595800900).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Wu, J., Yan, Z., Sun, Q. (2020). Multiple Faults Detection of Three-Level NPC Inverter Based on Improved Deep Learning Network. In: Abawajy, J., Choo, KK., Islam, R., Xu, Z., Atiquzzaman, M. (eds) International Conference on Applications and Techniques in Cyber Intelligence ATCI 2019. ATCI 2019. Advances in Intelligent Systems and Computing, vol 1017. Springer, Cham. https://doi.org/10.1007/978-3-030-25128-4_195
Download citation
DOI: https://doi.org/10.1007/978-3-030-25128-4_195
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-25127-7
Online ISBN: 978-3-030-25128-4
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)