Abstract
This paper presents an application of artificial intelligence for the analysis of total losses in power transformers. The method is based on a multilayer feed-forward neural network that uses the Levenberg-Marquard algorithm to adjust the network parameters. The analysis was carried out on a three-phase dry transformer 1000 kVA, 6000/400 V. The data used for developing the neural network were obtained experimentally by measuring on low voltage side of the transformer. The inputs to the developed neural network are: the mean value of the load current, the temperature and the losses in the copper, and the output is the total losses. The database contains 1441 samples obtained by changing the load every 30 s in the interval of 12 h. The network model was developed for a temperature of 25 °C, and then the same model was used to determine total losses at a temperature of 68 °C. Obtained results from the developed neural network were compared with the measured data. The low error value indicates that this neural network can be used for different load and temperature.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Kalić, Đ.: Transformers, Institute for Textbooks and Teaching Resources, Beograd (1991)
de Souza, A.N., da Silva, I.N., de Souza, C.F.L.N., Zago, M.G.: Using artificial neural networks for identification of electrical loses in transformers during the manufacturing phase. In: Proceedings of the 2002 International Joint Conference on Neural Networks, IJCNN 2002, 12–17 May 2002, vol. 2, pp. 1346–1350 (2002)
Suttisinthong, N., Pothisarn, C.: Analysis of electrical losses in transformers using artificial neural networks. In: Proceedings of the International MultiConference of Enginers and Computer Scientists 2014, IMECS 2014, 12–14 March 2014, vol. II, Hong Kong (2014)
Naresh, R., Sharma, V., Vashisth, M.: An integrated neural fuzzy approach for fault diagnosis of transformers. IEEE Trans. Power Deliv. 23(4), 2017–2024 (2008)
Leal, A.G., Jardini, J.A., Magrini, L.C., Ahn, S.U.: Distribution transformer losses evaluation: a new analytical methodology and artificial neural network approach. IEEE Trans. Power Syst. 24(2), 705–712 (2009)
Meng, K., Dong, Z.Y., Wang, D.H., Wong, K.P.: A self-adaptive RBF neural network classifier for transformer fault analysis. IEEE Trans. Power Syst. 25(3), 1350–1360 (2010)
Georgilakis, P.S., Hatziargyriou, N.D., Doulamis, N.D., Doulamis, A.D., Kollias, S.D.: Prediction of iron losses of wound core distribution transformers based on artificial neural networks. Neurocomputing 23, 15–29 (1998)
Yadav, A.K., Azeem, A., Singh, A., Malik, H., Rahi, O.P.: Application research based on artificial neural network (ANN) to predict no load loss for transformer’s design. In: International Conference on Communication Systems and Network Technologies (2011)
Štrac, L.: Modeling electromagnetic properties of steel for calculating stray losses in power transformers, doctoral dissertation, FER Zagreb (2010)
Konjić, T., Švenda, G.: Making decision and optimization with application in the electricity system, 1st edn. Tuzla (2010)
Kupusović, A.: Impact of harmonics on losses and life time of transformers, Master’s thesis, Faculty of Electrical Engineering, University of Tuzla (2017)
Jelušić, P.B.: Development of an automated calculation system using neural networks, Graduate thesis, University of Zagreb, Faculty of Graphic Arts, Zagreb (2016)
Matlab: Neural Network Toolbox, User’s Guide. The Matworkss Inc. (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Čerkezović, E., Konjić, T., Tešanović, M. (2019). Evaluation of Losses in Power Transformer Using Artificial Neural Network. In: Avdaković, S. (eds) Advanced Technologies, Systems, and Applications III. IAT 2018. Lecture Notes in Networks and Systems, vol 60. Springer, Cham. https://doi.org/10.1007/978-3-030-02577-9_39
Download citation
DOI: https://doi.org/10.1007/978-3-030-02577-9_39
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-02576-2
Online ISBN: 978-3-030-02577-9
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)