The subject of the article is relevant, since the proposed method for performing a regression analysis of the operation of asynchronous motors does not have special requirements to the accuracy of measuring the quantities used in the regression analysis and to the volume of a training sample, so it can be used in modern embedded diagnostic systems.
The research methods are based on the use of the support vector machine applied in the regression representation. In this approach, the parameters of a regression model are determined by solving a quadratic programming problem having only one solution. To determine the values of the parameters used to train the model, the general theory of transients in electric machines, methods of mathematical modeling, computational mathematics and methods for determining the symmetric components of generalized vectors are used in the paper.
The regression model based on the support vector machine is used to determine the number of damaged rods of the short-circuited rotor of the asynchronous motor. The efficiency of the model has been confirmed by experimental studies. It has been established that a regression model with the radial basis kernel function has the least value of the mean square deviation. Thus, in cases where a regression relation between controlled coordinates must be used, the use of machine learning methods based on the vector space model whose purpose is to find dividing surfaces between classes located as far as possible from all points of the training set has considerable prospects.
- O. D. Goldberg, I. M. Abdulaev, and A. N. Abiev Automation of parameter control and diagnostics of asynchronous motors, 1991. (Russian)
- E. M. Kovarskiy and Yu. I. Yanko, Testing of electrical machines, 1990. (Russian)
- S. Maldonado, J. Merigo, and J. Miranda, “Redefining support vector machines with the ordered weighted average”, Knowledge-Based Systems, vol. 148, pp. 41–46, 2018.
- O. Sheremet and O. Sadovoy, “Using the support vector regression method for telecommunication networks monitoring”, in Proc. Third International Scientific-Practical Conference «Problems of Infocommunications. Science and Technology», pp. 8–10, Kharkiv, (Kharkiv National University of Radioelectronics, Institute of Electrical and Electronics Engineers), 4-6 October 2016.
- A. Arora, J.J. Lin, A. Gasperian, J. Stein, J. Maldjian, M. J. Kahana, and B. Lega, “Comparison of logistic regression, support vector machines, and deep learning classifiers for predicting memory encoding success using human intracranial”, EEG recordings. Journal of Neural Engineering, vol. 15, no. 6, pp. 1–18, 2018.
- J. I. Park, N. Kim, M. K. Jeong, and K. S. Shin, “Multiphase support vector regression for function approximation with break-points”, Journal of the operational research society, vol. 64, pp. 775–785, 2013.
- V. V. Vyugin, Mathematical foundations of the theory of machine learning and forecasting, 2013. (Russian)
- “SVMlight”, http://svmlight.joachims.org.
- Chih-Chung Chang and Chih-Jen Lin, “LIBSVM. A library for support vector machines”, ACM Trans. Intell. Syst. Technol, vol. 2, pp. 27:1–27:27, 2011.
- “PRTools. A MATLAB toolbox for pattern recognition”, http://prtools.org.
- “Signals, images, systems (ISIS) Research Group. Support Vector Machines”, http://www.isis.ecs.soton.ac.uk/resources/svminfo.
- V. S. Garmash, “The method of monitoring the health of the rotor rods of a short-circuited asynchronous motor”, Energetics, vol. 10, pp. 50–52, 1990. (Russian)
- V. F. Sivokobyilenko and Nuri Abdelbasset, “Diagnostics of the state of squirrel cage rotors of asynchronous machines”, Electricity, vol. 3, pp. 25–26, 1997. (Russian)
- “Scikit-learn. Machine Learning in Python”, http://scikit-learn.org/stable/.
- “Jupyter Notebook Tutorial: The Definitive Guide”, https://www.datacamp.com/community/tutorials /tutorial-jupyter-notebook.