The use of computers now in almost all areas of activity for solving a wide variety of tasks, including optimization tasks, leads to the need for a specialist in this field to be able to correctly set and solve problems of this kind. The use of cloud computing Microsoft Azure Machine Learning (AML) Studio allows to develop predictive analytic projects efficiently and with minimal developer influence on the quality of calculations. In this paper, we consider the deployment of an induction motor with squirrel cage rotor optimization project. Compared with the standard method offered by the Microsoft AML service, a set of results of experimental data of a series of electric machines is used to evaluate the quality of the model and predicted target values. In addition, in a separate metadata block, the Python script for an electric machine automated design is included, considering the range of initial variables. Other metadata blocks in Python are used in order to generate series of initial data. It is also shown how to implement reuse of the created project with the cloud web service and Microsoft Excel tables.
[1]
David E. Goldberg,et al.
Genetic Algorithms in Search Optimization and Machine Learning
,
1988
.
[2]
Sean R Eddy,et al.
What is dynamic programming?
,
2004,
Nature Biotechnology.
[3]
V. Pliugin,et al.
Using object-oriented design principles in electric machines development
,
2016
.
[4]
Maria Sukhonos,et al.
Using Azure Maching Learning Cloud Technology for Electric Machines Optimization
,
2018,
2018 International Scientific-Practical Conference Problems of Infocommunications. Science and Technology (PIC S&T).
[5]
Andreas Müller,et al.
Introduction to Machine Learning with Python: A Guide for Data Scientists
,
2016
.
[6]
Stephen P. Boyd,et al.
Convex Optimization
,
2004,
Algorithms and Theory of Computation Handbook.
[7]
V. E. Pliugin,et al.
IMPLEMENTING OF MICROSOFT AZURE MACHINE LEARNING TECHNOLOGY FOR ELECTRIC MACHINES OPTIMIZATION
,
2019
.