首页|Machine Learning Approaches for Predicting Excitation Current in Synchronous Motors

Machine Learning Approaches for Predicting Excitation Current in Synchronous Motors

扫码查看
原文链接
  • NETL
  • NSTL
© 2024 Praise Worthy Prize S.r.l.-All rights reserved.The excitation current has a significant impact on the operational efficiency of synchronous motors, which hold a pivotal position in several industrial applications. The main contribution of this study has been to conduct a comprehensive analysis of various machine learning models to make accurate predictions regarding the excitation current. Normalization and feature selection have been applied to the dataset before using various regression models, including Linear, Lasso, Ridge, and Elastic Net Regression, as well as Decision Tree and Random Forest Regression. The Random Forest Regression model demonstrates a test accuracy of 97.01%, indicating its superior performance compared to other models. The good performance of the linear models indicates a linear relationship between the predictors and the excitation current. Regression model has achieved 98% accuracy during training. The findings from this work have extensive implications, encompassing enhanced energy efficiency, extended motor lifespans, and reduced instances of operational interruptions. While the models show promise, it is crucial to recognize their limitations and the need for more research, particularly in the realms of nonlinear models and empirical verification in real-world scenarios.

Data PreprocessingExcitation Current PredictionMachine Learning RegressionOperational EfficiencySynchronous Motors

Marji G.S.

展开 >

Balqa Applied University Alhuson University College

2024

10.15866/iree.v19i3.24432

International review of electrical engineering

International review of electrical engineering

ISSN：1827-6660
年,卷(期)：2024.19(3)