Open AccessDetermination of power engineering equipment’s defects in predictive analytic system using machine learning algorithms
Author(s) -
Ivan Shcherbatov,
G N Turikov
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1683/4/042056
Subject(s) - machine learning , random forest , predictive power , logistic regression , artificial intelligence , computer science , gradient boosting , boosting (machine learning) , algorithm , power (physics) , philosophy , epistemology , physics , quantum mechanics
An approach to forecasting of power equipment’s defects and failures is considered in this paper. The main operations of predictive analytic system for forecasting turbine’s regulation system’s defects is represented. Special attention on machine learning models tuning for explored forecast problem is paid. The following machine learning algorithms: Logistic Regression, Random Forest, Extreme Gradient Boosting, ensembles of these algorithms are explored. According to the results of optimal machine learning models determined, their comparison is done and the conclusion on the appropriateness of the use in predictive analytic system is made.