Machine learning models in predicting failures of hydroturbine components

Abstract

This paper explores the application of machine learning models for predicting failures in hydroturbine components within energy systems digital transformation. Various models including Random Forest, XGBoost, SVM, LSTM, and CNN are analyzed for their effectiveness in processing operational data from sensors monitoring parameters such as vibration, temperature, and pressure. The research demonstrates that intelligent models can identify hidden patterns in degradation processes and enhance system reliability in real time.