PCA and Health Indicators: Predicting Machine Failures Through Resistance Analysis

Predictive maintenance is crucial for ensuring industrial equipment's reliability and operational efficiency. This research aims to develop accurate health indicators to monitor real-time equipment conditions based on current signals. The methodology involves several key stages: collection of degradation data in current signals, data processing and mining, analysis using Principal Component Analysis (PCA), and development of health indicators. This study presents a comprehensive approach to converting raw degradation data into meaningful health indicators for effective engine prognostics and health management (PHM). Leveraging current signal data, we apply data mining and processing techniques to extract statistically significant features, including Standard Deviation, Peak to Peak, Root Mean Square (RMS), Crest Factor, Impulse Factor, Margin Factor, and Kurtosis. PCA is then used to reduce the dimensionality of the processed data, highlighting the principal components that capture the most significant variance indicating the machine's health. The resulting health indicators, derived from PCA, show a clear correlation between changes in additional load and increasing trends of PCA components and health indicators, thus validating the effectiveness of this approach in monitoring and predicting machine conditions. This methodology provides a robust real-time machine health assessment framework, facilitating timely maintenance and reducing the risk of unexpected failures. The results show that increasing resistance over time (t) leads to improved health indicators in a nonlinear manner, providing valuable insights for timely intervention before critical failure occurs. This analysis demonstrates a strong correlation between daily incremental resistance changes and machine condition as monitored by PCA and health indicators. Consistent upward trends in PCA scores and health indicators validate the effectiveness of this technique in tracking engine health under varying resistance conditions.