<![CDATA[Indramayu Steam Power Plant (PLTU Indramayu) is one of the strategic power generation facilities that plays a vital role in supplying national electrical energy. One of the operational issues frequently encountered in the power generation system is Plugging in the Pulverizer, which refers to the blockage of coal flow inside the mill. This condition can lead to reduced combustion performance, increased equipment load, and even unit trip events. Therefore, an early detection and prediction system for Plugging conditions based on operational data is required.This study aims to develop a Plugging forecasting system for the Pulverizer at PLTU Indramayu using the Random Forest Classifier method. The data used in this research were obtained from historical Pulverizer operational records, including mill motor current, coal flow rate, primary air flow rate, and mill outlet temperature. The research stages consist of initial data preprocessing, feature selection, labeling of operating conditions (normal and Plugging), and model testing using the Random Forest Classifier algorithm. Model performance was evaluated using accuracy, precision, recall, F1-score, and cross-validation metrics. The results show that the Random Forest Classifier model is capable of classifying Plugging conditions with high and stable accuracy, indicating its effectiveness as a decision-support tool for predictive maintenance systems. With the implementation of this system, operators are expected to take preventive actions at an earlier stage, thereby minimizing operational disturbances and enhancing the reliability of the power generation system.]]>
Copyrights © 2026
