A Web-Based Smart Real-Time Motor Control Center (MCC) Monitoring and Analytics System for Early Detection of Motor Anomalies in Muara Karang Power Plant

Abstract

This study proposes a comprehensive web-based real-time monitoring and analytics framework for 380V Motor Control Center (MCC) assets at PLTGU Block 1 Muara Karang, addressing the challenge of early identification of motor performance degradation under variable operational conditions. The system integrates a multi-layer architecture comprising sensor-level data acquisition, a real-time processing pipeline, and an analytics layer incorporating statistical modelling, multivariate trend assessment, and anomaly scoring algorithms. An Early Warning System (EWS) is implemented using hybrid rule-based and data-driven thresholds, enabling automated detection of abnormal operating patterns. A historical knowledge engine supports Work Planning and Control (WPC) by modeling temporal degradation signatures, enabling structured diagnostic interpretation and follow-up maintenance actions. The framework further incorporates an event-streaming notification mechanism through Telegram API for low-latency dissemination of contextualized alerts. Validation under field operation demonstrates successful detection of early-stage faults, including a bearing defect in motor 88BT GT1.2 identified before crossing critical failure limits. These interventions prevented unit trips and reduced risk of cascading failures. Experimental results indicate improvements in detection accuracy, diagnostic efficiency, and operational reliability. The proposed framework exhibits strong scalability potential for fleet-wide deployment across PLN’s induction motor systems.