AI-Powered Machine Learning Approaches for Fault Diagnosis in Industrial Pumps

TL;DR

This paper introduces a machine learning framework using sensor data and synthetic fault signals for early fault detection in industrial pumps, demonstrating high accuracy and real-time applicability for proactive maintenance.

Contribution

It presents a hybrid approach combining dual-threshold labeling and synthetic fault injection, improving early fault detection in industrial pump systems with scalable ML models.

Findings

Random Forest and XGBoost achieved high accuracy in fault classification.

The method effectively detects rare and emerging faults.

The framework is suitable for real-time deployment and adaptable to other machinery.

Abstract

This study presents a practical approach for early fault detection in industrial pump systems using real-world sensor data from a large-scale vertical centrifugal pump operating in a demanding marine environment. Five key operational parameters were monitored: vibration, temperature, flow rate, pressure, and electrical current. A dual-threshold labeling method was applied, combining fixed engineering limits with adaptive thresholds calculated as the 95th percentile of historical sensor values. To address the rarity of documented failures, synthetic fault signals were injected into the data using domain-specific rules, simulating critical alerts within plausible operating ranges. Three machine learning classifiers - Random Forest, Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM) - were trained to distinguish between normal operation, early warnings, and critical alerts. Results showed that Random Forest and XGBoost models achieved high accuracy across all classes, including minority cases representing rare or emerging faults, while the SVM model exhibited lower sensitivity to anomalies. Visual analyses, including grouped confusion matrices and time-series plots, indicated that the proposed hybrid method provides robust detection capabilities. The framework is scalable, interpretable, and suitable for real-time industrial deployment, supporting proactive maintenance decisions before failures occur. Furthermore, it can be adapted to other machinery with similar sensor architectures, highlighting its potential as a scalable solution for predictive maintenance in complex systems.