Machine Condition Monitoring System Based on Edge Computing Technology

Area of Science:

• Industrial Engineering
• Computer Science
• Artificial Intelligence

Background:

• Effective monitoring of production equipment is crucial for minimizing downtime and optimizing maintenance schedules.
• Traditional monitoring methods often lack real-time processing capabilities and advanced analytical insights.
• Integrating artificial intelligence (AI) and edge computing (EC) offers a promising approach to enhance equipment diagnostics.

Purpose of the Study:

• To design and evaluate a novel system for assessing the condition of production equipment.
• To leverage artificial intelligence (AI) and expert systems (ESs) with edge computing (EC) for real-time sensor data processing.
• To develop a robust solution for predictive maintenance through integrated sensor subsystems and data analysis.

Main Methods:

• Development of a sensor subsystem for monitoring selected production process parameters.
• Implementation of data processing algorithms utilizing AI and expert systems (ESs) at the edge (EC).
• Design of a communication infrastructure and database for storing triggers and facilitating predictive maintenance.
• Laboratory testing and experimental operation of the designed system.

Main Results:

• Successful integration of AI and EC for direct, on-site processing of sensor data.
• Generation of triggers in a database for effective predictive maintenance.
• Demonstration of the system's capability to evaluate equipment condition in real-time.
• Development of an interactive online map for visualizing production system status.

Conclusions:

• The designed system effectively enhances equipment condition evaluation through AI-driven, edge-processed sensor data.
• The integration of AI, ESs, and EC provides a powerful framework for predictive maintenance in industrial settings.
• The developed solution offers a significant advancement in real-time monitoring and operational status visualization for production systems.