Toward Predictive Maintenance for Aerospace Hot Press Furnace and the Current Monitoring With Machine Learning

• Published in: IEEE Access
• Main Authors: Chandra Wijaya, Jing-Zhang Liu, I-Jan Wang, Jen-Kai King, Chao-Tung Yang
• Format: Article
• Language: English
• Published: IEEE 2025-01-01
• Subjects: Autoclave; energy audit; LSTM; PHM; RNN
• Online Access: https://ieeexplore.ieee.org/document/11052301/

In this research, we propose a Prognostics and Health Management (PHM) system that integrates an energy monitoring framework to enhance the efficiency of maintenance strategy of aerospace autoclaves, which are usually subjected to expensive and complicated inspection processes. It provides real-time health indicators, predicts the condition of equipment, and performs energy audits. A health evaluation logic was built upon the current balance through the analysis of the three-phase current fluctuation amount of heater units and circulation fan motors. Using a main-sub health assessment structure, the PHM system processes data from individual equipment units, identifies anomalies, and notifies maintenance personnel to carry out corrective actions. It uses complex AI techniques such as RNN, GRU, LSTM, etc., for trend prediction of health for proactive maintenance. It also accounts for energy consumption and carbon emissions per batch of production, thereby contributing to sustainability in the aerospace industry. Ultimately, this implementation makes autoclaves more reliable, longer-lasting, and more efficient, while minimizing maintenance costs, as well as energy consumption and carbon emissions. The end-to-end solution offers a powerful means of enhancing autoclave performance, and of enabling green aerospace operations.