MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123178 A) filed by Shree Venkateshwara Hi-Tech Engineering College; Dr. V. Saminathan; Dr. R. S. Kamalakannan; Mr. K. Balakrishnan; and Mr. K. S. Vivek, Erode, Tamil Nadu, on Dec. 6, 2025, for 'deep learning-based predictive maintenance system for electronics industry.'
Inventor(s) include Dr. V. Saminathan; Dr. R. S. Kamalakannan; Mr. K. Balakrishnan; and Mr. K. S. Vivek.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The rapid advancement of automation and Industry 4.0 has intensified the need for intelligent maintenance strategies in the electronics manufacturing sector, where unplanned equipment failures can cause production losses exceeding 15-20% annually. This study proposes a Deep Learning-Based Predictive Maintenance System (DL-PMS) tailored for the electronics industry to minimize downtime and enhance operational efficiency. The system integrates sensor data (temperature, vibration, acoustic signals, and current consumption) collected at 1 Hz frequency from 25 surface-mount technology (SMT) production units over 6 months. A hybrid deep learning model combining Convolutional Neural Networks (CNNs) for feature extraction and Long Short-Term Memory (LSTM) networks for temporal pattern recognition was developed. Experimental results demonstrate that the proposed DL-PMS achieves an accuracy of 96.8% in fault classification and a Mean Absolute Error (MAE) of 0.037 in remaining useful life (RUL) prediction, outperforming conventional Random Forest and SVM-based models by 9.2% and 11.5%, respectively. Furthermore, early fault detection reduced unplanned machine stoppages by 27%, translating to a 12.4% improvement in production throughput. The model's adaptive learning capability ensures robustness under varying operational conditions, including temperature fluctuations of 5 C and vibration deviations up to 8%. The system was deployed on a cloud-based platform using TensorFlow 2.14 and AWS IoT Core, supporting real-time anomaly alerts with an average response latency of 0.42 seconds. The implementation results validate the feasibility of integrating deep learning for proactive maintenance in high-speed electronic manufacturing environments. In conclusion, the proposed DL-PMS demonstrates a scalable, data-driven approach to predictive maintenance, offering measurable benefits in productivity, equipment reliability, and lifecycle cost reduction - essential for sustaining competitiveness in the modern electronics industry."
Disclaimer: Curated by HT Syndication.
