MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641101073 A) filed by Meenakshi Sundararajan Engineering College on August 21, 2026, for Predictive Maintance System With Edge Ai.
Inventors include Sudhan Vikram K; Naveen Gb; Pragashree B; Sujith Kumar S; Tarun C; Vinoth Kumar R; Selvarani N; Nadhiya S; Md Nooruzzaman Khan; Sengole Elavarasi K; Dr K Amutha; and P. Ajitha.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: Abstract Predictive Maintenance System With Edge Al The increasing adoption of industrial automation has created a growing need for intelligent monitoring systems capable of detecting equipment faults before they result in machine failure. Conventional maintenance methods primarily rely on periodic inspections or reactive maintenance, which often lead to unexpected breakdowns, production downtime, increased maintenance costs, and safety risks. To address these challenges, this work presents an AI-Based Acoustic Monitoring System for Predictive Maintenance that combines acoustic sensing, digital signal processing, machine learning, and IoT- based remote monitoring for real-time machine health assessment. The proposed system continuously captures acoustic signals generated by industrial machinery using a high-sensitivity MEMS microphone. The acquired signals are preprocessed to remove noise and converted into the frequency domain using Fast Fourier Transform (FFT). Relevant spectral features are extracted and analyzed using a trained machine learning model to distinguish between normal and abnormal operating conditions. The model identifies potential faults by recognizing deviations in machine sound patterns, enabling early fault detection before severe mechanical damage occurs. The system is implemented using an ESP32 microcontroller, which performs real-time data acquisition, signal processing, and machine learning inference. When an abnormal condition is detected, the system activates visual and audible alerts through an OLED display, RGB LED indicators, and a buzzer. Simultaneously, machine health data and prediction results are transmitted to a cloud dashboard via Wi-Fl using the MQTT communication protocol, enabling continuous remote monitoring and predictive maintenance. The proposed system provides a compact, low-cost, and scalable solution for industrial condition monitoring. By integrating acoustic analysis, artificial intelligence, embedded systems, and IoT communication, the invention reduces unplanned downtime, lowers maintenance costs, improves equipment reliability, and enhances operational safety. The system is suitable for monitoring motors, bearings, pumps, compressors, HVAC systems, elevators, conveyor systems, and other rotating industrial machinery.
Disclaimer: Curated by HT Syndication.
