MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641108269 A) filed by Shashi Kant Mishra; Guna Gayathri Praseetha K; Prakash Shanmurthy; Dr. Veerubotla Bhaskara Murthy; A. Lavanya; E Goma; and Dr. B. Gunasundari on September 09, 2026, for Edge Ai-Based Industrial Iot Fault Prediction System Using Temporal Sensor Patterns.
Inventors include Shashi Kant Mishra; Guna Gayathri Praseetha K; Prakash Shanmurthy; Dr. Veerubotla Bhaskara Murthy; A. Lavanya; E Goma; and Dr. B. Gunasundari.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: ABSTRACT Industrial Internet of Things (IIoT) systems continuously collect sensor data from machines and industrial equipment. Predicting faults at an early stage is important to reduce machine failures, maintenance costs, and production downtime. The proposed system, an Edge AI-Based Industrial IoT Fault Prediction System analyzes temporal sensor patterns for realtime fault detection. The proposed method uses a ID Convolutional Neural Network (1DCNN) combined with Bidirectional Long Short-Term Memory (BiLSTM) to learn important features and temporal patterns from sensor data. Sensor values such as temperature, vibration, pressure, and motor current are normalized using Z-score normalization and converted into sequences using a sliding-window technique. The ID-CNN extracts useful sensor features, while BiLSTM learns the changes and dependencies over time. The trained model is optimized using TensorFlow Lite quantization for deployment on an edge device such as a Raspberry Pi or NVIDIA Jetson Nano. The system predicts normal and faulty machine conditions with low inference latency. The proposed approach provides a fast and efficient solution for real-time industrial fault prediction while reducing the need for continuous cloud processing
Disclaimer: Curated by HT Syndication.
