MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641097967 A) filed by Sri Eshwar College Of Engineering on August 12, 2026, for Ai Driven Multi-Sensor Fusion System For Pre-Explosion Prediction In Fireworks Manufacturing.
Inventors include Mr. Gowrisanker M; and Gowshalya P.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: The FireSafe AI ecosystem addresses the lack of real-time pre-explosion prediction systems in fireworks manufacturing through a tri-modal sensor fusion framework that integrates mechanical stress analysis, electrostatic discharge prediction, and infrared thermal anomaly detection. The system uses a Random Forest Classifier with SHapley Additive exPlanations (SHAP) to process mechanical parameters such as friction, impact force, and vibration, generating a Mechanical Hazard Index (MHI) based on the pyrotechnic composition of the active batch. A Long Short-Term Memory (LSTM) based Electrostatic and Environmental Prediction Engine analyses Electrostatic Discharge (ESD) voltage, humidity, grounding resistance, and atmospheric conditions to identify potential electrostatic discharge risks before critical levels are reached. An Infrared Thermal Anomaly Detection Engine, using a Vision Transformer (ViT) with Slicing Aided Hyper Inference (SAHI), analyses Long- Wave Infrared (LWIR) thermal images to identify abnormal temperature patterns, hot spots, and possible smoldering conditions. The outputs of all three engines are combined using a Composition-Aware Gated Fusion Logic to generate a unified Pre-Explosion Risk Index (PERI).
Disclaimer: Curated by HT Syndication.
