MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202631102158 A) filed by National Institute Of Technology Rourkela Nit Rourkela on August 24, 2026, for Hybrid Time-Varying Ensemble Method For Real-Time Air Quality Index Forecasting And A System Thereof.
Inventors include Sahu, Baikuntha; Panigrahi, Prof. Sibarama; Sa, Prof. Pankaj Kumar; and Sahoo, Prof. Manmath Narayan.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: HYBRID TIME-VARYING ENSEMBLE METHOD FOR REAL-TIME AIR QUALITY INDEX FORECASTING AND A SYSTEM THEREOF The disclosure relates to a method (900) and system (300) for real-time air quality index (AQI) forecasting and closed-loop control of an air purification system. The method (900) comprises receiving (902) pollutant measurements from environmental sensors (106), generating (904) an AQI time series using a processing unit (112) of a cloud server (110), decomposing (906) the AQI time series using Variational Mode Decomposition (VMD) into intrinsic mode functions (IMFs) and a residual component, processing (908) the IMFs and residual component using heterogeneous prediction models (208) to generate prediction outputs, dynamically aggregating (910) the prediction outputs using Kalman Filter-based time-varying weighting (206) to generate a predicted AQI, determining (912) an air purification strategy, transmitting (914) a corresponding control command to an actuator controller (318), operating (916) purification components of the air purification system (126), and receiving (918) operational feedback through an electrical control interface (320) for closed-loop control. Figures 3, 9
Disclaimer: Curated by HT Syndication.
