MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202521125790 A) filed by Yugandhara Ganesh Kasture; and Dr. Sushree Sangita Mishra, Maharashtra, on Dec. 12, 2025, for 'system and method for predictive modelling of seismic-sloshing dynamics in overhead water tanks of buildings using recurrent neural network integrated with harmony search optimization.'
Inventor(s) include Yugandhara Ganesh Kasture; and Dr. Sushree Sangita Mishra.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "The invention provides a system and method for real-time predictive modelling of sloshing dynamics in overhead water tanks under seismic loading, integrating sensor fusion (accelerometers, ultrasonics, strains), RNN (LSTM-GRU stack), and harmony search optimization for parameter tuning. Features extracted at 200 Hz (frequencies 0.5-5 Hz, phases, spectra) feed bidirectional RNN yielding slosh height/pressure predictions (error 8% RMSE). HSO (HM=40, HMCR=0.9) minimizes loss balancing accuracy and effort, driving PID control of dampers reducing peaks 40%. Edge-deployed (Jetson, 100 ms), validated on Indian spectra, achieves 25x speedup vs. CFD with 94% correlation, industrially applicable for retrofitting urban infrastructure enhancing seismic resilience per NBC 2016."
Disclaimer: Curated by HT Syndication.
