MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621050232 A) filed by Yuvraj Bharade; Dr. Ankush Kudale; and Dr. Chandrani Singh on April 20, 2026, for A System And Method For Real-Time Adaptive Feature Extraction Using Evolving Modular Pipes And Thermodynamic Regulation.
Inventors include Yuvraj Bharade; Dr. Ankush Kudale; Dr. Chandrani Singh; Dr. Bhagwat Kaulwar; Shobha Mishra; Dr. Ravinkant Zirmite; Dr. Dattatray Ishwar Navale; and Meenakshi Jadhav.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT A System and Method for Real-Time Adaptive Feature Extraction using Evolving Modular Pipes and Thermodynamic Regulation A computer- implemented Thermo-Fluidic Evolutionary Network (TFEN) system for real-time adaptive feature extraction from high-entropy, non-deterministic data streams is disclosed. The system continuously ingests dynamic data and computes thermodynamic metrics including a Kinetic Heat scalar representing information entropy and volatility, and a Global Gravity vector field representing macro-stochastic forces. An Active Pipe Matrix contains multiple autonomous Adaptive Logic Pipes, each having fixed heuristic anchors and an adaptive elastic layer. When the Kinetic Heat exceeds a flexibility threshold, a Topological Fork mechanism clones and mutates pipes, while an XOR Synthesis Gate resolves logical contradictions by generating temporary catalyst pipes. A Regime-Tagged Pipe Vault archives successful pipes to prevent catastrophic forgetting. A Mutation Authenticity Gate and Survival-of-the-Fittest Pruning Protocol ensure quality and resource optimization. The Tessellated Tree-Matrix Architecture (TTMA) 3D Chromatic Matrix Interface provides intrinsic explainability by mapping evolutionary history, confidence levels, and emerging logic clusters in real time. The system enables dynamic structural adaptation, complete lineage-based transparency, and robust performance in highly non-stationary environments without requiring retraining.
Disclaimer: Curated by HT Syndication.
