MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202641118772 A) filed by Srm Institute Of Science And Technology, Ramapuram Campus; and Easwari Engineering College on October 05, 2026, for Self-Powered, Non-Intrusive Grid-Edge Analytics System For Harmonic-Resolved Load Disaggregation And Autonomous Billing Audit.
Inventors include Mr. Laksman D; Dr. R. Subhashini; Dr. V. R. Nagalakshmi; and Dr. G. V. Shrichandran.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: ABSTRACT This paper presents a clamp-on analytics device designed for low-voltage residential service lines, specifically optimized for the spatial constraints of dense urban environments. The hardware installs directly around an insulated phase conductor, utilizing a split-core design with high-permeability materials influenced by flux-gate principles. By employing distinct windings for sensing and energy scavenging, the system achieves a micro-watt cold start and maintains stable milliwatt- range operation from primary line currents as low as 5 A—completely eliminating the need for external power or batteries. An integrated micro-power management circuit conditions this harvested energy to drive a microcontroller. This controller digitizes current waveforms at a minimum sampling rate of 10 kHz and performs fixed-point harmonic analysis up to the 15th order. These datasets feed a quantized Long Short-Term Memory (LSTM) neural network running natively on the chip. This LSTM engine executes on-device non-intrusive load monitoring (NILM), identifying specific household appliance states and their respective power consumption. Additionally, a tariff engine tracks usage against multi-tiered billing structures in real-time. Operating in tandem, a billing audit subsystem retrieves DLMS/COSEM data from utility management servers to cross-reference official meter readings with locally integrated energy data. By aligning these datasets, the system can autonomously detect meter drift, phase mis assignment, and potential energy theft by flagging discrepancies over consecutive intervals. Unlike conventional smart meters, this solution integrates parasitic harvesting, high-order NILM on resource-constrained hardware, and autonomous utility API auditing into a single, latency tolerant edge instrument.
Disclaimer: Curated by HT Syndication.
