Intellectual Property India Publishes Patent Application for Deep Learning Based Robust Method For Electricity Power Consumption Prediction

MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202621069464 A) filed by Kanchan Ashish Khedikar; Piyush Kumar Pareek; and Nitte Meenakshi Institute Of Technology, Nitte Deemed To Be University on June 03, 2026, for Deep Learning Based Robust Method For Electricity Power Consumption Prediction.

Inventors include Kanchan Ashish Khedikar; Piyush Kumar Pareek; and Nitte Meenakshi Institute Of Technology, Nitte Deemed To Be University.

The application for the patent was published on July 24, 2026, under issue no. 30/2026.

Abstract: ABSTRACT OF THE INVENTION Title: Deep Learning Based Robust Method for Electricity Power Consumption Prediction The present invention discloses an intelligent, deep learning-driven system for forecasting 350 electricity power consumption with high precision, tailored for rural and semi-urban energy distribution networks. The system integrates a hybrid modeling approach using Long Short-Term Memory (LSTM) networks for short-term temporal dependencies and Facebook Prophet for trend-seasonality forecasting, applied over a preprocessed and feature-engineered dataset. 355 The invention is capable of handling complex temporal patterns, high-cardinality categorical variables, and missing data scenarios, thereby making it suitable for real-world electricity board datasets. Feature engineering modules embedded within the system extract domain-relevant signals from input variables, while anomaly detection components flag unusual consumption behavior that may indicate power theft, faulty meters, or data inconsistencies. 360 The framework has been trained on actual consumer-level electricity usage data from the Sankh subdivision, Sangli Circle, Maharashtra (2021–2023), incorporating attributes such as Consumer Number, Tariff Code, DTC Code, Connection Date, and Unit Consumption (KWH). It supports consumer-specific and seasonal prediction, continuous model retraining, and real-time forecasting capabilities—facilitating proactive grid management, efficient 365 resource allocation, and fraud mitigation. This invention offers a scalable, modular, and interpretable solution for electricity utilities to enhance operational intelligence, regulatory compliance, and digital transformation across smart grid infrastructures.

Disclaimer: Curated by HT Syndication.