What question did this study set out to answer?

The research aims to optimize electricity usage and reduce energy bills through a deep learning approach.

May 6, 20260 citationsOpen Access

CNN-LSTM–Based Electricity Usage Optimization for Bill Reduction Using Climate Factors, Load Type, and Monthly Budget

Key Points

The research aims to optimize electricity usage and reduce energy bills through a deep learning approach.
Developed a CNN-LSTM hybrid architecture for electricity optimization
Analyzed multivariate time-series data including climate factors and load types
Implemented feature engineering and data preprocessing techniques
Used a Flask-based dashboard for real-time predictions and recommendations
Achieved RMSE of 2.31 ₹/kWh and MAE of 1.74 ₹/kWh
R² score of 0.943 indicating strong model performance
Demonstrated approximately 18.7% monthly savings on energy bills

Abstract

This research presents a deep learning–based approach for optimizing electricity usage and reducing energy bills using a hybrid CNN-LSTM architecture. The model analyzes multivariate time-series data, including climate factors, load types, and user-defined monthly budgets, to predict electricity prices and recommend optimal appliance scheduling. The system integrates feature engineering, Min-Max normalization, and sliding window techniques for data preprocessing. Experimental results demonstrate strong performance with RMSE of 2.31 ₹/kWh, MAE of 1.74 ₹/kWh, and R² score of 0.943, achieving approximately 18.7% monthly savings. A Flask-based dashboard provides real-time predictions and budget-aware recommendations, making the system practical for real-world energy optimization.

AIに質問

Bookmark

View Full Paper