Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM69280
Volume:
10 |
11 |
Nov 2026
Pages:
1-6
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Engineering and Technology

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Engineering and Technology

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Engineering and Technology

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Engineering and Technology

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Engineering and Technology
Households usually learn how much electricity they used only when the monthly bill arrives, by which time wasteful usage can no longer be corrected. This paper presents an AI-powered energy consumption optimizer, a web-based platform that tracks appliance-level usage, predicts the monthly bill under the Karnataka Gruha Jyothi tariff rules, and enforces a daily energy budget. Four features distinguish the work: 1) an AI Budget Guard that derives a daily allowance from the same month of the previous year, reserves energy for essential appliances, and divides the rest among the other appliances by priority-weighted shares; 2) a reward wallet that grants one point for every 10 kWh saved against the previous year, each point reducing the payable bill by ₹1; 3) a Random Forest recommendation model trained on 11,026 synthetic records derived from three monitored homes, which reached 88.40% test accuracy on three action classes; and 4) EnergyBot, a retrieval-augmented chatbot built on Hugging Face embeddings, MongoDB Atlas Vector Search, and a Groq-hosted language model. The system is implemented with React, Flask, and MongoDB, uses OpenWeatherMap data, and demonstrates appliance control through a simulated ESP32 controller in Wokwi.
KEYWORDS
This paper presented an AI-powered energy consumption optimizer that predicts bills under the Karnataka Gruha Jyothi tariff rules, allocates a daily energy budget by appliance priority, rewards savings against the previous year, recommends appliance actions with a Random Forest, and answers questions through a RAG chatbot. A simulated ESP32 controller demonstrates the link between software decisions and appliance outputs. The Random Forest reached 88.40% test accuracy on synthetic data, and further validation with measured data and a pilot deployment are the next steps. The proposed system provides an intelligent and efficient approach to monitoring and optimizing household energy consumption using AI. Future work includes integrating smart meters and physical ESP32 hardware with properly rated switching components, validating the models on real household data, handling class imbalance (for example, through class weighting), comparing deep forecasters such as the LSTM reported in [3], updating tariffs from verified sources, adding multilingual (including Kannada) chatbot support and a mobile application, and running a pilot with real households.
[1] N. Raj, S. Bhardwaj, A. Sinha, and A. L. Yadav, “AI-powered energy consumption optimization for smart homes using IoT,” in Proc. Int. Conf. Comput. Intell. Commun. Appl. (ICCICA), Samalkha, India, 2024, pp. 231–236, doi: 10.1109/ICCICA60014.2024.10585239. [2] P. Biswas, A. Rashid, A. Biswas, M. A. A. Al Nasim, S. Chakraborty, K. D. Gupta, and R. George, “AI-driven approaches for optimizing power consumption: A comprehensive survey,” Discover Artif. Intell., vol. 4, no. 1, Art. no. 116, 2024, doi: 10.1007/s44163-024-00211-7. [3] A. M. Durrani, A. U. Asar, A. Aziz, W. Khan, M. Z. Yousaf,U. Farooq, M. Altayeb, and M. S. Geremew, “AI-driven optimization of energy consumption and demand response in smart homes,” Energy Explor. Exploit., vol. 44, no. 3, pp. 1382–1419, 2026, doi: 10.1177/01445987251403607. [4] Meta Open Source, “React documentation.” [Online].Available: https://react.dev [5] Pallets, “Flask documentation.” [Online]. Available: https://flask.palletsprojects.com [6] MongoDB, Inc., “MongoDB Atlas Vector Search documentation.” [Online]. Available: https://www.mongodb.com/docs/atlas/atlas-vector-search/ [7] Hugging Face, “Hugging Face documentation.” [Online].Available: https://huggingface.co/docs [8] Groq, Inc., “GroqCloud documentation.” [Online]. Available: https://console.groq.com/docs [9] Espressif Systems, “ESP32 series of SoCs.” [Online].Available: https://www.espressif.com/en/products/socs/esp32 [10] Wokwi, “Wokwi documentation.” [Online]. Available: https://docs.wokwi.com
Publication Date
11/21/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
0
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...