This paper presents a uniquely designed Smart Energy Monitoring and Control System that integrates Internet of Things (IoT), Artificial Intelligence (AI), and GSM communication to create an energy-efficient and predictive home automation solution. The system continuously monitors household electrical parameters using ACS712 and ZMPT101B sensors and transmits data via an Arduino Uno. A standout feature is its offline-capable design, where a GSM module (SIM900A) allows users to control appliances and receive alerts through SMS without the need for internet. In addition, a powerful Long Short-Term Memory (LSTM) neural network model is implemented to forecast future energy consumption using real-time data stored in a local MySQL database. The system includes a fully functional website with user login, real-time monitoring dashboard, prediction graphs, and a feedback module, providing an interactive user experience. This modular, low-cost, and intelligent system is specially developed for regions with limited infrastructure and has demonstrated excellent performance in real-time conditions.
Introduction
With rising household energy consumption due to more electrical appliances, traditional meters fail to provide actionable insights, leading to inefficiency and higher bills. This project introduces an affordable, offline-capable energy management system designed for Indian homes, combining real-time monitoring, AI-based usage prediction, and remote appliance control via SMS.
Key Components & Workflow:
Uses Arduino Uno with current (ACS712) and voltage (ZMPT101B) sensors, relay modules for appliance control, and a GSM module (SIM900A) for SMS communication.
Data collected is processed locally through Node-RED and stored in a MySQL database.
An LSTM deep learning model predicts next-day energy consumption, aiding users in managing and preventing overloads.
Users control appliances remotely via SMS or a web dashboard that shows live, historical, and forecasted energy data.
Innovations:
Operates without internet relying on SMS and local servers.
Combines monitoring, prediction, and control in one cost-effective, user-friendly system.
Web interface includes registration, login, live dashboard, prediction visualization, and feedback collection to improve the model and user experience.
Results:
Sensor accuracy: ~92%
GSM control responsiveness: ~98%
Prediction accuracy (LSTM): ~93%
Dashboard updates with <1 second latency
The system is scalable, secure, and improves energy efficiency by empowering users with insight and control.
Conclusion
This paper presented an innovative and self-reliant energy monitoring system tailored for real-world conditions. The integration of IoT sensors, GSM control, and AI-based prediction offers a comprehensive and intelligent approach to smart energy management. Unlike conventional systems, this project emphasizes modular design, affordability, and adaptability, making it highly suitable for environments with limited or no internet access. The paper provides a clear understanding of the exact functioning of the system, detailing how real-time energy monitoring is achieved through sensor integration and micro-controller processing. It also highlights the application of LSTM in forecasting energy consumption, allowing users to make informed and proactive decisions. The use of SMS-based control adds significant value in rural or connectivity-constrained areas by enabling remote access without relying on the internet. Additionally, the design showcases how user-centric considerations, such as privacy through encrypted credentials and a structured feedback mechanism, enhance trust and interaction. Overall, the system bridges the gap between technology and accessibility, promoting efficient energy use while ensuring user engagement and data security. Future enhancements aim to incorporate solar energy data, intelligent fault detection using auto encoders, mobile application interfaces, and scalable cloud-based access to support a broader user base.
References
[1] Gungor, V. C., Sahin, D., Kocak, T., Ergut, S., Buccella, C., Cecati, C., & Hancke, G. P., \"Smart Grid Technologies: Communication Technologies and Standards,\" IEEE Transactions on Industrial Informatics, vol. 7, no. 4, pp. 529–539, Nov. 2011.
[2] Rault, T., Bouabdallah, A., &Challal, Y., \"Energy Efficiency in Wireless Sensor Networks: A Top-Down Survey,\" Computer Networks, vol. 67, pp. 104–122, 2014.
[3] Hochreiter, S., &Schmidhuber, J., \"Long Short-Term Memory,\" Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.
[4] Zhao, B., Zhang, X., & Zhao, J., \"A Review on Energy Management in Smart Grids,\" Renewable and Sustainable Energy Reviews, vol. 70, pp. 896–911, 2017.
[5] Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., ... & Zheng, X., \"TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems,\" 2016. [Online]. Available: https://www.tensorflow.org
[6] DFRobot, \"ACS712 Current Sensor,\" DFRobot Wiki, 2021. [Online]. Available: https://wiki.dfrobot.com/ACS712
[7] DFRobot, \"ZMPT101B Voltage Sensor Module,\" DFRobot Wiki, 2021. [Online]. Available: https://wiki.dfrobot.com/ZMPT101B
[8] Baccour, N., Koubaa, A., Mottola, L., Zuniga, M., Youssef, H., Boano, C. A., & Alves, M., \"Radio Link Quality Estimation in Wireless Sensor Networks: A Survey,\" ACM Transactions on Sensor Networks, vol. 8, no. 4, pp. 1–33, 2012.
[9] Goodfellow, I., Bengio, Y., & Courville, A., \"Deep Learning,\" MIT Press, 2016. [Online]. Available: https://www.deeplearningbook.org
[10] Mohsin, M., & Kumar, N., \"A Review of IoT-Based Smart Energy Metering Systems,\" International Journal of Advanced Computer Science and Applications, vol. 11, no. 5, pp. 123–130, 2020.