AI Assisted Renewable Energy Performance Monitoring and Predictive Analytics Using ESP32

This paper presented an AI-assisted renewable-energy performance monitoring architecture based on a 12 V solar panel, calibrated sensors, an ESP32, USB serial communication, Python analytics, and a Flask dashboard. The design measures electrical and environmental conditions, preserves raw and calibrated records, and computes power and energy only from time-aligned valid samples. The predictive workflow compares persistence, linear, tree, ensemble, and kernel regression methods using fixed features, chronological folds, and common metrics. The evaluation intentionally separates model development from the final test period and records all configuration needed for reproduction. Experimental accuracy will be reported only after hardware calibration and representative data collection are complete. Future work can extend the acquisition period across seasons, add calibrated irradiance and weather inputs, estimate forecast intervals, detect model drift, and evaluate transfer between panels. These additions should retain the same traceability principles so that changes in hardware, data, and models remain distinguishable.

1. Espressif Systems, ESP32 Series Datasheet and ESP-IDF ADC Calibration Driver Documentation, 2026. 2. A. P. Dobos, PVWatts Version 5 Manual, National Renewable Energy Laboratory, NREL/TP-6A20-62641, 2014. 3. B. Marion et al., Performance Parameters for Grid-Connected PV Systems, National Renewable Energy Laboratory, 2005. 4. National Renewable Energy Laboratory, PVWatts Calculator and Photovoltaic Data Tools, accessed 2026. 5. Scikit-learn Developers, TimeSeriesSplit, Model Selection, and Regression Metrics Documentation, 2026. 6. Pallets Projects, Flask Documentation 3.1, Application Factories, Blueprints, Security, and Deployment, 2026. 7. L. Breiman, Random Forests, Machine Learning, vol. 45, pp. 5-32, 2001. 8. J. H. Friedman, Greedy Function Approximation: A Gradient Boosting Machine, Annals of Statistics, vol. 29, no. 5, 2001. 9. C. Cortes and V. Vapnik, Support-Vector Networks, Machine Learning, vol. 20, pp. 273-297, 1995. 10. T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd ed., Springer, 2009.