- Harmonic Management with AI for Networks to intelligently enhance EV charging and grid-tied micro-generation at customer sites

This project is developing and using distributed lightweight AI models to analyze harmonics in power networks. The goal is to evaluate Power Quality (PQ) compliance under various future scenarios, such as simultaneous electric vehicle (EV) charging, variations in battery state of charge (SoC), and integration of renewable energy sources like photovoltaic (PV) systems and combined heat and power (CHP) units at customer premises.

Key components include:

- Implementing AI models using the TinyML framework and federated learning for accurate harmonic analysis at EV charging points.

- Developing a smart energy management system (EMS) that optimally schedules charging and discharging events of EVs and battery energy storage systems. It also integrates energy harvesting from PVs and interacts with CHP to manage building energy use, aiming to reduce peak demand.

- Leveraging the TinyML framework to integrate PQ-compliant optimal correction schemes into the EMS. The system maintains Maximum Import Capacity (MIC) by controlling charging rates and implementing cooperative control for simultaneous EV charging in real time.

- Validating the developed network architecture and control scheme using hardware-in-the-loop test setups to assess effectiveness in simultaneous EV charging and harmonic reduction.