Load Management Solutions for Electric Vehicle Chargers in a Software as a Service Platform

Abstract

Electric vehicle charging poses a problem to the current electrical system infrastructure. The total energy consumption can rise to an exceptionally high level when multiple EV charging stations are using high power simultaneously. Electrical system overload is risky because it can lead to power outages and high-cost electrical bills. The problem can be resolved by upgrading hardware on the charging site. However, it is not cost-efficient.

This thesis presents an alternative solution by utilizing load management. The main goal of this project was to design and implement load management solutions for electric vehicle chargers in the cloud platform. Solutions were developed utilizing microservice architecture, and various technologies were used: Docker, Kubernetes, React.js, Node.js, Express, Redis, PostgreSQL, and TimescaleDB. Power consumption on the site was measured with an external hardware module. Solutions support chargers that operate on Open Charge Point Protocol version 1.6 JSON.

Dynamic Load Management distributes available power evenly among all the active charging stations on the site. Adaptive Load Management reacts to the real-time change in energy consumption on the site and distributes available power accordingly to the charge points. Both solutions were continuously tested with genuine chargers and electric cars. Upon test results, needed changes were implemented.

This thesis also covers current implementation disadvantages and suggests possible improvements.