Javascript must be enabled for the correct page display

Developing EV Charge Station Health Indicators to Support Proactive Maintenance and Troubleshooting at GreenFlux

Mubarok, Mohammad Wahyu Syafiul (2026) Developing EV Charge Station Health Indicators to Support Proactive Maintenance and Troubleshooting at GreenFlux. Master's Internship Report, Energy and Environmental Sciences.

[img]
Preview
Text
mEES2026MohammadWahyuSyafiulMubarok.pdf

Download (2MB) | Preview

Abstract

This internship thesis investigates how operational data from electric-vehicle charging infrastructure can support more efficient maintenance and troubleshooting. Using anonymized CPMS/OCPP data, including charger status notifications, session records, fault-related signals, telemetry, and asset metadata, the study develops separate station-level and connector-level analytical views for representative AC and DC chargers. The methodology combines data auditing, issue-family construction, feature engineering, recurrence analysis, rule-based screening, and interpretable baseline modelling evaluated through precision-recall metrics and lift. The results show that raw error frequency alone is insufficient: signal relevance depends on charger type, entity level, vendor diagnostics, recurrence, and session impact. Several issue families demonstrate useful ranking value for maintenance review, but the evidence does not support autonomous physical root-cause diagnosis. The thesis therefore recommends that predictive maintenance should be developed in stages, beginning with explainable uptime intelligence: issue-family dashboards, review queues, reason-code explanations, and structured feedback loops. This approach offers GreenFlux a feasible path from reactive monitoring toward more reliable, data-driven maintenance decision support.

Item Type: Thesis (Master's Internship Report)
Supervisor name: Medjroubi, W. and Laugs, G.A.H.
Degree programme: Energy and Environmental Sciences
Thesis type: Master's Internship Report
Language: English
Date Deposited: 16 Jul 2026 14:51
Last Modified: 16 Jul 2026 14:51
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/37440

Actions (login required)

View Item View Item