Building a cloud-based EV charging monitoring platform with real-time AI analytics

This article describes EVBrain, a cloud-based platform built on AWS by Iberdrola-BP Pulse and GaleoTech for monitoring EV charging infrastructure across Spain with real-time AI analytics.

• Processes OCPP messages in real-time from hundreds of charging points with sub-minute latency
• Automatically detects 26 distinct incident types including connector faults and hardware anomalies
• Increased platform uptime from 82% to 97% through proactive incident detection
• Serverless architecture using IoT Core, SQS, Lambda, DynamoDB, Aurora, and Tinybird
• AI-powered assistant using Amazon Bedrock for natural language queries on incidents and metrics
• Extends monitoring to Low Voltage Cabinets powering charging stations via Modbus TCP
• Automatically remediates 80% of incidents before escalation to field services
• Handles 500,000+ OCPP messages daily; stress-tested to 25 million messages per day
• Monitors 1,152 charging points with 1,973 connectors; projects 10,000-12,000 by 2030

EVBrain demonstrates how serverless AWS architecture enables reliable, scalable EV charging network operations through real-time processing, advanced analytics, and integrated AI capabilities.