DataPortia

DataPortia is a self-hosted industrial data acquisition and AI analysis platform that connects to automation systems via OPC UA protocol. It runs entirely on-premises — your data never leaves your infrastructure. AI Analysis Features DataPortia integrates with locally-running Ollama LLM models to provide AI-powered analysis of industrial time-series data: - Anomaly Detection — Z-score based anomaly identification with severity classification. Automatically detects unusual patterns in sensor data, temperature readings, power consumption, and other industrial measurements. - Forecasting — Trend and periodicity analysis with future projections based on historical patterns. - Cost Optimization — AI identifies energy waste, inefficiencies, and potential savings from production and utility data. - Natural Language Queries — Ask questions about your data in plain language (Finnish, English, Swedish, or German). - Structured Analysis Reports — Automated 6-section comprehensive reports on operational data. - Multi-turn Chat — Continue conversations about your data with full context retention. You choose the AI model — any Ollama-compatible model works (Gemma, Llama, Mistral, DeepSeek, Qwen, etc.). Analysis runs fully locally with no cloud dependency. Thinking models (DeepSeek-R1, Qwen3, QwQ) are supported with automatic think-block filtering. Industrial Data Platform Beyond AI analysis, DataPortia provides a complete industrial data management solution: - Multi-connection OPC UA data acquisition (10+ simultaneous connections with redundancy) - Real-time dashboard with gauges, charts, bar charts, and tables - Interactive trend visualization with ECharts and color customization - OPC UA Alarms & Conditions monitoring, analytics, and export - Comprehensive CSV and PDF reporting - Automated report scheduling (daily, weekly, monthly, custom intervals) -OPC UA history read from server historians - Calculation circuits (cumulative and non-cumulative formulas) - Tag data transfer, copy, and merge between connections - TimescaleDB time-series database with compression and retention policies -Continuous aggregates (minute, hourly, daily, weekly) for fast queries -Automated database backup and restore with scheduling - Role-based user management (local accounts or Azure AD) - HTTPS with automatic certificate management - 4-language web interface: Finnish, English, Swedish, German Use Cases - Factory energy monitoring and optimization - District heating network analysis - Power plant performance tracking - Process industry automation data acquisition - Machine tool monitoring - Water treatment facility management - Waste incineration plant monitoring Licensing - 30-day free trial included - Perpetual licenses by connection count - Floating licenses for multi-site deployments - AI analysis add-on license System Requirements - Windows 10/11 / Windows Server 2016+ or Linux - PostgreSQL 18 + TimescaleDB (included in Windows setup) - Minimum 8 GB RAM (16 GB+ recommended) - Minimum 100 GB disk space (depending on data volume) - Ollama for AI analysis (separate installation)