IT / Data integration

Target systems

SCADA, historians, Grafana, cloud, and Python integration patterns.

← Home

This page maps common industrial targets to integration patterns. All paths assume you already have decoded JSON from the LNS or your ingestion service.

Time-series database + Grafana

Fit: Most greenfield IT projects, SMEs, pilots.

  • Ingest MQTT/webhook → PostgreSQL + TimescaleDB (or InfluxDB)
  • Normalize to (time, deveui, metric, value) — see Data model
  • Grafana dashboards: fleet health (storageVoltage, RSSI), process variables, alarm state

Historians (PI System, Aveva, etc.)

Fit: Large industrial sites with existing OT historians.

  • Use MQTT or OPC-UA connector/gateway between your ingestion bus and the historian
  • Map DevEUI + asset ID from your CMMS/EAM registry
  • Separate device health tags from process tags in the historian namespace

SCADA

LNS → integration service → OPC-UA/MQTT → SCADA (Ignition, WinCC, etc.). Branch from your normalized database or real-time bus after decoding.

Cloud (Azure IoT Hub, AWS IoT Core)

  • Forward LNS webhooks to IoT Hub / IoT Core
  • Device twin or registry table for static metadata; telemetry stream for measurements
  • Route rules to Functions/Lambda for normalization and storage (Timestream, RDS, etc.)

Python analytics / ML

  • Read from telemetry_raw or subscribe to Kafka topic fed by ingestion
  • Use pandas for batch analysis; keep decoder version in replay scripts
  • Example consumers (pick one): examples/README.md

Node-RED (prototyping)

  • MQTT in → JSON → function node (normalize) → database or dashboard
  • Fast validation before hardening into Python/Go services

Recommended dashboard views

ViewMetrics
Fleet healthLast seen, storageVoltage, boardTemperature, RSSI/SNR
ProcessPort keys per device config (pt_1, vib_1_rms_hf, …)
Alarmsstatus = 2, alarm source flags, event log
Energy proxythermogenVoltage, baseTemperature gradient indicator