Node-RED

Node-RED#

Node-RED is a flow-based, low-code development tool commonly used in IoT and industrial automation. It allows users to connect data sources, processing nodes, and outputs using a visual editor, making it ideal for rapidly building streaming pipelines without writing extensive boilerplate code.

In the STAMM demo, Node-RED acts as a data replay service: instead of directly connecting to a physical bioreactor, it replays historical trajectories from IndPenSim and publishes them as time-series data, so that STAMM receives them as if they were coming from a live penicillin fermentation.

Node-RED data replay service#

The Node-RED flow used in this demo:

  • Loads pre-recorded fed-batch trajectories from IndPenSim (CSV files such as batch2.csv, batch15.csv, batch60.csv, batch86.csv, batch91.csv, …).

  • Parses each row and converts the columns into standardized observations.

  • Streams measurements into InfluxDB 2.7 using a unified schema:

    • measurement: device_obs

    • Tags: device_id, project_name, batch_id, source

    • Fields: observed_property, value

From the perspective of the time-series database, workflow engine, and dashboards, this virtual data stream is indistinguishable from a real bioreactor publishing online sensor data.

Node-RED dashboard#

Node-RED dashboard selecting an IndPenSim batch and starting the real-time data replay into InfluxDB

The Node-RED dashboard exposes an operator-friendly UI that allows users to:

  • Select a batch file (e.g. batch2.csv, batch15.csv, batch91.csv).

  • Choose the project name (e.g. penicillin, pichia, ecoli).

  • Choose the device ID (e.g. R1–R4).

  • Start or stop the real-time data replay.

  • Reset the internal row counter and timestamps.

  • Trigger a metadata bootstrap that populates the stamm_metadata bucket with human-readable labels, units and recommended precision for each process variable.

Every few seconds (by default, every 20 s), the data replay service sends the next row of the selected batch, preserving the original dynamic profiles and relative timing. This makes it possible to test the whole STAMM stack—time-series storage, soft sensors, and dashboards—without needing physical access to a 100 m³ bioreactor.

How to deploy the Node-RED service (Docker-based)#

The Node-RED data replay service is distributed as a Docker-based stack that can run on its own or alongside a full STAMM deployment.

  1. Requirements

    • Docker and Docker Compose installed

    • A running InfluxDB 2.7 instance (local or remote), with buckets such as stamm_raw and stamm_metadata

  2. Environment configuration

    In the Node-RED simulator repository (e.g. penicillin_nodered), copy .env.example to .env and set:

    # Node-RED
NODERED_PORT=1880

# InfluxDB
INFLUX_BASE_URL=http://influxdb:8086
INFLUX_ORG=stamm
INFLUX_BUCKET_RAW=stamm_raw
INFLUX_BUCKET_META=stamm_metadata
INFLUX_TOKEN=YOUR_INFLUXDB_TOKEN_HERE

# Default tags for the data replay
DEFAULT_PROJECT_NAME=penicillin
DEFAULT_DEVICE_ID=R1
DEFAULT_BATCH_FILE=batch2.csv
STREAM_INTERVAL_SECONDS=20

---

Explore the source code and contribute on GitLab: View on GitLab

Contents