Speed up operational insights on industrial machinery
- Get precise control over industrial machine data with real-time processing.
- Build a global view machinery at distributed remote sites by ingesting data using the MQTT protocol.
- Improve the operational efficiency of machines and impress customers with responsive data-driven optimization.
- Ensure no insights are missed from temporarily disconnected machines by using windowing to process late-arriving data.
Main project components
Machine Telemetry Feeds
Telegraf agents collect sensor data from industrial equipment such as temperature, pressure, motor health, and custom metrics. Data is sent from machines in locations with unreliable connectivity using the lightweight MQTT protocol while supporting store-and-forward for offline operation.
HiveMQ MQTT Broker (on-premise)
Receive and buffer MQTT data from remote machines.
MQTT Connector
Read data from MQTT broker and send it to a Kafka topic hosted in a Quix Edge cluster.
Signal Processing Service
Reads from Kafka and transforms raw sensor data into standardized metrics. Quix Edge is deployed on on-premise servers for processing close to the source.
Machine Cycle Detection
Reads from Kafka and segments continuous data streams into discrete machine cycles for analysis. Includes configurable windowing operations.
Metrics Calculation Engine
Reads from kafka and calculates equipment KPIs and performance metrics using Python-based statistical functions.
InfluxDB Sink
Reads from Kafka and writes processed metrics to InfluxDB or other time-series databases for long-term storage and visualization.
Technologies used
Using this template
This use case template showcases the following features of Quix Cloud and Quix Edge:
- Support for both edge and cloud processing
- Containerized deployment for resource isolation
- Built-in handling of connectivity issues
- Python-native processing of sensor data
- Flexible storage options for processed metrics
