Bridge OPC-UA data from industrial equipment to cloud via an MQTT gateway

Steps

1. Deploy an OPC-UA to MQTT bridge tool (e.g., EMQX Neuron, Unified Automation UaGateway, or a custom stack using an open-source OPC-UA client library like open62541 or node-opcua) on an edge gateway with network access to the OPC-UA server
2. Connect to the OPC-UA server endpoint (e.g., opc.tcp://<plc-ip>:4840) using the appropriate security mode and authentication (anonymous, username/password, or X.509 certificate); browse the server's address space to identify the NodeIds of the variables to monitor
3. Configure monitored items or subscriptions on the OPC-UA server for the target NodeIds with a sampling interval matching the process update rate; the server pushes data changes to the gateway client via Publish responses
4. Map each OPC-UA NodeId to an MQTT topic (e.g., site/<plant>/<machine>/<variable>) and serialize the value, timestamp, and status code into a JSON payload; publish to the MQTT broker on data change or at a maximum publication interval
5. Implement a store-and-forward buffer on the gateway: queue messages in local storage (e.g., SQLite) when the MQTT broker is unreachable and flush the queue in order when connectivity is restored to prevent data loss during network outages
6. Apply payload normalization: OPC-UA data types (Int16, Float, DateTime, StatusCode) must be explicitly cast to JSON-compatible types; include the OPC-UA quality/status code alongside the value so consumers can filter bad-quality readings