Sensor to Cloud using Azure IoT

This article explains the step by step procedure to send sensor data to Microsoft azure cloud using IIoT starter kit (A secured gateway). Industry 4.0 induced the OEMs to migrate towards a scalable and interoperable architecture with demand for a vendor neutral communications standard. Open Platform Communication Unified Architecture (OPC UA) is a platform independent, service-oriented modeling framework that is capable of meeting this demand from the OEMs. IEC 62541 OPC UA is fast emerging as the application data modelling standard to enable interoperable data exchange between devices from different manufacturers.

Among many OPC UA stack available open62541 is the one which support both pub/sub for OT communication and secure client/server for the cloud communication. Meanwhile, this article explains the below use case and step by step procedure to send the data from machine to cloud layer. The end-to-end system was visualized by simulating the machine data with multiple raspberry pi devices, windows pc acting as an IoT-edge gateway device. The data from raspberry pi was sent to the azure IoT cloud using OPC UA standard in the IoT -edge and monitored using the application provided by Azure. Here Azure IIoT Starter-Kit acting as secured gateway between the cloud and factory.

• OPC UA secure server compiled from open62541 stack runs in Raspberry pi.
• The OPC UA server has all the data in its address space and waits to accept a trusted connection to send the data.

• All the plug and play embedded devices are connected to one common switch (to the factory level network). The gateway system running a docker engine is also in connection with the factory network.
• IoT edge (Microsoft IoT opc publisher) application runs in one of the docker container. The IoT edge (OPCUA client) connects to embedded device running open62541 OPC UA server and registers to a node to receive push notification during data change. This service in OPC UA concept is known as Monitored Item, client server based publish subscribe model.
• Whenever the data changes, the change is notified to registered client. This exchange is secured using RSA algorithm which is established initially by exchanging their certificates.
• OPC UA server (open62541) sends the data in UA Binary encoded format) send data to IoT edge. The data received at the IoT edge is interpreted and encoded to JSON format.

• The JSON data is sent to cloud server (IoT hub).
• The data from the IoT hub is stored in a database and it is further used for time series analytics, predictive maintenance and decision making (Machine Learning).

Industry 4.0 has a need to store huge amount of data which come from disparate sources and in different format. To address this a data lake in cloud is used. Data lake is a centralized repository that stores all structured and unstructured data at any scale. In which different types of analytics, from dashboards and visualizations to big data processing for taking better business decisions and operational decisions. Likewise, digital twin is also one service in cloud that is the digital replica of physical assets that lets to take better business decisions.

Technologies are emerging in Industry 4.0 and but they are starting to take shape. Standard way so doing things is starting to emerge. This article showcases the technical steps involved in getting your data from shop floor to cloud layer considering a simple use case of any Industry. Smart factory & Industry 4.0 will not only benefit the manufacturers but also revolutionize other industries such as farming, agriculture, dairy etc.