This IoT use case shows how you can use the Record Evolution platform for IoT and AI to connect any device to the cloud, from interface-free prototypes to large machines, and collect data. You’ll learn how easily apps can be installed on IoT devices around the world, and how live data can be analyzed in the browser so that all your team members can learn about the status of their devices from anywhere, at any time.
To demonstrate all this, we have set up a use case with an electric vacuum pump from the automotive sector:
The setup of our IoT use case
The IoT device is a central element in the entire setup as it ensures the secure connection between the pump and the Record Evolution platform.
Authorized users can access the edge device from any browser. This gives us the advantage that the vacuum pump prototype can remain in one place while the app for the use case is developed remotely, directly on the device.
The vacuum pump app installed on the edge device sends data coming from six IoT sensors attached to the vacuum pump. The data is sent to a protected data pod on the Record Evolution platform where it is automatically analyzed and visualized in near real-time.
A valve attached to the bottom of the base plate can be used to create an error situation by preventing the target vacuum from building up.
The vacuum pump is controlled by the G2 18v25 motor control unit from Pololu. The control unit and some of the sensors are connected to the edge device via an I2C bus system.
In our IoT use case, the edge device is a Raspberry Pi 4. However, you can also use controllers from WAGO, Beckhoff, and other Unix-capable computers.
The current is measured using an HLSR 32-P sensor from LEM. In addition, we collect measurement data from the speed sensor and measure the voltage by means of two resistors:
Connecting to the Record Evolution Platform
To connect the Raspberry Pi 4 (our edge device) to the platform, it is necessary to burn a micro SD card with the Record Evolution operating system ReswarmOS.
The Record Evolution Reflasher will take care of this in just a few minutes. Details about the Reflasher and download links can be found in the platform documentation.
The edge device then automatically connects to the Record Evolution platform after a short time. As soon as the device lights up in green in the browser, the connection to the platform is established.
On the platform, you can create device groups containing multiple edge devices and get an overview of the different device locations. The data coming from all pumps is analyzed safely in the browser and can be used for product development, maintenance services, or usage-based billing.
Developing an IoT app for the vacuum pump
A professional development environment is another component of the Record Evolution platform. With less than 30 lines of code, you seamlessly record the signal of a temperature sensor and store the data in the cloud.
From the dropdown menu, you select any of your connected edge devices to run the app. This allows developers to work with the real sensor data and get working results much faster.
The code is packaged as an app and then deployed to the Record Evolution App Store. Now the app can be quickly rolled out to many more edge devices.
Other apps from the platform’s public app store show, for example, the GPS location of a device, or send a message in the event of an error:
Sending measurement data from the pump to the cloud
The Record Evolution Plattform für IoT und Data Science offers the possibility to set up different data sources. For our setup, we have selected the IoT source:
The data is automatically transformed and stored in a target table.
The created IoT data source pulls the data sent by the edge device and stores it in a raw table in the cloud. In the platform’s data pipes, the data packages are transformed into a format that is accessible for analysis and stored in a long-term target table. A preview gives you a glimpse of the live incoming data. You can also download the data in an Excel format:
In the cloud, all data is stored in a secure central location and new data is added seamlessly to the existing data pool. This enables the location-independent analysis of almost unlimited amounts of data. However, it is also possible to run the entire system on-premises in your own data center or on a virtual private cloud.
Analyzing the pump measurement data
In the platform’s data science workbooks, the query shown below filters the existing measurement data of the pump for a specified time interval. In addition, the following measured variables are derived:
- Motor torque
- Electrical power
- Mechanical power
- Efficiency
- Cumulative energy cost
Live performance graph in the browser
Under the following link, you can inspect the pump performance and zoom into the data by dragging a window on the screen with the mouse: https://studio.record-evolution.com/infoserve/marko/vak27/1142/a3c7d7631fa1227b0f610c44e1934b6b.
And that’s just the beginning. From here, we can develop many different IoT examples and use cases. For instance, it’s now easy to expand your fleet and scale to multiple vacuum pumps working simultaneously. All you need to do is flash more Raspberry Pis and add them to your swarm or group of IoT enabled devices.
This IoT use case illustrates what is possible within the context of a variety of industrial applications and IoT application scenarios, including predictive maintenance, remote monitoring, energy consumption tracking, operational efficiency analysis, condition monitoring, asset tracking, as well as more sophisticated cases across the manufacturing industry.
