A brief overview of IoT
The physical embedded devices are integrated with the web-enabled software, to create an Internet of Things. These devices may be of different kinds, such as personal wearables or devices that are used in homes or offices, but they all perform some function. The data that is generated from this function is uploaded to the web. The connectivity is provided by ethernet or wireless LAN, and the data can either be stored on internal dedicated servers that are in the network, or can be uploaded on the cloud.
IIoT – Industrial Internet of Things
Mr. Kamat said that industrial devices have been using electronics systems like data acquisition systems and instrumentation systems for a very long time. An example of this is a system that measures and displays the temperature of a boiler. The data that is collected from the instrumentation system can also be used to directly control the operation of the plant, using controllers like a PID controller. He mentioned that the industry trend is to interconnect all the industry devices, collect all the data in one place, store it on a server and use it to make well-informed decisions. Since there are different devices like programmable logic controllers(PLCs), PID (proportional-integral-derivative) controllers and CNC (computer numerical control) machines involved, gateways and data concentrator units(DCUs) are used to convert the protocols from one form to another. The DCUs collect the data from the industrial devices, process the information, and then upload it onto the servers.
The number of small scale industries that have adopted IoT connectivity is not very high. Such industries need to first ensure that all their equipment is standardised. Retrofitting can be done for all the industry equipment. The process may take about 2 years, depending on the size of the industry. Smaller industries may need assistance from larger industries at this stage.
Industrial automation systems
IIoT can be integrated with industrial automation systems. There are certain stages involved in this process. One of them is sensor fusion, which is a process by which the data coming from a group of sensors is combined together. This is more efficient than using any one of the sensors at a time. There is also the data concentration stage, during which the combined data from all the sensors is temporarily stored and then uploaded onto the servers.
Mr. Kamat also explained a few more concepts related to IIoT. One of them was computational intelligence. “Computational intelligence is a part of cognitive science. Using computational intelligence, we can manage the internet of things, because the internet of things is a network of objects, and for communication interfaces like Zigbee, you use a mesh topology. In mesh topology, every object talks to every other object in the network. They are all peers. And all these peers, when they talk to each other, there can be issues – the distance between two peers, the availability of signal conditions in that particular premise, all these are of concern. And that’s where computational intelligence comes into effect. There are a lot of paradigms. Many of them are quite old, they have been existing for few decades now – things like artificial intelligence and artificial neural networks. Then we have fuzzy systems, and nowadays we have evolutionary computational intelligence and swarm intelligence.” He mentioned that concepts like these go a long way in optimising designs for IIoT systems, by improving the speed of communication, reliability of data and reducing the emission of radio frequency signals, which is considered to be a health risk.
