The shifting vision of a smart factory

18 April, 2024

piece of equipment may, for example, be derived from an existing sensor that is used for some other purpose in its PLC programme. Other equipment energy consumption may be measured once so that subsequent consumption can be extrapolated from the operating hours of the equipment as provided by daily manual recordings.

Experienced shop floor workers probably know the conditions when equipment is running but could be idle. When existing intelligence is used smartly a lot can be gained. Nevertheless, at some point, it will be necessary to implement automated data collection. Maintenance staff and system integrators with networking experience are perhaps best equipped to help define where the installation of new measuring devices would be most efficient and most easily achieved. IT can provide the necessary data repositories and backups while ensuring cyber security.

Get the data structured and harmonised

While I have tried to avoid using buzzwords so far, I would like to introduce a paradigm that some may consider a buzzword: edge computing. Data can be collected from many energy meters and other devices as fast as 100 milliseconds, generating large

amounts of data.

While I can agree that data is the new gold, this idea does need to be put into perspective. It’s gold when you have the right kind of data to answer a pressing question. Of course, the price of gold fluctuates with the trends in the worldwide economy, interest rates, and the nervousness of investors. Similarly, the value of data also fluctuates. Energy consumption data has less value when energy prices are low. It’s worth more during an energy crisis. And having huge amounts of consumption data without context related to other sensory and production information, may even prove to be as valuable as fool’s gold.

Deriving real value

Many energy meters provide their information through the Modbus protocol. It has no mechanism for conveying timestamp information or quality information. Challenges around data consistency must also be considered. When an energy meter needs to be replaced, a sudden change in the energy consumption reading can throw off downstream logic, algorithms and reporting. While standards are emerging for unifying information over Modbus – such as Sunspec – not all meters support this. Other data used as virtual energy consumption data needs some transformation and harmonisation in order to be useful.

Ideally, data should be reduced to only what is needed downstream for visualisation, reporting, and decision making. All these tasks can be performed or computed locally, close to production, i.e. at “the edge”. With COPA-DATA’s software platform, zenon as your edge computing engine, it’s easy to provide qualified, timestamped, harmonised, aggregated, consistent, and context-related data. With zenon version 12 available on Linux for selected communication protocols and functionality, it’s even easier.

A fully autonomous, self-optimising, self-maintaining, self-healing, perpetual process factory may today be science fiction. But, in future, we will be needing factories that are smarter still. Think of facilities operating on the ocean floor in the deep sea or

on the moon to produce oxygen or hydrogen. These Smart Factories of the future will be an even bigger challenge.

For further information please visit: https://www.copadata.com/en

https://twitter.com/copadata

https://www.linkedin.com/company/copa-data-headquarters/