Big data

Analysing the data behind plant processes is a very hot topic right now, but the truth is the boxes that collect this information have been around for a long time.

Data acquisition systems are a critical tool for plant-floor decision making because of their ability to obtain and analyse data.

 If you look at today’s complex networks, you’ll realise a lot of data is being generated but it requires a lot of analysis to make use of it

Frost & Sullivan’s Sujan Sami

But analysts say that integration of data acquisition systems with cloud and mobile technologies are what will really change the game for these solutions.

What is driving their adoption is big data analytics, says Sujan Sami, program manager for measurement and instrumentation at Frost & Sullivan.

“If you look at today’s complex networks, you’ll realise a lot of data is being generated but it requires a lot of analysis to make use of it,” he says.

End users also want to store the data for much longer periods of time as quality assurance regulations on a global basis become more stringent. 

Integration with other technologies is another factor fuelling innovation, says Sami, in the quest for open architecture and greater bandwidth.

Monitoring process safety is another important function says Paul Beardsworth, sales manager at HBM.

“This might include measurement of stresses, strains, loads, temperatures or pressures over the lifetime of a product,” he says.

Another popular application is for monitoring plant equipment beyond its lifespan - such as that running in old power stations. 

“The equipment designer may say the user can get another five years of life if they measure and monitor this particular part of the process or equipment,” says Beardsworth.

Many users of data acquisition technology for process monitoring are now also asking for smaller units to squeeze into remote locations or more flexible units that will allow more pieces of equipment that can feed data into a single unit.

“There is always the challenge to go that little bit smaller and more flexible, to create a box that can measure everything,” he says.

Michele Loseto, business development manager for Oil & Gas at Honeywell Process Solutions, says: “When I interact with customers, my mantra is that instead of pushing for more and more data, we advise they use the data they already have.”  

An example is separation processes in the upstream oil industry. 

Basic instrumentation might enable you to measure wellhead pressure and temperature but it can also provide interesting information on what is happening at the wells to a distributed control system (DCS), he says.

“The DCS is the cabin and from there the user can control the field and facilities.” 

With access to both real time and historical data, he says the user can also run analyses from the past to verify whether equipment is running within normal parameters.

Exception-based surveillance uses this condition monitoring data to track abnormal conditions that trigger a notification, so someone can look into a situation before it becomes a big issue.

He says Honeywell once provided such a solution “that helped a user to avoid replacing equipment to the tune of £200,000 per well”.

OPC UA has been around for a while now, but it’s far from falling by the wayside of the rapidly advancing juggernaut that is the industrial automation industry

European Automation’s Jonathan Wilkins

With this method you can also direct a person to the well that actually needs attention, instead of them having to inspect every single one.

“Corrective maintenance is four times more expensive than preventative maintenance,” he says.

Another important goal for many plants now is to capture data constantly with no errors, and keep it forever, says Nick Milner, European director of professional services at Aspentech.

“[Data acquisition] is a lot easier than it used to be, but the underlying requirements have remained pretty much the same.”

The addition of modern analysis technology, however, can yield extremely useful results.

“On top of a base layer of data acquisition you can add a number of systems to analyse and make use of data, and one of the things we provide is a really powerful search mechanism,” he says. 

“So if you type in the name of a particular unit, the system will search through any databases, process graphics or engineering models you may have.”

Once you’ve got this data you can find the root cause of an incident and replay it to establish trends, says Milner.

“We are increasingly seeing people wanting to do very specific things with the data – so we are providing tools such as ‘Golden Batch’ to make this easier for them,” he says.

Here, the user picks a number of batches that represent exceptional quality and overlays their data onto new batches for comparative purposes.

“What we are increasingly doing is integrating data with other products and making it a lot easier to access,” says Milner.

He says the emergence of new technologies such as OPC UA have helped to simplify this process.

Developed by the Object Linking and Embedding for Process Control (OPC) Foundation, OPC UA is a set of industrial standards for the exchange of data between different vendors.

This helps to facilitate rapid and automated communication between physical systems and virtual networks, says Jonathan Wilkins, director at European Automation. 

“OPC UA has been around for a while now, but it’s far from falling by the wayside of the rapidly advancing juggernaut that is the industrial automation industry,” says Wilkins. 

“OPC UA is helping support industry trends, such as Industry 4.0, big data and the Internet of Things (IoT) by offering end users plug-and-play communications solutions.“We see OPC UA playing an essential role in the development of how products make their way through a production line and the ordering of replacement parts.”