Digital Energy Journal- IMRANND - better ways to work with inspection data

Join | Existing Member Login

You are Home » News » View Article

IMRANND - better ways to work with inspection data

Thursday, May 7, 2020

IMRANND of Aberdeen is helping oil and gas companies make better use of their inspection data, with a combination of data science and integrity engineering.

IMRANDD of Aberdeen has developed AIDA, a suite of technologies designed to provide a new service that will help oil and gas companies better understand and manage the integrity of their assets. AIDA uses extraction, mining, cleansing, analysis and visualization of available data, starting initially with topside pressure system pipework on offshore structures.

The company employees are a mixture of oil & gas professionals, mathematicians, data scientists and domain experts - integrity engineers and corrosion engineers. With this breadth of experience, they use their combined competence to get a better understanding of what is happening on a given asset.

The aim of the new technology is both to better understand degradation and have more confidence in the readings. It enables companies to better manage their risks and inspection spending.

Oil and gas operators typically spend millions of pounds a year on inspection, gathering hundreds of thousands of wall thickness readings, but not all of the data is directly usable. The data needs expertise to be interpreted and effectively applied.

The analysis work Begins by utilizing data cleansing tools to identify inaccuracies in raw data. The visualization tool then uses data clustering to identify hotspots and blind spots in need of attention.

Using AIDA, it is possible to take a very large, dense data set and group it on a single graph, enabling the user to zoom in on and interrogate specific areas within it, where the biggest concerns are.

Within the software, there are data visualisation tools which can easily show up 'blind spots' on an asset where there is not enough data, enabling operators to focus their inspection efforts where they are needed with greater accuracy.

AIDA can facilitate assessment of the quality of the gathered data and allow a user to cleanse the 'bad' data out, leaving only low variation, trendable data behind. The technology's data analytics give the user advice about threat levels on the asset, possible imminent failures, rate of degradation, and make estimates about remaining lifetime.

The user can see where problems are on the pipework - test points that look likely to fail in the next few years, and where the risks appear highest.

The insight enables engineers to increase the intensity of inspections in areas of higher risk, says Innes Auchterlonie, managing director of IMRANDD.

Companies often plan to inspect at a point of half the estimated remaining life. So, if they expect a piece of pipework to have 10 years life, they'll inspect in 5 years. If they expect 5 years of life, they'll inspect in 2.5 years. It is a sensible method, which focusses most inspection on areas with the shortest life. But it only works if you have a good estimation to begin with.

IMRANDD says that its data processing algorithms are much faster than competitors. For example, one UK offshore asset had 103,000 pieces of wall thickness data acquired over 3 decades'. One service company had spent a year looking at the data to try to assess its quality. 'We took that same data set and processed it in under an hour, using AIDA' Mr Auchterlonie says.

IMRANND estimates that the better analytics of inspection data can lead to big reductions in the requirement to spend on inspections - perhaps 20 per cent savings. At the same time, companies have a better insight into pipework condition, so can avoid costly failures which can put equipment out of action.

Confidence in data

The lack of confidence in inspection data is itself a problem for oil companies, and a target challenge for IMRANND, says Mr Auchterlonie.

Data can be poor both due to poor data collection on site, missing data, or compromises to data after it has been collected.

As a result of so much poor-quality inspection data being around, it is common for people to just disregard data which is outside the expected parameters, when it may have some important insights within it, Mr Auchterlonie says.

Spotting trends and patterns

Trends are very important in analysing corrosion - including spotting that a certain section of pipework is corroding much faster than would be expected, or if a corrosion area is spreading.

So analysing trends can also give a better prediction of remaining life of the element.

But oil companies tend to work by exception - taking action when something is obviously wrong, such as having corrosion to a dangerous level, rather than look for trends, Mr Auchterlonie says.

But often there is so much data missing, trends and patterns are very hard to see, he says.

The company

Innes Auchterlonie, CEO, formerly worked in asset management at Talisman (now Repsol Sinopec) and Hess.

So far, AIDA has largely been used to analyse pipework on offshore structures - work is going on to develop similar methods for pressure vessels.

The company reports that it has grown from 6 employees in 2016 to 40 now and has business development activities in West Africa and the Middle East, along with in its base in Aberdeen.

The product is currently offered primarily as a data analysis service, although it is possible that it could be sold as a software, together with coaching for how to use it, Mr Auchterlonie says.

The company is able to demonstrate its full methodology to clients. It is also able to show ways that its claims - such as to support reduction in inspection spending by 20 per cent - can be verified.

Associated Companies
» IMRANDD

CREATE A FREE MEMBERSHIP

To attend our free events, receive our newsletter, and receive the free colour Digital Energy Journal.