You are Home   »   News   »   View Article

Enhanced production data

Wednesday, November 2, 2011

The more you want to do with production data, the harder it is to fit it all together, explained Magnus Svensson, Senior system consultant with Dong Energy, at the Digital Energy Journal London conference in June

Managing production data is easy enough when you're just sending a figure to the regulators or partners once a month or on a daily basis about how much you produced.

Where it gets harder is when people want to incorporate the production data into all kinds of software systems - when you discover that people in different departments define terms such as 'net oil volumes' in very different ways.

'From the production domain net oil volume is a crude oil excluding sediment and water. From economics perspective it would be associated with the net revenue associated,' said Magnus Svensson, Senior system consultant with Dong Energy.

He was speaking at the June 2nd Digital Energy Journal / Finding Petroleum London conference, 'Developments with digital oilfield infrastructure'.

There can be problems when one engineer moves to a different field, which uses a different definition for a term such as 'volume management.'

The oil industry has never had a standard dictionary. On the Norwegian shelf EPIM (www.epimg.no) and OLF (www.olf.no) have (as part of a joint operator effort on the NCS) put together a 'reference database' with around 3,000 terms in it (ISO15926), which describes how things should be described within theproduction and drilling domain, including subsea equipment.

DONG stands for Dansk Olie og Naturgas A/S. It is 75 per cent owned by the Danish government.

The country is the largest power producer in Denmark (generating 49 per cent of electricity) and has exploration and production in the Southern Norwegian North Sea and Danish part of the North Sea, also West of Shetland, Faroes and mid Norway. It also has drilling activities in the Barents Sea and offshore Greenland.

Dong has participated in efforts in Norway to standardise production reporting to authorities since 2000, in work which was ultimately incorporated as part of the ISO 15926 standard.

This standard has been used in production since 2006 on a daily basis and forms the basis for the PRODML standard (PRODML started off after the initial effort on the NCS to standardize daily production reporting and as such the xml schemas and definitions have the same basic foundation), and most of the fields in the Norwegian continental shelf use it to report on a daily basis, he said.


Enhanced production reporting

On the NCS a joint operator effort under the OLF/EPIM umbrella recently added enhancements to the data to be sent to governments, including stating who owns what information (ownership stream tracking), how volumes in pipelines are allocated, stock accounting and covering vessels and their cargo (typically liftings), health and safety reports, shutdowns and deferments (when production was less than anticipated). A lot of these requirements come from the Norwegian government but also as part of the NCS standardization of monthly partner reporting.

'We have tackled a lot of different things here,' he said.

Dong wants the data to be generated automatically out of the existing IT systems, not pulled together manually from a range of sources.

Many of its rigs have different processes for collecting the data, and it is handled in different ways. 'One of the challenges here is to find the correct data and to trust the source of the data,' he said.

'The really old fields can be a nightmare to tackle when it comes to standardisation of business processes,' he said. 'That is because of decisions that were taken 10 years ago. They made a certain decision of how to hook e.g. a well up to the platform.'

'We have had several locations where we needed to call up a pensioned production engineer to ask him, what did you actually mean when you tried to introduce this business process? It wasn't documented. The people have been doing it for 20 years.'

There are many software tools being developed which can help find the right data in corporate systems, but it is still very difficult, he said.

Manually entered data is gathered together in forms (ie a list of different pieces of data required for a well), rather than individual pieces of data.

One of the challenges implementing the standardized production reporting has been that all the benefit is perceived to go to head office, not the people who have to do the work of entering the data into a system, rather than their spreadsheets.

'It's been a hard nut to crack convincing people that this is actually beneficial to them,' he said. 'As always, the problem relates to Excel. We can't get rid of it, we just need to tackle it.'


It is possible that standardizing data means that individuals actually end up with less data available to them than before, and that means it is hard to persuade them. Similarly, if you want people to start entering new data, there can be resistance to that.

Working with the data

Then the next challenge is bringing the data into an infrastructure, so it can be interpreted the data, particularly to calculate key performance indicators and do budgeting.

Data for internal use can be generated on time intervals shorter or longer than monthly.


'When you start moving up in the value chain - corporate reporting, budgeting, you have a lot of challenges just relating to gathering and aggregating the data and ensuring that it is correct.'

At the end of the chain, when the data enters financial systems, 'this type of data is the cashflow within the company,' he said.

'A lot of these processes are still done through e-mailing and spreadsheets. It's an old way of doing things. But it's up and running and it's working with some major strains,' he said.


There are people in the oil industry who still wants to receive and work with the data manually, he said.


Sustaining it

Once the methods have been implemented, you have to make sure that people carry on doing them in the same way.

The company does not allow people to add their own data into the system, because it might not all be set up right. 'Then we have the same nightmare over and over again,' he said.

Over time, the overall quality of data should improve. If the data is in PRODML or WITSML format, it is easier to make automatic quality control checks on the data.

If the underlying data is better, it should be easier to develop new reporting and analysis systems or look for new trends.

'We have a common asset model for the whole Norwegian continental shelf as part of the joint effort from all of the operators,describing things like fixed equipment and field setup' he said.







Associated Companies
» Dong Energy
comments powered by Disqus

CREATE A FREE MEMBERSHIP

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

FEATURED VIDEO

Clustering Considerations in the Machine Learning Workflow – Examples with Exploration Data
Philip Lesslar
from Precision DM

DIGITAL ENERGY JOURNAL

Latest Edition Jul-Aug 2019
Aug 2019

Download latest and back issues

COMPANIES SUPPORTING ONE OR MORE DIGITAL ENERGY JOURNAL EVENTS INCLUDE

Learn more about supporting Digital Energy Journal