You are Home   »   News   »   View Article

Bringing analytics to the control room

Thursday, April 18, 2019

There's a lot of call to use the strengths of data analytics to help control room operators - but do they really just need a system to provide them with the data they need? Julian Pickering of Geologix presented an example.

There is a lot of call for 'analytics' or deeper insight into operations for control room staff, providing deeper insight than is available from control room software, for example about an emerging problem, said Julian Pickering, CEO of Geologix Systems Integration.

Analytics doesn't need to mean petabytes of data, he said. You can use the term 'analytics' for a system which reveals useful information which is otherwise not immediately apparent, he said.

As an example, consider a gas cooling system, with heat exchangers and a cooling liquid. Over time, the heat exchanger plates will gradually get fouled (dirty) so they are less effective at transferring the cooling. The system will see that the outlet gas is warmer, and pump more cooling liquid through to compensate. No alarms are activated and the control system operator does not necessarily know.

Over time, the heat exchanger fouling gets worse, and the pumping rate of the cooling liquid gets higher. Eventually, the cooling liquid is pumped at its maximum flowrate, the system is unable to do anything more to cool the gas, the gas is not cooled to the desired outlet temperature, the alarm sounds and the whole system shuts down, leading to enormous cost.

It would have been much easier to fix the problem earlier, when there was only a small amount of fouling, similar to how it is much easier to unblock a sink when it only has a partial blockage, he said.

Any system for an operator needs to be simple to use. 'People have a lot of pressure and responsibility on them. Anything which slows them down is seen as negative,' he said. There's a danger from giving people too much data.

Any warning must be given early enough to enable someone to make a decision with it.

The data must be clean. 'Decisions are only so good as quality of data. If errors creep into data they proliferate through the system,' he said.

The analytics system needs to either be embedded in the operations system, or to sit alongside it. It must also show people what they need to do, not just tell them what is happening. It needs to be very usable, perhaps not needing an instruction manual.

A similar approach could be applied to other scenarios where problems develop slowly, such as non-productive time in drilling, kicks in drilling, premature bit wear, loss of drilling fluids and wellbore instability, availability of critical equipment.

Some companies manage losses by adding in safety margins, but 'it tends not to be the way to operate in the modern world,' he said.

Associated Companies
» Geologix
comments powered by Disqus


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


Latest Edition May-June 2021
May 2021

Download latest and back issues


Learn more about supporting Digital Energy Journal