Getting technology implemented
Feature Articles, Nov  27  2009 (Digital Energy Journal)

- Steve Jacobs, president of RMI, an oil and gas marketing consultancy company in Houston, is on a Society of Petroleum Engineers (SPE) panel to try to accelerate the rate of acceptance of new technology. “It’s a pretty thankless task I might add,” he said.

Mr Jacobs put together a discussion session at Offshore Europe to discuss ways to get new technology used, including discussions of the implementation of wireless seismic devices and integrated asset models.

Steve Jacobs, president, RMI

The oil industry has been fast to adopt new technology when it sees an obvious benefit, he said. It is the “second largest user of computers in the world after the entertainment industry.”

If a technology could be developed which would enable 40 to 60 per cent oil recovery, that should be adopted quickly as well. “The argument for peak oil would be pretty moot,” he said.

If a technology has an enormous benefit, you can expect much faster take up. For example the first Measuring While Drilling (MWD) technology introduced in 1978 had a high additional cost ($4,000 to $5,000 a day), poor reliability for nearly 10 years after its introduction, but a massive benefit. “We weren’t drilling in the wrong direction for 10,000 feet,” he said.

Meanwhile logging while drilling (LWD) did not have such a great additional cost, but initially did not have such a great additional benefit – because logging by wireline gave more accurate results. So the take-up was not so fast.

Technology companies might be better off trying to work out what oil companies are prepared to pay for, than working out what they say they want, Mr Jacobs said.

“If you can do something which does something people need, it can have a fast adoption,” he said.

“Many service companies have an incremental view – bring in incrementally better technology at a disproportionately higher price. That’s not technology – its just business.”

True to the theme of implementing new technology, the audience was given handheld voting machines, which they used to say which kind of company they were from. Of an audience of 50 people, 40 per cent were service providers, 33 per cent from oil operators and 22 per cent from other and 4 from academies.

The audience was asked what they think is the most important technology for the future, with options including improved seismic, automated drilling, enhanced CTD coiled tubing drilling, wireless telemetry, nano technology, multifrac horizontals, smart completions, enhanced EOR, look ahead LWD.

Audience said enhanced EOR 26 per cent; nanotech 23 percent; seismic 20 per cent, with a small number going for the remaining options.

The audience was asked if they thought oil companies focussed too much on incremental innovation. 64 percent said there was too much focus in incremental innovation, 8 per cent said there was too much ‘breakthrough’ innovation and 28 per cent said the balance was about right.

When asked what they thought was the main obstacle to breakthrough innovation, 19 per cent said a conservative strategy, 36 per cent said a short term focus, 26 per cent said a risk averse climate, 9 per cent said linear development proceses and 9 per cent said too much internal orientation.

Introducing wireless seismic

Chris Friedemann, senior VP corporate marketing with ION, talked about his company’s experience introducing a new technology – wireless seismic geophones.

Aware that new technologies take 35 years on average to reach 50 per cent penetration in the oil and gas market, ION wanted to look for ways to accelerate adoption as much as it could.

It decided that a good way to do this was to involve two oil companies, BP and Apache Corporation, in the field trials. As well as getting them involved in an early stage, both companies committed $8m each to spend on the wireless geophones.

This meant that it had a budget to build 10,000 stations for the field trial – otherwise ION would probably only have built ‘hundreds’, he said. Having a large number of stations was important in being able to see how far the system could go.

BP wanted to find ways to really “challenge” its reservoir development and improve the subsurface image. “They asked, how will we use the data to improve the image,” he said.

Since the first trial, the system has been used many times, including in China and Mexico.

There is something of an art to building good relationships with oil companies, Mr Friedemann says. “You don’t need to have senior management falling in love with your product, but you do need someone within the oil company who thinks it is a good idea and encourages his colleagues to see things the same way.”

You also need people to accept that they are working with an unproven technology and things might go wrong. “It needs a lot of upfront relationship / trust building,” he said. “You have to acknowledge the real likelihood of setbacks.”

In hindsight, it might have been better to field test it on a smaller scale rather than go for a big project at the start, he said.

Some oil companies say that they will only get involved in the technology development stage if they can have ownership over the intellectual property, but that means it would be impossible for the technology company to sell it to anyone else.

One of the biggest horrors when trying to introduce new technology is the oil company procurement manager, he said – they will often try to use methods to reduce expenditure which really don’t work with new technology.

“Typical procurement metrics don’t apply. When the procurement guys show up and try to drive cost out, you just can’t do that. They are not quite set up to work with new technologies that are emerging. We are really scared of this,” he says.

When doing partnerships with oil companies, two is probably the right number. “Involving more than two is likely to prove challenging.”

Mr Friedemann was asked what he would do differently if he had a chance to do it again. “We got pulled into the field trial a little earlier than we wanted to. We would have liked to be one year later so we had a chance to evolve the equipment,” he said. “We got a minor black eye because some performance metrics weren’t where people thought they should be.”

The investors were very supportive of the new development before the economic meltdown. “We were seen as a company that spent a lot on R&D and became a growth story,” he said.

Although some investors were expecting a much faster up-take than ION was able to achieve – with mass adoption within 2-3 years, like the iPod.

The audience was asked if they thought technology diffusion rates in oil companies was accelerating. 51 per cent thought it was accelerating, 41 per cent said they thought it was staying the same and 7 per cent said slowing down.

The audience was asked if they thought onshore seismic was tougher than offshore. 57 per cent said they thought onshore was a greater challenge, 21 per cent said the same as offshore, and 21 per cent said it was less of a challenge.

Integrated asset models

Richard Ella, production business manager for EAF at Schlumberger Information Solutions, talked about his efforts encouraging oil companies to install Integrated Asset Models (IAM) – where you have the reservoir, production facilities and economics of a well in a single computer model.

It is common for the same oilfield’s operations to be captured in three separate models, for subsurface, surface and economics. The geoscientists and engineers have a common earth model of the reservoir; the facilities engineers have a model of the surface equipment and the mass flow rates of fluid through it and the energy required to operate it; and the economists have an economic model for the business.

“But it’s the same asset – and it all has the same sensitivities and uncertainties,” he said. “The system is all connected – any change affects what happens upstream and downstream.”

Communications between the models and departments is normally done “with a lot of manual work and effort.”

There are plenty of questions which come up regularly which need co-ordinated input from all of the models to be able to answer. For example, working out how much injection water is required and how it should be distributed, or working out a program for drilling new wells to ensure use of the production facility is optimised.

But there have also been plenty of problems with putting together Integrated Asset Models.

There are different attitudes towards time – production engineers are only interested in the instantaneous flow rate (i.e. in litres per second) but reservoir engineers are more interested in what will happen over the next few years. “That was a major challenge for a long time,” he said.

A recent development is simplified “proxy” models, which can model what is happening in the reservoir much faster than standard models. They do the modelling in less resolution, but the trade-off is welcome because high resolution models update themselves too slowly.

Integrated Asset Models, which bring everything together, were first built in 1993, and are now used in around 20 per cent of wells, Mr Ella estimates.

There are two different ways to do it. One method (called ‘implicit’) builds a gigantic model with everything included. The second method, which is much easier to implement practically and becoming more common, is to try to connect to individual models (subsurface, surface, financial). This is known as ‘explicit’.

Reservoir engineers tend to end up being the primary custodians of the Integrated Asset Models (IAM), possibly because they have the most expertise with models in general, he said.

There are many success stories already, for example Petrobras in Ecuador reckons it could increase production by 20 per cent as a result of using an integrated asset model, to identify which wells to do workovers on and plan its drilling campaign.

ConocoPhillips used an IAM model for a water alternating gas development in Alaska, where it needed to decide how much water and gas to inject in a number of different wells. This was on the basis that all the produced gas needed to be ultimately re-injected (none of it was exported), all produced water was re-injected, and there were constraints on the energy available to compress the gas across the entire field.

BP used the system to model a 500-well onshore gas field, which had 30,000 horsepower of mobile compressors, to work out how it could get the best flow with the injection gas compression power available.

StatoilHydro used IAM to model a water alternating gas flooding scheme on its Snorre B field, and claims an improved net present value of the field of 30-35 per cent. “We had to create a simplified reservoir model so it would run much faster,” he said.

Pemex used a system in Mexico on a well which had slugging problems, where slugs were constraining the amount of production which was possible.

The audience was asked if they thought full field optimisation models, from “pore to pump”, with real time data being fed in the model to continuously optimise it, would be a “reality” in the next 5 years. 54 per cent said yes, 46 per cent said no.

When asked if they were more likely to adopt integrated asset models in their companies as a result of listening to the speech, 83 per cent said yes.

<<BACK