The impact of Reservoir Pressure on Artificial Lift Design

You know the scene, an ESP failed overnight, the production dance is going on. You need to get the well back online asap…..

What information should you give your vendor(s) to perform the design and what information has the most impact on design? This article will show why it is important to use a high reservoir pressure if you want to guarantee your desired production….

Industry Crew Change

In the last 10 years we’ve had a massive exit of “grey hair and no hair’ from our industry and with it goes a wealth of experience. That is the case for operators and service companies. The result is that the people performing designs for our ESPs or galsift completions may be recently graduated and the operator staff reviewing designs have a similar level of experience.

IPR Tracking

While working with a client recently, it became apparent that the design proposed for an ESP could be improved. In this article I’ll show you the impact on the design of using too low a reservoir pressure.

Clients that use our Pump Checker software get IPR tracking over time. For every test a bottomhole flowing pressure (pwf) and productivity index (PI) are automatically determined as shown in the following graphs.

Rate vs pwf over time

Productivity index determination vs time for constant reservoir pressure

To calculate the PI a reservoir pressure is required. The two questions we get asked all the time is, “what reservoir pressure should I use?” and “how often should I update the reservoir pressure”.

At risk of being flippant our response is “who cares, it’s wrong anyway”. Our advice is to pick a sensible number and then leave it as is, don’t worry about updating it. The ESP sensor pressure prior to ESP start, corrected to perforation depth is as good a number as any. The reality is its changing with time and the pressure in the reservoir near the wellbore can be very different from the reservoir pressure, especially in a tight reservoir.

Impact of low reservoir pressure

The following graph shows a well with the last representative test showing a flowrate of 650 bfpd with a pwf of 903 psig. The IPR and consequently the PI and Absolute Open Flow Potential (AOFP) are very different depending on what is assumed for reservoir pressure. The graph shows an example for a reservoir pressure of 3600 and 1200 for the same test. Which is correct - who knows?

Impact of Pr on derived PI and AOFP

The Design

What is interesting for this well, is what happens when the ESP fails and we need to design a replacement. What information should be provide our vendor to design the replacement?

In this example we know from our ESP Diagnosis that the ESP was worn, so instead of designing for 650 bfpd, we’re going to be greedy and design for 700 bfpd.

This means that we’re designing for a rate that is not the same as the previous producing rate and so knowing the slope (governed by the PI) of the IPR curve is important. When we do the design for the two scenarios, we see completely different results.

Counter-intuitively, designing with a higher reservoir pressure results in a 12% higher TDH requirement.

In other words, if we design using a low reservoir pressure, we use a higher effective PI and consequently calculate that we require less pump stages to achieve the desired rate. The danger is that we end up with a pump with insufficient stages to give us the production we desire. 

My advice is stick with a higher reservoir pressure and don’t worry about trying to account for reservoir pressure depletion. Make sure you have enough stages in your pump or enough gaslift valves with appropriate closing pressures to allow you to produce the rates you want.

ESP Design Course Coming Soon

If you found this article insightful, we’ll be running an ESP design course later this year, click here if you’d like to register an interest in attending.