F.A.S.T. RTA™ is designed specially for analyzing production data &endash; both rate and flowing pressures. Use it to:
F.A.S.T. RTA™ has eight major modules, with different degrees of analytical sophistication:
F.A.S.T. RTA™ is another practical toolkit from Fekete that enables you to conduct advanced decline analysis on both production and flowing pressure data at the same time. It allows you to determine expected ultimate recovery, gas-in-place, permeability and skin and to perform material balance analysis without having to shut-in a well.
F.A.S.T. RTA™ can be used in both oil and gas reservoirs. All analyses (typecurves, analytical models) can be used in RTA if the fluid system in the reservoir is single phase. For oil, that means it must be undersaturated or the pressure above the bubble point. Multiphase oil systems can be analyzed using the Numerical Model. In oil reservoirs, PVT properties and relative permeability data are extremely important: These include Formation Volume Factor (FVF), viscosity, compressibility and bubble point pressure. For multiphase oil systems, user defined relative permeability and PVT properties (if available) should be entered into the Advanced Properties section in RTA. When analyzing oil reservoirs in F.A.S.T. RTA™, it is important to have good flowing pressure data.
F.A.S.T. RTA™ is primarily designed to analyze boundary-dominated performance in wells. If your well’s performance is clearly boundary (depletion) dominated, then the likelihood of obtaining a unique and robust estimate of original hydrocarbon in-place (OHIP) is very high. However, this may not be true of the transient (infinite acting) portion of the same data set. Therefore it is entirely feasible that multiple transient type-curve matches will appear to be reasonable with little or no impact on OHIP.
When this occurs, you must rely on other (independent) well information (well test analysis, core data etc.) to judge which is the best transient match.
There are three wellbore correlation options:
1. Dry gas wells (no liquids) &endash; F.A.S.T. RTA™ defaults to a multi-step Cullender Smith procedure,
2. Gas wells with some liquids in flow stream &endash; Beggs & Brill or Gray. Beggs & Brill is the default correlation. Gray is more robust in conditions where you have high pressure and high volume. It is recommended that you run both to see if you get similar answers, and
3. Oil wells &endash; Hagendorn & Brown.
Your confidence in F.A.S.T. RTA’s results will depend on how repeatable those results are. If several correlations give you the same or a similar answer, you can have a high degree of certainty in those results. Additionally, if you have a large amount of smooth (not too noisy), reliable pressure and production data, it will increase your confidence with F.A.S.T. RTA's results.
The reliability of pressure and production data does not necessarily pertain to cumulative production time. Analysis of a one-month producing well, where the data has been collected hourly, is likely to be more reliable than a well that has been producing 6 months, where only monthly data is available. In general, the more data points you have, the more descriptive the type curve match will be.
Condensates may be imported into F.A.S.T. RTA’s production table as oil volumes. These volumes will be used in the wellbore multiphase pressure loss correlation. However, single-phase gas is assumed in the reservoir. The condensate rates are NOT converted to equivalent gas volumes.
There are two issues the user should be aware of when analyzing gas containing condensate volumes:
1. Pressure loss in the well &endash; If possible, it should be determined whether or not condensate volumes are liquid in the wellbore. In most cases, condensates drop out of the gas at the surface (condensates remain dissolved in the gas while in the well). In these cases, the condensate volumes should not be entered into F.A.S.T. RTA™. However, the wet gas gravity should be input into the program,and
2. Equivalent gas volume of condensates &endash; The measured condensate volumes also have an impact on the material balance. However, below a CGR of approximately 100 bbl/MMscf, the contribution of the condensates is insignificant. For CGRs above this, the condensates should be converted into equivalent gas volumes and recombined with the gas rate stream. This step must be done outside of F.A.S.T. RTA™.
The amount of history needed will depend on what information you are trying to determine. If you are trying to estimate reserves, you need to have sufficient production so that all boundaries have been observed in the reservoir. This "time to stabilization" depends not only on reservoir size, but also on permeability and porosity. If permeability is high and the reservoir is limited, then it may take only hours or days to collect enough data to see the boundaries. However, if you have more extensive, tight reservoirs, it may take in the order of months or even years to see the boundaries. If you do not have enough data to determine your reserves, you need to continue collecting data until such time as you observe all the boundaries.
If you are trying to establish permeability (k) and fracture half-length (Xf), or k and skin (S), your collected data does need not to span a large time interval. However, you need to collect the transient data very frequently. This is important. A high density of data aids the analysis of transient data for k and S or Xf.
At a minimum, use an Excel spreadsheet with three columns; TIME, RATE, and PRESSURE:
Additional data required to run F.A.S.T. RTA™, which are of lesser importance; Reservoir properties such as reservoir temperature, net pay thickness, porosity, water saturation, and gas analysis. If using surface pressures, you will need to convert them to bottom-hole; in this situation, a wellbore schematic will be required.
A screening level analysis is possible with a minimum amount of data using only initial pressures; Assumptions can be made about most other properties.
F.A.S.T. RTA™ is an excellent tool for determining if there is an active water drive or some other form of pressure support. In fact, the advantage of using F.A.S.T. RTA™ over traditional material balance based methods is that regular shut-ins to establish the reservoir pressure are not required. F.A.S.T. RTA™ can often detect the influence of a water-drive on reservoir pressure much earlier than with traditional methods based upon production data and flowing pressures alone. It is possible to determine if an aquifer is acting on the reservoir (before a drop of water is ever produced) through analysis of the pressure transient response.
Identification of reservoir pressure support in F.A.S.T. RTA™ simply requires a comparison of your production response against a boundary-dominated (depletion drive) type-curve stem. If your data follows that stem for a short period then drifts to the right of it (above the stem), that would be an indication of pressure support.
Yes, but for a pumping oil well you must have knowledge of fluid levels over time in order to calculate flowing bottom-hole pressure for direct input into F.A.S.T. RTA™.
If the well operates as "pumped off," you know or can calculate the bottom-hole flowing pressure. Your pressure source needs to be specified as the annulus and the static column fluid must be set equal to gas. Then, the wellbore pressure loss scheme in F.A.S.T. RTA™ is valid for pumping wells.
Once the internal economics have been run to identify a possible candidate
with a sufficient reserve base to warrant stimulation or
re-stimulation, then F.A.S.T. RTA™ can be used to determine if the
well is damaged or stimulated. An obvious answer would be if a skin is
observed on a well where there is a very large reserve base. F.A.S.T.
RTA™ can determine this with a minimal amount of high quality production
data which has minimal fluctuation/noise.
For gas reservoirs, estimating EUR analytically is fairly straightforward. This is because most gas reservoirs are single-phase compressible drive systems. F.A.S.T. RTA™ enables the user to establish EUR using any of the advanced methods by inputting an abandonment pressure.
For oil reservoirs, the situation is more complex because, in general, they are not single-phase compressible drive systems. For instance, an oil reservoir under waterflood is a displacement-dominated system, with a negligible compressible drive component. With oil reservoirs, it is important to establish the oil-in-place early in its life when the reservoir is still above the bubble point and system behavior is simple.
With different recovery schemes, there is no good analytical way to establish a recovery factor. We suggest estimating the recovery factor by analogy to similar pools with similar schemes, then applying it to determine your EUR. The traditional decline curves are also a valuable tool for estimating oil EUR. These can be used in combination with OOIP estimates from the advanced methods to determine recovery factor.
Yes. F.A.S.T. RTA™ is ideal for tight gas reservoirs. The reason F.A.S.T. RTA™ is an attractive option for tight gas is due to its material balance capability. It is difficult do get a material balance for tight gas reservoirs as shutting them for sufficient time to establish average reservoir pressure is not economical. F.A.S.T. RTA™ has the ability to obtain these values using flowing data instead; i.e., using flow rates and flowing pressures. Seeing the boundaries within tight gas reservoirs can be a long process due to the low permeability. However, once boundary dominated flow is established, you can then use the data to calculate reserves and EUR just as you would with a conventional gas reservoir.
It depends. For example, if you have a complex reservoir with natural fractures, an odd shape and multiple layers, and all the boundaries of that reservoir have been reached, then none of the complexities matter. An equivalent reservoir that is cylindrical in shape with a single permeability could be used to represent the entire reservoir. Again, if all the boundaries have been reached, complexities do not hinder the analysis. It is when all the boundaries have not been reached that the analysis becomes difficult. In these situations you can attempt to get a good model match but the level of certainty is low.
With the acquisition of properties being so commonplace today, this situation is very common. Even though you may make an attempt to start collecting accurate wellhead data in order to start analyzing your well, your ability to use the data to give you reliable results is greatly diminished. However, having "some" data is still valuable. You can still make assumptions about what you think the well is doing. For instance, if it is a high draw down well, then the pressure may not have been greatly impacted, allowing it to remain constant. If the reservoir is floating on line pressure, you can use a constant value all the way back. If it has a "certain" shape to it for the last couple of years, then you can extrapolate back to initial conditions by using the perceived trend. Even better, if you can use spot pressures which correlate to where you think things have changed and you know the pressure change, then use that information to connect the points or if you know the initial flowing pressure from an initial deliverability test, then you can use that point to anchor your results. This approach is far more valuable than assuming a constant pressure all the way back.
It is also recommended to do a sensitivity analysis. If you assume the pressure is constant and looks one way, then you make an assumption and it looks another, compare those results. You may see that when you compare the two, that they look the same, if that is the case, then there is no sensitivity to pressure. If they are completely different, then the reliability is very low.
Yes. The fluid flow behavior of heavy oil above the bubble point is no different from that of conventional oil. As with conventional oils, the flow of heavy oil below the bubble point is more complex due to multi-phase issues.
Yes. As long as you have a clear understanding of operation of the system and can determine the bottom-hole pressure, you can use F.A.S.T. RTA™.
No. Waterflood is a system with injectors and producers. Once an oil reservoir is under waterflood, the dominating mechanism in that system is no longer a compressible drive. F.A.S.T. RTA™ is designed for compressible drive systems.
Even though F.A.S.T. RTA’s capability with waterflood analysis is limited, it is ideal for early diagnostics of waterflood performance. If you have just started a waterflood and want to see if you are getting the desired pressure support, F.A.S.T. RTA™ will show you that qualitatively. While F.A.S.T. RTA™ can tell you if you are getting the pressure support, it will not be able to calculate your reserves or model what is going on within the reservoir. However, if you want to know if your well is seeing the pressure support from the waterflood, F.A.S.T. RTA™ will tell you that sooner than any other method.
Yes. As long as your reservoir pressure is above the dew point. F.A.S.T. RTA™ requires single-phase flow conditions in order for reliable results to be generated. Note too that the recovery for a condensate reservoir is lower than that of a conventional oil reservoir. Unfortunately, at this time, we do not have a method in place to help with determining what that recovery factor should be.
Make sure you have no calculated bottomhole pressures that exceed the initial pressure you entered on the Reservoir page. This causes negative numbers which cannot be plotted on type curves. This is why they are greyed out. If the problem is a result of a ’r;rogue’ data point, simply delete the point from the production editor (F.A.S.T. RTA™ will interpolate any missing points). If all of the calculated bottomhole pressures appear to be too high, in relation to the initial pressure, double-check your wellbore configuration (for example, tubing size and depth) and production and reservoir data. If this is not the problem, try using a different pressure loss correlation.
Type curve data can often become distorted, or end up on a distant corner of the plot (which sometimes makes it invisible to the user) if reservoir parameters and/or raw production data are continually altered. Simply press the "Reset Plot Attributes" button on the toolbar to restore the data to its initial conditions.
© 2007 Fekete Associates Inc.
The information in this document is subject to change without notice. No part of this document may be reproduced or transmitted in any form or by any means, electronically or mechanical, for any purpose, without the written permission of the Fekete Associates Inc.