Skip to main content Skip to secondary navigation

Distributed Temperature Sensing

Main content start

Investigator: Zhe Wang

Distributed Temperature Sensing is a newly developed measurement technique, and was introduced into the oil industry in recent years. DTS can provide high resolution, real-time and continuous temperature information along a wellbore. With the increasing need for monitoring reservoir production, more and more permanent downhole gauges are installed in oilfields. Compared with traditional measurement tools (e.g. PLT), DTS has many advantages, making it more suitable for permanent installation.

Together with the good news that we have large volume of downhole temperature data, there comes a new challange for petroleum engineers: how to make use of these temperature data. Already proven successful in oild field, DTS data can be well used to predict flow profiling (Johnson et al[i]), allocate zonal flow (Brown et al[ii]) and steamflood monitoring (D.K.Nath et al[iii]).

In our research group, as an attempt to explore DTS applications, we propose a method of solving downhole gauge pressure data integrity problem, based on the distributed temperature information acquired from DTS. Next is a brief introduction of the problem itself.

Downhole gauge placement problem

In well testing operations, downhole pressure gauges are often placed above Midpoint of Production (MPP). Usually, we assume that the wellbore is in an isothermal environment, which means that there is no heat transfer between the fluid in wellbore and the formation; however, in the oilfield operations, the wellbore is in a nonisothermal environment, and heat transfer along the tubing will affect the pressure profile, so placing pressure gauge above the MPP can result in misdiagnosis result from the pressure data acquired.

Figure 1: effect of gauge placement on well testing

As shown in an experiment Kabir et al.[v], (Figure 1) to put the gauge at different points along the wellbore can result in totally different pressure data, which will show different character in well test interpretation. When we put the gauge at the MPP, pressure derivative shows a plateau, while putting the gauge at the wellhead, the pressure derivative shows a dip, which is misleading in this case. This example shows that gauge placement will affect data interpretation result. 

This gauge placement problem is especially severe in high transmissibility reservoirs or gas wells. In high transmissibility reservoirs, when shutting in or opening the well to do transient tests, the time interval for pressure diffusion would be very short. On the other side, heat transfer is large in this kind of reservoir, so heat diffusion will overwhelm pressure diffusion, which may even lead to a pressure increase in a drawdown test and pressure decrease in a buildup test. In gas well, as gas pressure is more sensitive to temperature variation, it is also possible to encounter similar situations as in high transmissibility reservoir.

Figure 2 presents a roadmap of our way to address this problem. We need to build a numerical wellbore thermal model, which involves T, q and p.

Figure 2: Roadmap

In the first block of the algorithm, we input temperature data from DTS and generate a flowrate profile along the wellbore, and use T and q together to calculate pressure distribution. Then we add to this model to make it capable for well testing processes, i.e. transient and multiphase effects. So from the input temperature data, we can obtain pressure variation along the wellbore, which can correct the downhole gauge pressure data. This is the general concept of solving this problem, and following, I presents results we have so far.

Results so far

The first step has been to build a steady-state numerical thermal wellbore model, which can solve both constant (no inflow or outflow along the tubing) and variable flowrate conditions.

The result shown in Figure 3 is to input a constant flowrate profile (at 2500 bbl/d in this case) and generate temperature profile. The result is compared with the result from an  analytical model. The heat transfer coefficient is a function of time, so cases of different production time are compared.

Figure 3: temperature prediction 

This model is also applicable for variable flowrate profiles. The input of a flowrate profile that is variable with two perforations is shown in Figure 4.

 Figure 4: Variable flowrate case

The inverse version this model, i.e. input temperature profile and estimate flowrate profile, is what we need to solve the problem. The result is shown in Figure 5, in which temperature is input and we can obtain the flowrate profile.

Figure 5: Flowrate profiling

These are my research work so far, welcome comments and discussion:

Email: wangzhe@stanford.edu

References: 

[i]    D. Johnson, J. Sierra, J. Kaura, and D. Gualtieri, Halliburton Energy Service Inc.,SPE 103097
[ii] G. Brown, SPE and D. Field, Schlumberger, and J. Davies, SPE, P. Collin, SPE and N. Garayeva, SPE, SPE 95419
[iii] D.K. Nath, Halliburton Energy Service, Riki  Suglanto, PT Chevron Pacific Indonesia, and Doug Finley, Halliburton Energy Service, SPE 97912
[iv]  B.K.Drakeley, SPE, E.S. Johansen, SPE, E.J.Jisk, SPE, and F.X.Bostick, SPE, Weatherford Intl. Ltd., SPE 99696
[v]   C.S.Kabir, SPE and A.R.Hasan, SPE 36527