Selection of the optimal completion of horizontal wells with multi-stage hydraulic fracturing of the low-permeable formation, field C

Author: A.M. Bozoev1 and E.A. Demidova2


1-Gazpromneft Muravlenko, Muravlenko, Russia, 2-Tomsk Polytechnic University, Tomsk, Russia



At  the  moment,  many  fields  of  Western  Siberia  are  in  the  later  stages  of development. In this regard, the multilayer fields are actually involved in the development of hard to recover reserves by conducting well interventions. However, most of these assets may not to be economical profitable without application of horizontal drilling and  multi-stage hydraulic fracturing treatment. Moreover, location of frac ports in relative to each other, number of stages, volume of proppant per one stage are the main issues due to the fact that the interference effect  could  lead  to  the  loss  of  oil  production. The  optimal arrangement of horizontal  wells  with  multi-stage  hydraulic  fracture  was  defined  in  this  paper.  Several analytical approaches have been used to predict the started oil flow rate and chose the most appropriate for field C reservoir J1. However, none of the analytical equations could not take into account the interference effect and determine the optimum number of fractures. Therefore, the simulation modeling was used. Finally, the universal equation is derived for this field C, the reservoir J1. This tool could be used to predict the flow rate of the horizontal well with hydraulic fracturing treatment on the qualitative level without simulation model.

1.  Introduction



Most oil companies are beginning to develop field with hard to recover reserves due to growing demand. However, development of these types of deposits  cannot be achieved an uneconomical production rate by conventional method (i.e. vertical wells with hydraulic fracturing treatment). The new  technology  (horizontal  wells  with  multi-stage  hydraulic  fracturing  treatment)  replaces  the previous one due to high production rate.


Further, many scientists have analyzed this issue since 1984 years, for example, such scientist as Giger et al. suggested the first analytical solution for evaluation of horizontal wells intersecting fractures; however, this equation could not manage with flow in the fracture and in the reservoir properly. Moreover, Hujun Li (1996) and Economides and Martin (2010) presented the more practical application of mathematical models. In addition, the next mathematical model was introduced by Boyun Gou and Xiance Yu (2008). The proposed equation includes several flows (the reservoir radial and linear flow, fracture linear and radial flow respectively).T.M. Herge (1994) presented accurate methods of determining a productivity of multifractured horizontal wells as well as Hong Yuan (2010) showed the new approach for calculation of horizontal wells productivity with and without fractures [1-8].


In addition, these models are intended for horizontal wells with transverse fractures. The horizontal well with transverse fractures will be used for field C. Geological and petrophysical properties of the field C reservoir J1 are presented in the Table below.

Unfortunately, none of these methods give good result due to the following reasons:


  • Fracture radial flow is not taken into account
  • Low permeable reservoir
  • Interference effect


However, Economides method demonstrates best convergence in comparison with other methods. Moreover, the wells that have been exposed by the interference effect can be detected on the following chart (only 16 wells) (Figure 1). The coefficient of correlation showed a good result. It equals 71.4 % without wells with interference effect.

Furthermore, none of these analytical solutions which were listed above, take interference effect between adjacent fractures  into account. Therefore, the simulation model seems to be an appropriate approach to solving this problem (optimum completion of the wellbore).

2.  Simulation approach



The size of this sector model is 1400 meters in length (X-direction) and 800 meters in width (Y-direction), moreover the number of the cells in the X-direction, Y-direction and Z-direction is 28, 16 and 32 respectively.  Also, the dimensions of the sector model were selected with respect to the well pattern.


After that, the horizontal well (length of horizontal section is around 600 meters) was added in the center of the sector model with fractures along the horizontal wellbore. Moreover, the fractures were set by the function WFRA (well fracture) in the Tempest software. In addition, there is no pressure support, so horizontal well is on depletion drive during three years (normal lifetime of the fractures after multi-stage hydraulic fracturing treatment).


The simulation model was provided for the next cases:


  • 1-7 hydraulic fractures along the horizontal wellbore (the maximum 7 fractures, because the minimum distance between fractures should not be less 80 meters due to technological constraints).
  • The  mass  of  proppant  varies  from  30  ton  to  167  ton  per  one  stage  (there  is  a  risk  of breakthrough in the overlying water-saturated reservoirs above 167 ton of proppant per one stage).


Location of frac ports relative to each other should be uniform because uniform distribution of frac ports along horizontal part of the well gives the effective involvement of the reserves by a well and exclusion of interference effect.

In general, forty seven models have been done with different fractures and mass of proppant. The cumulative oil production schedule is presented with 100 ton of proppant per one stage for different number of fractures.


Based on this graph, it is possible to conclude that interference effect can be observed with increasing number of fractures (reduction of rate of cumulative oil production). Moreover, this effect also could be detected at the start of the well production. The following chart demonstrates that the production rate does not increase linearly, the production rate remains the same with 7 fractures in comparison with 6 fractures (Figure 2). In addition, the flow rate from the simulation model with 5 fractures (Qliquid =76.86 m3/day) is approximately the same in comparison with the real data, for example the well H5 (L=593 m, Qliquid =78 m3/day).

Selection of the optimal horizontal well arrangement is directly dependent on the economic results. As the rate of cumulative oil production reduces with fracture number increase, however, it is still the highest in comparison with others. However, the expenses also increase with the number of fractures [9-11]. The chart of NPV was constructed for different cases. The obtained results are demonstrated in Figure 3.

The optimum number of fractures of horizontal well is five with 100 ton of proppant per one stage based on the NPV results. Moreover, the fractures with higher mass of proppant per one than 100 ton show lower NPV. It could be explained by insufficient increase in fracture geometry (width and length) that can cover expenses.


Finally, the geometry of the fracture (100 ton of proppant) is listed below:


  • Propped Half-Length 151 meters
  • Total propped Height  56 meters
  • Average Frac Width 3.65 mm

3.  The universal equation



When the recommended distance between fractures is determined, the universal equation could be

derived through the simulation modeling data. The principle of universal equation is the representation of fractures (horizontal well) as vertical wells with hydraulic fracturing [12]. This equation can be used to determine the starting flow rates for horizontal wells with multi-stage hydraulic fracturing treatment on a qualitative level. However, there are several conditions for the application of this equation:


  • The presence of a vertical well in the proposed area of horizontal well drilling;
  • Hydraulic fractured vertical wells;
  • The properties and geometry of the fracture should be identical as for vertical well and for horizontal well with multi-stage hydraulic fracturing;
  • Flow  from fractures into the  wellbore (or  negligible  flow from matrix into the  horizontal section).


First of all, the geometry and properties of the fracture were chosen for 100 ton of proppant per one stage for which the universal equation will be derived. Therefore, the average fracture properties are almost similar to the real fractures after hydraulic fracturing treatment.


Secondly, the correspondence of the interference effect on the multiplicity of the starting flow rates (QL120 /QLn<120) were obtained from the simulation model (5 fractures), with a distance between the fractures being less than 120 meters (limiting distance).

Thirdly, the starting liquid flow rate depends on three variables:

,                                                                    (1)

where       – the starting liquid flow rate of vertical well with hydraulic fracturing, N- number of fractures of the horizontal well,         – reservoir pressure in the region where the planned horizontal well will be drilled.

In  addition,  this  function  does  not  depend  on  the  reservoir  properties  (KH),  because  the depositional  environment  is  shallow  marine  so  the  reservoir  properties  do  not  tend  to  vary significantly. The following chart demonstrates the multiplicity of the starting liquid rate with different number of fractures in comparison with the starting flow rate with one fracture at different reservoir pressure (Figure 8). According to this chart, the equation was calculated based on the linear regression through the least-square deviation method that was derived in the special software Kompas.


 Finally, the universal equation of determining the starting liquid flow rate for horizontal well with

multi-stage hydraulic fracturing was derived by combining these two equations:




If L≥120 than L=120

There  are  two  main  factors  leading  to  a  production  decline  at  Field  A,  i.e.  liquid  loading  and condensate banking.

To compare the obtained results (through the universal equation) with the starting liquid flow rate of horizontal well (with Hydraulic fracturing) 47 wells were chosen (Figure 4).


4. Conclutions



1.  The horizontal wells with multi-stage hydraulic fracturing seem to be more preferable than vertical wells with hydraulic fracturing treatment in the Field C, reservoir J1.


2.  The technology (Frack ports or Ball dropping process) of multi-stage hydraulic fracturing is applied in the Field C, reservoir J1.


3.  Analytical approaches do not take into account the interference effect. However, Economides method should be used with caution, despite the fact that the possible wells with interference effect are observed.


4.  The simulation model tends to be an appropriate approach to choosing the optimum completion of the wellbore


5. The optimum arrangement of the horizontal well is five fractures with 100 ton of proppant per one stage based on the NPV results. The geometry of the fracture (100 ton of proppant) is listed



  • Propped Half-Length 151 meters
  • Total propped Height 56 meters
  • Average Frac Width 3.65 mm


6. The universal equation gives the appropriate correlation factor (R=84,1%) and could be used to

predict the flow rate of the horizontal well with hydraulic fracturing treatment on the qualitative level

without simulation model.



[1]     Herge T M and Leif Larse 1994 Productivity of Multifractured Horizontal Wlls SPE 28845, European Petroleum Conference (London, UK October 25-27)


[2]    Hong Yuan, Desheng Zhou and Xian Shiyou 2010 A New Model for Predicting Inflow Performance of Fractured Horizontal Wells SPE 133610, SPE Western Regional Meeting, Anaheim, California, (USA, May 27-29)


[3]     Guo B and Schechter D S 1997 A Simple and Rigorous Mathematical Model for Estimating Inflow Performance of Wells Intersecting Long Fractures SPE 38104 SPE Asia Pacific Oil and Gas Conference and Exhibition (Kuala Lumpur Malaysia April 14-17)


[4]     Boyun Guo and Xiance Yu 2008 A Simple and Accurate Mathematical Model for Predicting Productivity of Multifractured Horizontal Wells SPE 114452 CIPC/SPE Gas Technology Symposium 2008 Joint Conference (Cakgary, Alberta, Canada, June 16-19)


[5]     Economides M, Valko P, Oligney R Unified Fracture Design (London. IBT Global) p 262


[6]     Wei Y and Economides M J Transverse Hydraulic Fractures From a Horizontal Well SPE 94671   SPE   Annual   Technical   Conference   and   Exhibition   (Dallas,   Texas,   USA, October 9-12)


[7]     Daal J  A and Economides:  M J Optimization of Hydraulic Fractured Wells in Irregularly Shaped  Drainage  Areas  SPE  98047  SPE  International  Symposium  and  Exhibition  of Formation amage Control (Lafayette, LA, February, 15-17)


[8]     Economides  M  J,  Martin  A  N,  Services  B  J  2010  How  to  Decide  Between  Horizontal Transverse, Horizontal Longitudinal and Vertical Fractured Completion SPE 134424 SPE Annual Technical Conference and Exhibition (Florence, Italy, September 19-22)


[9]     Mirza Talha Baig, Sami Alnuam, Muzammil Hussain Rammay and King Fahd Productivity Increase Estimation for Multi Stage Fracturing in Horizontal Wells for Tight Oil Reservoirs SPE-178030-MS SPE Saudi Arabia Section annual Technical Symposium and Exhibition (Al-Khobar, Saudi Arabia, April 21-23)


[10]   Crosby D G, Yang Z, Rahman S S 1998The Successful Use of Transverse Hydraulic Fractures from Horizontal Wellbores SPE 50423 SPE International Conference on Horizontal Well Technology (Calgary, Alberta, Canada, November 1-4)


[11]   McDaniel B W and Keith Rispler 2009 Horizontal Wells with Multistage Fracs Prove to be Best Economic Completion for Many Low-Permeability Reservoirs SPE 125903 SPE Eastern Regional Meetin (Charleston, West Verginia, USA, September 23-25)


[12]   Hemanta Mukherjee and Michael J Economides 1991 A Parametric Comparison of Horizontal and Vertical Well Performance SPE-18303-PA

Author: A.M. Bozoev1 and E.A. Demidova2


1-Gazpromneft Muravlenko, Muravlenko, Russia, 2-Tomsk Polytechnic University, Tomsk, Russia




IOP Conf. Series: Earth and Environmental Science 33 (2016) 012035         doi:10.1088/1755-1315/33/1/012035


Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. This content has been downloaded from IOPscience.


Published under licence by IOP Publishing Ltd