Simultaneous Fracturing and Drilling in Unconventional Shale Plays

By Robello Robello Samuel, March 26, 2014

of the reference well = radius of the error + radius of wellbore.

The radius of offset well = radius of error + radius of casing/wellbore + radius of the fracture length (half radius of fracture length).

The minimum separation distance along the direction vector between the two wells  – the offset well with the fracture length and reference well  – is the minimum allowable distance that a well being drilled can approach an offset well with the fracture of half-length. Several analyses have been attempted to use more refined methods for expressing the probability of collision; but, when the results were compared to the minimum separation-distance method, the differences were found to be negligible. The minimum separation-distance method provides two advantages:

It can be used to determine minimum separation distance.

It is conservative for the case when the casing diameters are large and the combined survey errors leave little margin. This method is also applicable for surface-collision analysis.

It is necessary to calculate the probability of intersecting the fractured well or the well that is being fractured. It depends on the positional uncertainty of the drilled well and the positional uncertainty of the fractured well, uncertainty of the fracture length, and uncertainty of the fracture orientation. Alternatively, the objective and reference wells might be different – the reference well could be in process of being fractured, while the objective well could have already been drilled. The error sources for both the wells can be the same or different. If the survey data, fracture data, and error sources are known, it is possible to calculate the covariance matrices, C, for the two wells, thus providing the uncertainty of the reference location at the reference well with respect to the fractured well location. 

In the calculation shown previously, the fracture length is fixed; only the orientation of the fracture is uncertain. Usually, the fracture length is estimated using published models; but, in reality, the fracture length underground will also be uncertain, and the length should be a random number. A new covariance matrix can then be calculated using the equation below:


In this example, the modified   can be substituted. 

Where,   and   are covariance matrixes of the points, Po and Pr, respectively. 

The complete analysis consists of two options. The first option is forward analysis from the drilled well by assigning the variables in the offset wells. The second option is the reverse analysis, assuming that the offset well is considered as the reference well and the reference well as the offset. Figure illustrates the simulation in a multiwell environment coupled with earth modeling. The uncertainty cones between the reference and offset wells can be observed. With real-time data stream, the solution can be updated and can be further used to provide advanced warning so that proper remedial actions can be taken. The accuracy of the analysis directly depends on the accurate measurements of the fracture length or other parameters considered. 

Robello Samuel has been a Halliburton