ABSTRACT

ABSTRACT:

We introduce an inter-well rock mechanics parameters prediction method using a combination of geophysical well logging data and 3-D seismic data. The method has been used to estimate Poisson¡¯s ratio (¦Í), Young''s modulus (E), fracture closure pressure (PCL), and fracture breakdown pressure (PBD) during fracture treatments, for any reservoir location in the rock volume analyzed. The statistical approach is based first on explicit correlations between static core analysis of rock mechanics parameters and dynamic multi-pole-array sonic logging. By further multiple linear regression analysis, these parameters can in turn be correlated to common well logging curves. Then, inter-well rock mechanics parameters are estimated through the use of the 3-D seismic database, constrained statistically using the logging data from individual wells. As a result, a 3-D rock mechanics properties model is developed for the volume covered. Relevant rock mechanics parameters at any point in the volume can be estimated from the model, allowing hydraulic fracture engineering design to be undertaken before an infill well is drilled.

1 INTRODUCTION

There has been growing interest in determining insitu dynamic rock mechanical (lithomecanical) parameters in the oil and natural gas E&P industries. Various methods are available for measuring rock mechanics parameters. Mechanical and petrophysical properties are usually obtained from core sample tests using standardized laboratory procedures. This method is limited by the availability of core and by the cost of testing. Although interactive multi- pole-array sonic logging is an important tool for analyzing lithomechanical parameters, logging cost is too high for such tools to be used extensively. A mechanical earth model is one of the key tools for rock mechanical properties characterization as well as for provision of data to implement fully coupled reservoir geomechanics simulation.

This content is only available via PDF.
You can access this article if you purchase or spend a download.