Nouvelles méthodes d'identification des fractures par diagraphie acoustique en full wave form

New Methods of Identifying Fractures by Full Wave Form Acoustic Logging

IUT Talence

Résumé

Les outils acoustiques de dernière génération permettent maintenant d'enregistrer l'ensemble des ondes générées par une source acoustique à l'intérieur d'une géométrie cylindrique telle qu'un puits de sondage. Le train d'onde qu'il est alors possible d'analyser se compose successivement de trois composantes majeures (l'onde de compression, de cisaillement et de Stoneley) dont nous avons une représentation pour chaque position de la sonde à l'intérieur du puits. Nous présentons, dans ce texte, trois méthodes originales et rapides (calculs possibles sur le site même) pour identifier, à partir du traitement de l'onde de Stoneley, les fractures ouvertes recoupées par un forage. Nous donnons, dans un premier temps, nos motivations pour le choix unique du traitement de l'onde de Stoneley pour, dans un deuxième temps, exposer les trois méthodes développées et montrer pour chacune d'entre elles une application pratique.

Abstract

Interest in recognizing and identifying fractures in a coherent formation for the petroleum, geothermal and storage (oil and gas, wastes) sectors has led to the development of indirect prospection methods inside boreholes such as acoustic logging. The latest acoustic tools are capable of recording all waves generated by an acoustic logging tool inside a cyclindrical geometry such as a borehole. The wavetrain that can then be analyzed is successively made up of three major components (the P compression wave, the S shear wave and the Stoneley wave) for which we have a representation for each position of the logging tool in the borehole. An example of a recording is shown in Fig. 1. Because of its specific features (high amplitudes, low frequency, high signal-to-noise ratio), the Stoneley wave is recognized to be a good indicator of open fractures. Therefore, we use simple digital processing to quantify the influence of fracturing on the propagation of the Stoneley wave. Three methods stemming from the digital processing of the Stoneley wave are proposed. For each of them, an application to data obtained from deep drilling illustrates their potential for use. The first method consists in quantifying, in the frequency domain, the frequency modifications of the Stoneley wave as it passes a fracture. For this we have adapted the coherency function (Fig. 2). The second method is based on an examination of the deformation of the shape of the Stoneley wave in the vicinity of or opposite fractured levels (Fig. 3). The third and last method adapts geometric inversion techniques to acoustic logging. This enables us to use the energy of the Stoneley wave (Fig. 4) to obtain attenuation logs with a maximum vertical resolution (Fig. 5). For all these methods, the vertical resolution obtained is equivalent to the displacement interval of the logging tool inside the borehole. A quick inventory of the methods normally used for detecting fractured zones has been made. The impossibility of such methods to position and assess the importance of each fracture making up a fractured zone, particularly because of low vertical resolution, shows the great contribution of the methods proposed for better identifying fractures.