Mechanics of Borehole Ballooning in Naturally-Fractured Formations

Abstract

Abstract The mechanics of borehole ballooning caused by opening / closing of a single fracture intersected during drilling in a naturally-fractured reservoir is explained by means of a coupled analytical model. Equations governing borehole ballooning are summarized for different types of fluid rheology. The effects of various parameters on the fluid flow during a ballooning event are analyzed numerically. Shortcomings of the developed modeling approach are outlined. Introduction Borehole ballooning is the term used to describe reversible mud losses and gains during drilling. Three main mechanisms of borehole ballooning are usually quoted:Variations in the temperature of the drilling fluid[1]: Temperature increase at great depth makes the drilling fluid expand, which may be incorrectly interpreted as formation fluid influx. Temperature decrease makes the fluid contract which may be incorrectly interpreted as a mud loss.Elastic deformation of the borehole walls[2]: Borehole pressure decrease results in a borehole volume decrease, while its increase results in a borehole volume increase; thus, mud gain and mud loss, respectively.Opening / closing of natural fractures intersected during drilling. In naturally-fractured formations, the third of the above mechanisms should play a major role. In this paper, we will consider a possible approach to modeling borehole ballooning induced by opening / closing of natural fractures. The main motivation for this work is the industrial need for an accurate detection and recognition of drilling problems such as mud losses and kicks. Most drilling simulators curently in use consider only fluid flow inside the borehole, without account for the complex coupled processes inside the fractures. An accurate model of the fracture-induced ballooning phenomenon should provide an aid in mud optimization and improve the well control procedures while drilling in naturally-fractured reservoirs. Basic Assumptions of the Model We consider mud loss / mud gain events that occur when a borehole intersects a single fracture. Fracture networks are not considered at this stage. Formation fluid and fluid in the borehole are assumed to have the same rheology and the same properties. Both fluids are incompressible. The fracture has impermeable walls, which is not unreasonable in carbonate reservoirs. The fracture is horizontal and has a circular shape. Why we consider a circular fracture? It is known from structural geology that circular shape is a good approximation for fractures in uniform or thin-bedded rocks[3,4]. Circular fractures were used in the fracture network model developed by Bruel et al.[5] Circular fracture is not a good approximation in case of fractures confined between two bedding layers, e.g. subvertical fractures in a limestone layer running between two shale layers. Why we assume linear fracture deformation? The key feature of fracture deformation (opening / closing) when the pressure inside the fracture is increased / decreased, is nonlinearity: fracture's compressibility is a function of pressure. Several attempts have been made to develop an analytical model of fracture deformation based on the assumptions about deformation of asperities on rough fracture surfaces[6–8]. These models, although they describe fracture deformation quite well, are still hardly appropriate for incorporation into a ballooning model since they require a good knowledge of the fracture surface geometry and properties of asperities. Rather a simple analytical relation between effective stress and fracture opening is needed for the ballooning model. Such relations can be provided by empirical fracture deformation laws established on the basis of laboratory and in situ tests. Several types of empirical fracture deformation laws are known at present. A hyperbolic function was used by Bandis et al.[9] and Cook[10]. Bruel et al.[5] quote an empirical exponential law given by: