In recent times, the oil industry has shown increasing awareness towards the maintenance of optimum well productivity through better drilling/completion practices. Attempts are being made to control the invasion of high permeability sands by mud solids through the use of sized Loss Control Materials (LCM) which can be difficult to clean up. Recent developments involve the use of sized carbonate and sized salt systems with mixed results. Whilst these special systems have been known to provide efficient fluid loss control, the cleanup efficiency especially in horizontal wells has been known to be poor in certain instances. Likewise, the completion of long horizontal sections through highly permeable, unconsolidated reservoirs has witnessed increasing use of prepacked screens many of which are reported to be plugged/damaged by filter cake debris and completion fluid solids as well as formation fines. Gravelpacks are also known to have been damaged not only by formation fines but also by solids in completion fluids injected from the wellbore especially during cleanup operations. The migration of sized particles in the drilling and completion fluids through high permeability reservoir sands, prepacks or gravelpacks is characterised by flow capacity decline which may be instantaneous or gradual depending on the migration process and the pore bridging phenomena. Therefore, accurate prediction of the prevailing pore blocking mechanism for a given pore throat-particle size relationship can provide a good basis for the estimation of the maximum allowable size of particles in various completion fluid systems and also provide a good guide to the design for optimum prepack and gravelpack performance through the use of properly sized gravels. In this paper, attempts have been made to analyse the impact of a number of key parameters on productivity impairment. The analysis has been based on the results of in-depth research into particle-pore bridging phenomena. Based on rigorous experimental studies to define the phenomena, a number of deterministic models have been developed to mathematically define the unconsolidated pay sand/pack sand systems permeability decline profile as a function of invasion pattern, migration/pore blocking mechanism, production/injection rate, production time and fines concentration and fines textural properties. Application of the results to the optimisation of drilling/completion fluids design as well as prepack/gravelpack design and analysis are illustrated with specific case studies. Introduction The oil industry has continued to use sized particulates in many facets of its well completion programmes either as dispersed solids in drilling/completion fluids or as pack sands to control the migration of fines from unconsolidated reservoirs. In all of these cases, particulate size distribution appears to be the only current criterion adopted as a basis for the design of the following: P. 355

As the first step in theoretical development of a fully coupled mathematical model for transport of immiscible contaminants in unsaturated soils, a conceptual model is introduced. The conceptual model is an expression of what happens in the nature. In this paper, the conceptual model, different components of the system and principal factors governing their interaction have been specified based on available theoretical and experimental works. Mathematical description of this conceptual model in term of suitable state parameters is then presented and a numerical solution for the model is performed by the finite element method using weighted residual approach.

In this paper, certain fluid recovery mechanisms that occur in stratified systems are analysed using experimental floods in layered beadpacks. In particular, the roles of viscous and gravitational crossflow are studied in viscous stable and viscous slug type displacement processes. The results from these floods are then scaled to field processes using a combination of scaling theory and numerical simulation. The effects of smaller sub-layer heterogeneity (within the larger scale layering) on the fluid recovery mechanisms are also examined in some detail. It is found that, although the levels of smaller scale heterogeneity studied in this work does not largely affect the total recovery profiles, they do have some impact on the detailed levels of fluid crossflow in the system. However, it is stressed that the level of heterogeneity description required for a given system must depend both on the type of system, the details of the fluid recovery mechanism operating in that reservoir and also actual balance of forces (viscous/gravity etc.) that exists in that participle situation. Introduction One of the most common idealizations used by reservoir engineers to represent a complex heterogeneous formation is the layered reservoir model [e.g. 1–11]. There is, however, a good foundation in geology for this first approximation to certain reservoir formations, although it is known to be very simplified in some respects. For example, the next level of complexity that can be envisaged within such a model is to consider sub-layer heterogeneity; that is with permeability heterogeneity within each layer. It is also possible to imagine further transformation of such a layered system (at least in principle) into one where local length correlations within the sub-layers start to dominate over the stratification structure, thus leading to so-called laborynth and jig-saw type reservoir structures. Here, we consider physical systems - both reservoirs and experimental packs where the stratified model is still valid. However, we also wish to consider the extension of the stratified reservoir concept when sub-layer heterogeneity is present which, as we note above, is a first step towards modelling a realistic system. In particular, we wish to consider the question: how particular, we wish to consider the question: how important is it to describe this sublayer heterogeneity compared with the overall description of the larger scale stratification? In fact, without further qualification, this question is not very meaningful since we must consider it in the light of a particular fluid recovery process in that reservoir system. In other process in that reservoir system. In other words, the importance of any level of reservoir heterogeneity - including the larger scale layering - must pertain to the fluid displacement or oil recovery pertain to the fluid displacement or oil recovery mechanism that is operating. Indeed, not only is the fluid recovery process itself important in considering the relevance of questions about heterogeneity, but the balance of forces may also be very important. That is, the relative importance of viscous gravitational and capillary forces may also dictate the level of heterogeneity description that is necessary in order to model a given process. In the work presented in this paper, we examine some specific fluid flow processes in paper, we examine some specific fluid flow processes in layered systems as described below. The scaling from experiment to field systems is first considered and then the effects of heterogeneity are examined using numerical simulation. Many experimental and simulation studies have appeared over the years which discuss various aspects of flow in layered systems. P. 485

In this paper the details are presented of a software package which represents a new strategic approach to the selection of an optimum gravel size for effective sand control. The new approach not only considers the formation sand size distribution in the choice of the gravel to use, it goes further to evaluate the response of the probable gravels to operating conditions and the overall effect on the well production capacity This software package which operates in a DOS 'pseudo' Window environment is developed on the premise that the best gravel size selection is a function of not only the formaton sand size distribution but also other characteristics such as sand sorting and shape, gravelpack structure which dictates the pack pore throat distribution, and operating conditions and the overall effect on the well production capacity.The design and evaluation tool is made up of:(a). A gravelpack structure simulator (package 1) whose primary function is to carry out a 2-0/3-0 simulation of the gravelpack structure from which the maximum pore throat distribution and corresponding poroperm data and pack structure coefficient are computed. (b). The gravel sizing package(package 2) which is the main subject of this paper is made up of three distinct but interrelated modules namely, the pack structure module, the bridging efficiency module and the well performance module. The possible gravels to use are selected with the pack structure module which permits history matching of the formation sand size distribution with the corresponding pore throat distribution of select gravels. The bridging efficiency module thereafter evaluates the bridging effectiveness of the selected gravels andReferences and illustrations at the end of paper 37 the probable gravel is finally selected on the basis of the overall well performance evaluation.Specific case studies have been considered to demonstrate the use of the package.

In this paper, the basic concepts of a 3-D random gravel packing modelling technique being developed to simulate typical gravelpack structures are presented. This is a followup on the 2-D modelling technique which was initiated in the Department of Petroleum Engineering at Heriot-Watt University as part of an on-going project on optimum gravel sizing for effective sand control.