Predicting Optimum Shut-In of Wells in Peripheral and Line Drive Waterfloods

Abstract

Here is a method for computing the rate, reserve, and economic performance of floods as a function of water production at the time a well is shut in. By using a steady-state, stratified reservoir computer model to predict performance of producing wells shut in at different percentages of water, performance of producing wells shut in at different percentages of water, the most economical operating plan can be determined. Introduction Many fields have been waterflooded using a conventional five-spot pattern where no troublesome decision had to made as to when the producers should be shut in - one merely kept producing until the economic limit was reached. Recently, more and more peripheral and line drive patterns have been chosen peripheral and line drive patterns have been chosen as the proper operation. Many of these are cycling 75 to 90 percent of the injected fluid through some of the producers; yet, at the same time, other producers have scarcely responded. producers have scarcely responded. The operational decision that is faced is when to shut in the high water producers. Up to a point, the longer the producing wells are kept on production, the better the volumetric sweep of the water front will be. On the other hand, tardy shut-in of the wells will lengthen the flood's life, increase operating costs, and decrease the flood's dollar value. Optimization of the operation is what must be achieved. To optimize, a series of performance calculations must be made under varying assumptions of shut-in water cut (percentages of water produced) for the first row of wells, the second row, and so forth. Naturally, the last row of producing wells must not be shut in until the flood's economic limit is reached. The reservoir phenomena that must be dealt with in this type of analysis include:the effects of mobility ratio,the growth of areal sweep with water cut,the effects of reservoir stratification,well productivities, and in some casesthe effects of cross-flow. In the past, many papers have described models used to evaluate producing performance. These models range from a simple, inverted five-spot pattern program to two-dimensional numeric models pattern program to two-dimensional numeric models capable of predicting field-wide performance under the various assumed producing conditions. In this paper we will describe the use of a simplified model paper we will describe the use of a simplified model that is computationally fast enough to make it especially desirable in handling optimization studies. Model Description Fig. 1 illustrates how the model used in this paper visualizes a portion of a field. The area included in the model normally covers the area affected by one injection well. Within this area, the element is viewed as being composed of layers. Normally, five layers are sufficient. A throughput or rate technology incorporated in the model is based on that for a line drive pattern with one injection well located at one end of the pattern and producing wells placed at varying distances pattern and producing wells placed at varying distances toward the other end. The flexibility of the program is illustrated by the patterns represented in Fig. 2. The simplest system consists of one injection well and one producer (as illustrated, only a portion of the injection volume enters the pattern and a portion of the production flows from another pattern into the producer). producer). JPT P. 497