Simulation Study of Development Strategies for a Gravity-Assisted, Cyclic-Steam Project

Abstract

Abstract New developments of a simulation study of a mature cyclic steam injection project in California are presented in this paper. Decline of production rates was observed in the field, and it forced the operator to look into strategies of infill drilling and recompletion of existing wells. Horizontal infill wells, recompleted vertical wells, and mixed strategies were considered as the alternatives for further development of the reservoir. In 1996 Chona et al. presented the results of a simulation study for this field, where productive sands of a 60 degrees dip were modeled as vertical layers. In the presented simulation study, a new model is used. The model includes the actual reservoir dip. Since the reservoir consists of 8 different sand layers with a limited communication, including the actual dip in the model leads to significant changes of the model performance. Using the developed model, a number of development strategies were simulated. Results of the study are compared to results of Chona et al. to illustrate changes caused by including reservoir dip in the model. Details and description of simulated alternative strategies are presented. For horizontal infill wells, results indicate strong dependence of the cumulative production on the well spacing, injection rate, and well location with reference to the oil column. Strategies are compared on the basis of cumulative oil production. All the presented strategies result in an increase of cumulative production. Introduction The Midway Sunset field is located in Kern County of Southern California (Fig. 1). It is a large oil field, with ultimate reserves of 2.73 billion barrels. Due to the low gravity (14 API) and high viscosity (2,300 cp. at 95 CF), thermal recovery techniques have been employed to produce oil from this field. Berry Petroleum Company has been operating this field using steam assisted gravity drainage, with approximately 7, 500 STB/D of oil production. The geology of Monarch Sands consists of thin units (2 to 10 ft.) aggregating about 1,000 ft. in stratigraphic thickness. The beds dip 60 to 70 degrees and the trap is provided by a truncating unconformity with 20 degrees dip (Fig. 2). The sands show high porosity (about 30%), high oil saturation (approx. 80%), and high OOIP of 1,800 barrels/acre-ft with an original oil column from 0 to 1,000 ft. Wells are tightly spaced (1/4 acre well spacing on average). A cyclic steam assisted gravity drainage with vertical wells at small spacing has been effective. Residual oil saturation is typically 10% in the depleted zones. During the operation, a large steam chest in the upper part of the reservoir has developed. The average steam chest pressure is low (about 18 psi) as the result of steam cycling with balanced injection and production rates. A decline of production rates was observed in the field. One of the alternatives to increase production is to recomplete vertical wells in the oil zone, currently up to 400 ft thick. Recompletion of all the vertical wells would not be practical because of the high cost. The other alternative is to drill horizontal infill wells. The major objective of the presented study was to evaluate a number of strategies for recompletion of vertical wells and drilling of infill horizontal wells. Initially, a conceptual model was constructed. The model accounts for all key features of the reservoir and includes the actual reservoir dip of 60 degrees. Using the model, current average oil and water production rates were matched for a simulated section of the field. After the match, a number of strategies including recompletion of vertical wells, infill horizontal wells, and mixed strategies were simulated. P. 291^