Actual Versus Predicted Horizontal Well Performance, Weyburn Unit, S. E. Saskatchewan

Abstract

Abstract The Weyburn Unit, located in the southeastern corner of Saskatchewan, has over 80 horizontal wells. These wells are producing from a complex, naturally fractured reservoir that has been on waterflood for over 30 years. Accurate prediction of horizontal well performance is required for waterflood development planning, facilities design and marketing. For more precise estimates, simulation modelling has been the tool of choice to incorporate the reservoir anisotropy and the effect of having a multi-layered reservoir. Specific area models were built, but the Weyburn Unit spans 83 Sections with over 160 potential horizontal well locations making full field simulation impractical. Therefore, analytical models, because of their flexibility, were used to estimate the performance of horizontal wells field-wide. To validate predictions from both the simulation and analytical models, all available information sources were combined. These included:off-trend vertical well performance,existing horizontal well performance in the Weyburn Unit and other Midale beds pools, andgeological models. As a result of this work, we were able to calibrate a published analytical model to better predict horizontal well performance in this complex reservoir and to optimize the design of future' wells. Introduction The Weyburn oil pool, located in southeastern Saskatchewan, was discovered in 1954 (Figure 1). By 1960, the field was fully delineated with vertical wells on 32 ha (80 acre) spacing. In 1964, it was placed on waterflood with 151 inverted nine-spot patterns. A year later, oil production peaked at 7,500 m3/d and then declined until 1985. Production decline was arrested with the drilling of 157 vertical infill wells from 1985 to 1992. About 27% of the 178 106m3 original-oil-in-place has been produced to date. A medium gravity crude oil (27 ° API) is produced from the Midale Beds of the Mississippian Charles Formation. These beds are divided into an upper chalky dolostone called the Marly and a lower limestone called the Vuggy. The original inverted nine-spot waterflood has been successful in sweeping the Vuggy, but has not been efficient in sweeping the Marly. The vertical infill wells targeted the Marly zone. By 1991, the number of potential economic vertical infilllocations was diminishing, so the Unit turned to horizontal drilling technology for further development. During 1991 and 1992, six horizontal wells were drilled, experimenting with different geological environments and well orientation. The lessons learned from this pilot were used in the design and execution of the commercial horizontal well development in 1993. Through a cost/benefit analysis it was decided to drill long horizontal wells (1,000 m) that parallelled the fracture trend. These wells were drilled underbalanced. This development drilling is expected to continue throughout most of 1995 bringing the total number of horizontal wells in the Weyburn Unit to over 80. Accurate forecasts of horizontal well production rate and reserves allow:a proper assessment of economic reward with infill drilling,a framework in which capital and operating costs can be controlled, andlead time to arrange oil transport to markets.