Abstract
A fresh or different perspective is essential to discovering new opportunity for successful field development. This paper summarizes a new type of analysis applied to reservoir, well, completion, and fracture stimulation information for the western shallow oil zone (WSOZ) completions in the Elk Hills Field of Kern County, California. The focus of this analysis was to identify parameters that are indicators of reservoir quality and to evaluate the effectiveness of stimulation/completion methods.
This evaluation used unstructured modeling methods supported by the use of standard engineering tools. Of particular interest was the identification of ways to help enhance the economic success of WSOZ completions at Elk Hills. When the information derived from this study was applied to new completions, well economics improved significantly. This improved success has justified additional infield development drilling. Specific objectives included:
Developing a model or tool to predict production for wells in the field
Identifying parameters that are indicators of reservoir quality
Determining which completion/stimulation methods and designs have been successful in the past and using this information to improve the success of future completions
Using the above understandings to improve fracture candidate selection
A general objective was to test the applicability and benefits of holistic modeling methods, such as using artificial neural network (ANN) technology to model WSOZ wells at Elk Hills. In this instance, ANN modeling was used to differentiate the contributions of reservoir quality from the stimulation/completion method on a given well production result. The understanding derived from this study influenced decisions about pay selection, perforating practices, operational procedures, and stimulation fluid and proppant selection.
