Tight Gas Well-Performance Evaluation With Neural Network Analysis for Hydraulic Propped Fracture Treatment Optimization

Abstract

This paper summarizes the application of data mining techniques and neural network multi-variable analysis to evaluate critical parameters impacting well performance and optimize hydraulic fracture treatment design for tight gas well completions in the Pinedale Anticline field in western Wyoming. Challenges and uncertainty associated with single variable analysis are discussed with consideration of the complexity of reservoir parameter variability and nonsequential coincident completion variable modifications. Neural network model development is summarized with emphasis on applications for well performance evaluation and hydraulic fracture treatment optimization. Results are presented from extensive trials performed during 2009 development activity at Pinedale.