Horizontal Shale Well EUR Determination Integrating Geology, Machine Learning, Pattern Recognition and MultiVariate Statistics Focused on the Permian Basin

摘要

The objective of this work is to accurately determine horizontal shale well EUR for an area integrating geology, machine learning, pattern recognition and statistical analysis using various parameters of nearby producing horizontal shale wells, as inputs. This work utilizes local geological information followed by execution of machine learning to identify critical well parameters that lead to better production. Then a pattern recognition step is performed while making sure the number of wells in each category are statistically significant. This also serves as a quality control measure by not basing the conclusion solely on the results of machine learning. The conclusions are verified using available literature on correlation between well production and various well parameters. Wells with optimum (controllable) parameters are selected to obtain a type curve for the target zone(s) in the area of interest. The above-mentioned methodology helped in making the type-curve/EUR determination process scientific, systematic and seamless. Machine learning helped in identifying the key well-parameters that correlate to better production. Visual pattern recognition strengthened the confidence in the relationships identified in the last step. Different parameters were shown to affect production in different areas/targets confirming that every shale asset requires a thorough research before reaching a reasonable conclusion. The type-curves were established for each Wolfcamp bench in the area of interest selecting wells with optimum completions. The optimum completions parameters were identified by the methodology prescribed in the paper. This assisted in identifying the area of interest's true economic potential with regards to horizontal shale well development. This paper prescribes a novel scientific data-intensive methodology to systemically use well data in a step-wise manner to identify the type-curve and EUR/well for an area, thereby determining the area's true economic potential. Along with the prescribed big-data mining methodology, the most important take away from this study is: for the optimum evaluation of shale assets it is critical to tie in the controllable well parameters to well production. Once this relationship is established, the type-curve determination and the EUR estimation can be done more accurately.