Unconventional shale resources are key hydrocarbon sources, gaining importance and popularity as hydrocarbon reservoirs both in the United States and internationally. Horizontal wellbores and multiple transverse hydraulic fracturing are instrumental factors for economical production from shale assets. Hydraulic fracturing typically represents a major component of total well completion costs, and many efforts have been made to study and investigate different strategies to improve well production and reduce costs. The focus of this paper is completion effectiveness evaluation in different parts of the Eagle Ford Shale Formation, and our objective is to identify appropriate completion strategies in the field.
A data-driven neural network model is trained on the database comprised of multiple operators' well data. In this model, drilling and mud data are used as indicators for geology and reservoir-related parameters such as pressure, fluid saturation and permeability. Additionally, completion- and fracture-related parameters are also used as model inputs. Because wells are pressure managed differently, normalized oil and gas production is used as a model output. Thousands of neural networks are trained using genetic algorithm in order to fully evaluate hidden correlations within the database. This results in selection of a neural network that is able to understand reservoir, completion and frac differences between wells and identify how to improve future completion/stimulation designs.
The final neural network model is successfully developed and tested on two separate data sets located in different parts of the Eagle Ford Shale oil window. Further, an additional test data set comprised of eight wells from a third field location is used to validate the predictive usefulness of the data-driven model. Under-producing wells were also identified by the model and new fracture designs were recommended to improve well productivity.
This paper will be useful for understanding the effects of completion and fracture treatment designs on well productivity in the Eagle Ford. This information will help operators select more effective treatment designs, which can reduce operational costs associated with completion/fracturing and can improve oil and gas production.