(393d) Predicting Future Production for Unconventional Resources: A Data-Driven Approach

Hydraulic fracturing for horizontal wells is essential for oil and gas production from unconventional low-permeability reservoirs. A key challenge is to predict the effect of basic fracture design and construction variables with anticipated production and ultimate recovery. Decisions on such variables would optimize a related criterion, e.g. production, recovery, or NPV. Ideally, it would be desirable to understand the effect of several decision variables as well as reservoir-related parameters on the chosen objective. Fundamental equations could be used that describe phenomena related to hydraulic fracturing and subsequent production from a fractured reservoir. This is currently the prevailing approach in industry. While effective, this approach is cumbersome. As a simpler alternative, one could capitalize on the significant body of data on production from unconventional fractured wells. An opportunity exists to explore how such data can be used to develop methodologies that efficiently suggest optimal fracturing designs through the use of data-driven predictive models. Such models would capture the combined effect of model input variables, namely fracturing-job variables and reservoir properties, on the chosen objective. In a previous presentation, we explored how models can be developed that predict future production based on the initial production. This presentation focuses on a more ambitious objective, namely predicting future production, well before production commences. This would be useful, as it could help guide well placement and fracture design, both important tasks. To be able to build such a data-driven model, it is a key requirement that historical pressure data should be available along with production data. The purpose of this presentation is to explore how to build a model in that case.

Standard multivariate modeling techniques are used to build a model connecting various input variables (such reservoir properties, wellbore geometry, fracture construction process, production pressure, and others) with production.

The data-driven models developed would be useful for design purposes and what-if analyses. They can also provide insight into the variables that have strong impact on fractured horizontal well performance. Therefore, the proposed models are envisioned to offer initial decision support for well placement and fracturing design. In addition, they may will also be able to indicate parameters that should be improved/modified to get desired results in existing wells.