An investigation of casing structural integrity in shale gas horizontal wells during hydraulic fracturing using FEA and machine learning algorithms.

Mohammed, Auwalu Inuwa

Abstract

Unconventional oil and gas operators are currently facing the challenge of casing lateral buckling and/or deformation during shale gas wells stimulation through hydraulic fracturing. Failure of the casing during this process can lead to huge financial loss, catastrophic consequences and even fatalities depending on the magnitude and circumstance of the incident. An in-depth literature review was conducted, focusing on casing lateral buckling /deformation, factors attributing to casing failure, failure mechanism, and the resulting failure mode in shale gas horizontal wells. The study covered casing types, failure modes and mechanism, and impact of material selection on casing failure, with the viewpoint of casing failure mechanism, cement and rock as an integrated system. In the follow up, the casing material selections were investigated using ANSYS Granta Edupack (CES) and multicriteria decision making (MCDM) for three different scenarios of buckling tendencies. A finite element analysis (FEA) was then undertaken, using two simulation scenarios for casing structural integrity in both radial and axial configurations under the mechanics of a combine system - casing, cement and formation rock. There was then a detailed novel study on casing structural integrity using both FEA and machine learning. This research revealed the effect of combined loading, using multiple parameters to establish the relationship and effect of each on stress, displacement and ultimately casing safety factor. Finally, real-time parametric prediction and optimisation using Lunar and Quasar (ODYSSEE software package) enabled the examination of the casing structural responses based on the pertinent parameters. A similar trend was found between "KNN" and Lunar predictions in terms of how parameters influenced casing buckling phenomena and the corresponding Mises stress. In conclusion, the FEA study showed that time-dependent rock slippage-creep during stimulation lead to an increase transverse displacement and corresponding stresses on the casing. This new understanding is a major breakthrough in establishing casing health status during shale gas well stimulation. The optimised design shows 89% reduction in total deformation and 87% reduction in von Mises, in comparison to unoptimised simulation result. Also, the optimised design gives a safety factor of 3.3 against the previous 0.8129 without optimisation. The Lunar optimisation provided the ideal parameter values for the attainment of pre-defined von Mises stress as a function of other factors during the design phase. This quick approach shows both accuracy and validity, since the two independent procedures arrived at the same conclusion. We found that concurrent investigation of the attributing factors to casing buckling, alongside optimisation using FEA and the ODYSSEE package, was sufficient to maintain casing structural integrity during the shale gas extraction process. The study also revealed alternative and better suited materials choices than those that are currently preferred commercially (P110 and Q125).

Citation

MOHAMMED, A.I. 2021. An investigation of casing structural integrity in shale gas horizontal wells during hydraulic fracturing using FEA and machine learning algorithms. Robert Gordon University, PhD thesis. Hosted on OpenAIR [online]. Available from: https://doi.org/10.48526/rgu-wt-1603216

• Thesis Type: Thesis
• Deposit Date: Feb 23, 2022
• Publicly Available Date: Feb 23, 2022
• Keywords: Finite element analysis; Multi-criteria decision-making; Machine learning; Casing integrity; Oil and gas engineering
• Public URL: https://rgu-repository.worktribe.com/output/1603216
• Publisher URL: https://doi.org/10.48526/rgu-wt-1603216
• Award Date: Sep 30, 2021