Data Driven Physics-Guided Casing Fatigue Life Estimation

Abstract Casing pipe or joint fatigue failure could happen either during drilling in rough offshore operations or production due to alternating temperatures and cyclic loading of pressure or during stimulation, specifically, multistage hydraulic fracturing, due to temperature differentials between the reservoir fluids and stimulation fluids. In all those cases, those cyclic variations induce thermal axial casing stress (compressive and tensile) depending on the direction of the change. As a result, two major effects may occur from those thermally induced stresses 1) casing hot-yield and the resultant casing collapse failure, and 2) casing fatigue. These cyclic loads to the casing structure often result in detrimental fatigue failures at essential casing joints and welds. Wrong cyclic strain calculations will lead to false estimation of casing fatigue life. This, in turn, would lead to abuse of casing material beyond its capability and increased possibility of failure. This paper highlights two instances where the predicted fatigue life was almost 10 to 600-fold higher than the actual fatigue life that was actually observed after failure. The authors attributed a few factors that may have led to an unexpected increase in total cyclic change of local strain at casing connections. Although authors in both Cases 1 & 2 managed to successfully account for fatigue effects in their casing design workflow; using various casing fatigue models (e.g., Manson Universal Slope Method and Energy-Based Fatigue Model), their conclusions were utterly false and far from observed values in history data. The major challenging aspect of the implemented physics-based models was their inability to account for various effects that were later found to have a great impact on the durability of different casing parts. Those factors are passive in nature, hence, can't be directly included in the implemented fatigue models. The correct calculation of total cyclic change of local strain at concentration points is key to the accurate estimation of fatigue life of various casing parts and, in turn, mitigation of casing failure in the future. However, accurate estimation of cyclic local strain could be challenging due to multiple factors. Using the proposed modified model, those passive effects can be integrated and accounted for without the need for explicit mathematical formulation. To mitigate casing failure that results from fatigue, correct estimation of fatigue life of different casing parts must be made. Two classes of factors impact local strains: 1) active/direct (such as temperature changes, casing material, etc.) and (2) passive/indirect factors (such as cement cracks or leaks, casing-cement contact pressures, material imperfections/wrong field handling, etc.). Although conventional models can account for direct factors, they are bound to fail to account for indirect factors, leading to false conclusions on casing fatigue life and abusive consumption of casing parts beyond their capabilities. In this paper, a data-driven alternative is proposed that takes as input the direct and indirect factors, and outputs the corresponding total local strain that reflect those effects. Then, using the casing material properties, along with estimated strains as input for Manson's Eq. and estimating the fatigue life of those casing parts. Based on estimated fatigue life, the model can give recommendations on changing casing parts that are abused throughout any process (such as steam injection, or hydraulic fracturing). This would, ultimately, prevent or reduce the chances of the occurrence of casing failure. The proposed model showed a huge difference between results of the proposed modified data-driven model against the conventional method. A total local strain half of that calculated using conventional models was enough to keep the casing integrity for a certain amount of well overall lifetime. This would better assist drilling engineers plan and design their casing and maintain their integrity without abusing the material or risking their failure.