Finite element analysis (FEA) is a powerful tool in OCTG connection design and qualification. Connection designers use FEA to examine impacts of design variables on connection performance. FEA is also used in connection evaluation protocols to select configurations of full-scale test specimens, and in comparative assessments to qualify connection product lines. While FEA has become a common tool, analysis of OCTG connections remains complex, and presents numerous challenges related to modelling assumptions, mathematical approximations and solution accuracy – which can vary depending on the inputs, methods and software. These challenges sometimes lead to variation in the quality of connection analysis results. Given increasing industry use for connection design and qualification, the ability to verify correctness and increase confidence in FEA performed by a third party has become an important goal – but no guidelines for such verification are available.
This paper broadly describes a benchmarking process that can be used to confirm that FEA simulations are performed to a suitable level of rigour for the connection modelling application of interest. The process steps include identifying key results of interest, establishing the governing physics and modelling parameters that could affect those results, determining the required level of benchmarking rigour, devising appropriate cross-checks that can be compared to the analysis results, and conducting those checks to increase confidence in the FEA model.
We reference thermal well casing premium connections as an example of a tubular application where FEA is relied upon heavily to support both connection design and service suitability evaluations. Specifically, the Thermal Well Casing Connection Evaluation Protocol (TWCCEP, published as ISO PAS 12835) employs finite element analysis as a means to select the connection configurations for full-scale testing. Connection analysis is used to determine the worst-case combinations of machining tolerances, pin material strength and make-up torque for galling and sealability evaluations. The modelling requirements for this extreme-service application include a number of nonlinearities, and appropriate benchmarking of the FEA is essential if confidence in the analysis results is to be achieved.
This perspective on FEA benchmarking is a suggestion for a more systematic approach to confirming connection analysis is well-conditioned, with flexibility to accommodate the specifics of the connection application and the need for rigour in FEA results. Use of this process for ISO PAS 12835 evaluations and other connection analyses purposes will increase end-user confidence and facilitate effective use of FEA by the oil-and-gas industry.