A simulation-based and data-augmented shear force inversion method for offshore platform connector

This study introduces a Simulation-Based and Data-Augmented method for shear force inversion to address the challenge of directly measuring shear force on connector pins in multi-module floating platforms. Stress sensors are strategically placed in adjacent areas. Extensive Finite Element simulation scenarios lead to the identification of optimal features sensitive to both force magnitude and direction. Subsequently, an Artificial Neural Network (ANN) is developed to distill the simulation data into characteristic sensor responses. Fine-tuning with physical measurements further enhances shear force inversion accuracy. Using simulated and experimental data, the method demonstrates a shear force inversion error below 3.2 % and an angular inversion error under 1.4 % across test conditions. This methodology provides essential load data for connector safety assessments and crucial guidelines for the assembly of multi-module floating platforms.