A Framework Incorporating Resistance Optimization for Rapid Design and Validation of 3D-Printed Bridge Pier Geometry

Abstract

This study presents a rapid design-and-validation framework incorporating resistance optimization for 3D-printed bridge pier geometries. A full-factorial experimental campaign comprising 16 reduced-scale solid pier sections is first conducted by systematically varying upstream fairing length and downstream fishtail length while maintaining constant maximum transverse width and cross-sectional area. The specimens are fabricated using fused deposition modeling (FDM) 3D printing with polyethylene terephthalate glycol-modified (PETG) material and tested in controlled towing experiments driven by a field-oriented control (FOC) motor, with motor torque signals recorded as a proxy for hydrodynamic resistance. The raw data are processed through steady-state trimming, null-test bias correction, and one-dimensional Kalman filtering, after which a root-mean-square (RMS) resistance metric is computed for each geometry. A key finding from the two-way analysis of variance (ANOVA) analysis reveals that fairing length exerts the dominant influence on resistance, followed by the fairing-fishtail interaction, whereas fishtail length alone plays a secondary role. The experimental ranking identifies S13, combining a short fairing with a long fishtail, as the optimal geometry, achieving a 33.6% reduction in RMS torque relative to the circular baseline. Bootstrap resampling confirms that the low-resistance cluster is a robust geometry family rather than a statistically fragile optimum. Then, independent COMSOL Multiphysics 6.3 (COMSOL) topology optimization and transient flow-field simulations are employed as morphology-level validation tools, with the optimized outline converging toward a streamlined profile qualitatively consistent with the experimental findings. The drag decomposition further indicates that pressure drag constitutes the dominant component, suggesting that shape-induced pressure redistribution is the primary mechanism underlying resistance reduction. The proposed framework thus provides a physically grounded, low-cost intermediate step between computational shape generation and detailed engineering validation for resistance-optimized bridge-pier sections, and it can be readily extended to a broader range of pier cross-sections or other hydraulic structures.