Steel cables are critical load-bearing components in modern infrastructure, and their tension is a core indicator for assessing structural safety. Conventional elasto-magnetic (EM) methods for tension estimation often suffer from limited accuracy, primarily due to their reliance on low-order models and inadequate feature extraction. To address these limitations, this study proposes an adaptive nonlinear EM sensing method under pulsed excitation. The integrated area of the induced voltage during the falling edge of the pulse is extracted as the key feature and its inherent nonlinear relationship with tensile force is modeled using polynomial regression. An optimization algorithm is developed to automatically determine the model-order-dependent optimal integration threshold by maximizing the coefficient of determination (R2), thereby enabling adaptive and robust feature extraction. Experiments were conducted on a steel strand specimen, and polynomial models of orders 1–6 were systematically evaluated. The results indicate that the optimal integration thresholds are consistently concentrated in the low-amplitude region of the signal and that the fourth-order polynomial model achieves the highest prediction accuracy. By synergistically integrating physical interpretability with data-driven optimization, the proposed method provides a high-precision solution that significantly outperforms existing EM techniques for cable tension estimation.
Liu et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: