Abstract Magnetorheological gel (MRG) excels in the material properties in term of adjustability and sedimentation performance, which could upgrade the performances of the current magnetorheological fluid based adjustable devices for structural control and vibration mitigation. However, the characterization and modelling of the stress-strain hysteresis responses of MRG has not been reported in the past, which are fundamental steps towards engineering applications. In this study, the stress-strain hysteresis of polyurethane based MRG sample with 60% carbonyl iron particle weight fraction was characterized under sinusoidal shear excitations with broad ranges of strain amplitude (5%–100%), excitation frequency (0.1 Hz–2 Hz) and magnetic fields (0–0.91 T). Significant stress overshooting phenomenon were observed under the application of low magnetic fields (0.27 T). A structurally-simple and accurate phenomenological model has been established to capture this unique nonlinearity. By validating the experimental results, the proposed model accurately predicts the hysteretic behaviour and the overshoot of the MRG under the excitation scenarios and the magnetic fields considered. Finally, the support vector machine (SVM) was implemented to provide the solution to the model generalization. The SVM-assisted model showed good agreement with the experimental data and can benefit the efficiency and viability in developing controllable MRG-based devices and system.

中文翻译：

---

磁流变凝胶非线性粘弹性行为的四参数现象学模型的开发

摘要 磁流变凝胶（MRG）在材料性能的可调性和沉降性能方面具有优异的性能，可以提升目前基于磁流变流体的可调装置的结构控制和减振性能。然而，MRG 的应力-应变滞后响应的表征和建模在过去尚未报道，这是工程应用的基本步骤。在这项研究中，具有 60% 羰基铁颗粒重量分数的聚氨酯基 MRG 样品的应力-应变滞后在具有广泛应变幅度 (5%–100%)、激发频率 (0.1 Hz–2 Hz) 的正弦剪切激发下表征) 和磁场 (0–0.91 T)。在低磁场（0.27 T）的应用下观察到显着的应力超调现象。已经建立了结构简单且准确的现象学模型来捕捉这种独特的非线性。通过验证实验结果，所提出的模型准确地预测了磁滞行为和 MRG 在所考虑的励磁场景和磁场下的过冲。最后，实现了支持向量机 (SVM) 以提供模型泛化的解决方案。SVM 辅助模型与实验数据显示出良好的一致性，有助于提高基于 MRG 的可控设备和系统的开发效率和可行性。通过验证实验结果，所提出的模型准确地预测了磁滞行为和 MRG 在所考虑的励磁场景和磁场下的过冲。最后，实现了支持向量机 (SVM) 以提供模型泛化的解决方案。SVM 辅助模型与实验数据显示出良好的一致性，有助于提高基于 MRG 的可控设备和系统的开发效率和可行性。通过验证实验结果，所提出的模型准确地预测了磁滞行为和 MRG 在所考虑的励磁场景和磁场下的过冲。最后，实现了支持向量机 (SVM) 以提供模型泛化的解决方案。SVM 辅助模型与实验数据显示出良好的一致性，有助于提高基于 MRG 的可控设备和系统的开发效率和可行性。