【摘要】 磁流变弹性体（magnetorheological elastomers，简称MRE）的力学性能呈现复杂的非线性特性，建立MRE减震器的力学模型以表征其动力学特性是进行智能振动控制应用的关键。针对有参模型参数识别困难、无参模型易陷入局部最优等问题，根据MRE减震器的力学特性试验结果，建立思维进化算法（mind evolution algorithm，简称MEA）优化的BP神经网络模型来描述MRE减震器的力学特性，并对比了参数化建模与非参数化建模的差异性。研究结果表明：线性K-C模型仅能描述MRE减震器的线性力学特性；Bouc-Wen模型能较为准确地表征其中心对称非线性力学特性；MEA-BP神经网络能准确预测MRE减震器的非线性力学特性。研究成果为MRE减震器的设计及应用提供了参考。更多还原

【Abstract】 The mechanical properties of magnetorheological elastomers(MRE) exhibit complex nonlinear characteristics. The key of intelligent vibration control is to establish the mechanical model of MRE shock absorber to characterize its dynamic characteristics. Aiming at the problems of parameter identification of parametric model and local optimization of nonparametric model, according to the experimental results of mechanical characteristics of MRE shock absorber, a back propagation(BP) neural network model optimized by mind evolution algorithm(MEA) is established to describe the mechanical characteristics of MRE shock absorber, and the differences between parametric modeling and nonparametric modeling are compared. The results show that the linear K-C model can only describe the linear mechanical properties of MRE shock absorber, while the Bouc-Wen model can accurately describe the nonlinear mechanical properties of its central symmetry, MEA-BP neural network can accurately predict the nonlinear mechanical characteristics of MRE shock absorber. The research results provide a reference for the design and application of MRE shock absorber.更多还原