基于遗传算法优化的SMABP神经网络本构模型

发布时间：2018-04-09 20:02

本文选题：形状记忆合金(SMA)　切入点：遗传算法　出处：《金属学报》2017年02期

【摘要】：系统研究了形状记忆合金丝(SMA)应力-应变曲线、特征点应力、耗能能力及等效阻尼比随材料直径、应变幅值、加载速率、加载循环次数的变化规律;由于SMA唯象Brinson等常见本构模型无法以数学模型方式精确描述SMA各影响因素对其力学性能的影响程度,基于SMA实验结果,本工作采用BP神经网络智能算法(一种利用误差反向传播训练的神经网络算法)对其进行非线性建模,同时利用遗传算法对神经元的初始权值和阈值进行优化,进而获得了一种基于遗传算法优化的SMA BP神经网络本构模型。利用该模型对SMA实验结果进行模拟,所得结果平均误差仅为1.13%,优于未优化的SMA BP神经网络模型。结果表明,基于遗传算法优化的SMA BP神经网络本构模型,能够精确地预测SMA在反复荷载作用下的超弹性性能,避免由于初始权/阈值取值不当引起的BP网络振荡而产生不收敛的问题,同时也充分考虑了加/卸载速率的动态影响,是一种良好的速率相关型动力本构模型。
[Abstract]:The stress-strain curve, characteristic stress, energy dissipation capacity and equivalent damping ratio of shape memory alloy wire (SMA) with material diameter, strain amplitude, loading rate and loading cycle times are systematically studied.Because SMA phenomenological Brinson and other common constitutive models can not accurately describe the degree of influence of various factors of SMA on its mechanical properties by mathematical model, based on the results of SMA experiments,In this work, BP neural network intelligent algorithm (a neural network algorithm using error back propagation training) is used to model its nonlinear model, and genetic algorithm is used to optimize the initial weights and thresholds of neurons.Then a constitutive model of SMA BP neural network based on genetic algorithm optimization is obtained.The model is used to simulate the experimental results of SMA. The average error of the obtained results is only 1.13, which is better than that of the unoptimized SMA BP neural network model.The results show that the constitutive model of SMA BP neural network based on genetic algorithm can accurately predict the hyperelastic performance of SMA under repeated loads.The problem of non-convergence caused by BP network oscillation caused by improper initial weight / threshold value is avoided, and the dynamic effect of loading / unloading rate is fully considered. It is a good rate dependent dynamic constitutive model.
【作者单位】：西安建筑科技大学土木工程学院;长安大学建筑学院;
【基金】：国家自然科学基金项目No.51678480 陕西省科技统筹创新计划项目No.2013SZS01-S02 陕西省工业攻关项目No.2014K06-34~~
【分类号】：TG139.6

【参考文献】