排序方式: 共有38条查询结果,搜索用时 15 毫秒
1.
Hamidreza Yazdani Sarvestani 《Advanced Composite Materials》2017,26(3):273-294
In this study, a new simple-input displacement-based method is used to study effects of layup sequences on stresses, strains, and deformations of thick laminated orthotropic cantilever straight tubes under transverse loading. Three-dimensional stress distributions are obtained based on the most general displacement field of elasticity. A layer-wise theory which includes the full three-dimensional constitutive relations is used. A non-dimensional simple coefficient is introduced to compare interlaminar radial stresses of different layup sequences. Finally, some design guidelines are proposed to consider effects of layup sequences of laminated thick composite tubes subjected to shearing load. 相似文献
2.
《国际智能与纳米材料杂志》2013,4(2):85-104
The present article considers the free-vibration analysis of plate structures with piezoelectric patches by means of a plate finite element with variable through-the-thickness layer-wise kinematic. The refined models used are derived from Carrera’s Unified Formulation (CUF) and they permit the vibration modes along the thickness to be accurately described. The finite-element method is employed and the plate element implemented has nine nodes, and the mixed interpolation of tensorial component (MITC) method is used to contrast the membrane and shear locking phenomenon. The related governing equations are derived from the principle of virtual displacement, extended to the analysis of electromechanical problems. An isotropic plate with piezoelectric patches is analyzed, with clamped-free boundary conditions and subjected to open- and short-circuit configurations. The results, obtained with different theories, are compared with the higher-order type solutions given in the literature. The conclusion is reached that the plate element based on the CUF is more suitable and efficient compared to the classical models in the study of multilayered structures embedding piezo-patches. 相似文献
3.
前向神经网络的神经元分层逐个线性优化快速学习算法 总被引:1,自引:0,他引:1
本文提出了一种新的前向神经网络快速分层学习算法.在此学习算法中,其优化策略为对输出层和隐层神经元的连接权值交替优化.对输出层权值优化算法采用基于广义逆的最小二乘递推算法,对隐层神经元的连接权值采取则对每个神经元逐个进行优化,而且采用正交变换加快每一步学习的计算速度和提高算法的数值稳定性.当学习过程停滞时采用随机扰动的方法摆脱过早收敛.数值实验表明,与BP动量因子法、牛顿型方法和现有的分层优化算法相比,新算法不仅学习速度快学习时间短,而且当网络规模增大时仍然比较有效. 相似文献
4.
5.
6.
Due to the nested nonlinear structure inside neural networks, most existing deep learning models are treated as black boxes, and they are highly vulnerable to adversarial attacks. On the one hand, adversarial examples shed light on the decision-making process of these opaque models to interrogate the interpretability. On the other hand, interpretability can be used as a powerful tool to assist in the generation of adversarial examples by affording transparency on the relative contribution of each input feature to the final prediction. Recently, a post-hoc explanatory method, layer-wise relevance propagation (LRP), shows significant value in instance-wise explanations. In this paper, we attempt to optimize the recently proposed explanation-based attack algorithms (EAAs) on text classification models with LRP. We empirically show that LRP provides good explanations and benefits existing EAAs notably. Apart from that, we propose a LRP-based simple but effective EAA, LRPTricker. LRPTricker uses LRP to identify important words and subsequently performs typo-based perturbations on these words to generate the adversarial texts. The extensive experiments show that LRPTricker is able to reduce the performance of text classification models significantly with infinitesimal perturbations as well as lead to high scalability. 相似文献
7.
地基非线性沉降计算的原状土割线模量法 总被引:7,自引:0,他引:7
地基的沉降计算如何较准确而又简便是工程实践中一直未很好解决的问题。为较好解决地基沉降计算这一重要难题,提出一种原状土割线模量法,就是利用原状土的压板试验曲线,假定压板试验曲线符合双曲线方程,利用Bussinesq解,建立不同荷载水平下土体的等效割线模量。对于实际基础,则根据土体不同深度的应力水平,从压板试验曲线确定原状土的割线模量,然后用于分层总和法进行地基非线性沉降计算,通过试验曲线的验证和实际工程的应用,证明该方法效果较好,可较准确计算地基的沉降。 相似文献
8.
地基非线性沉降计算的原状土切线模量法 总被引:22,自引:3,他引:22
根据原状地基的载荷试验曲线,建立原状土的切线模量与应力水平关系的切线模量方程,对地基不同点根据其应力水平由切线模量方程确定计算点原状土的切线模量,用该切线模量对地基沉降进行分层总和法计算,其特点是切线模量是由原位试验得到的,能反映原状地基土的特点,同时考虑应力水平的影响,反映了土的非线性特点。 相似文献
9.
以卷积神经网络为代表的深度学习算法高度依赖于模型的非线性和调试技术,在实际应用过程中普遍存在黑箱属性,严重限制了其在安全敏感领域的进一步发展。为此,该文提出一种由粗到细的类激活映射算法(CF-CAM),用于对深度神经网络的决策行为进行诊断。该算法重新建立了特征图和模型决策之间的关系,利用对比层级相关性传播理论获取特征图中每个位置对网络决策的贡献生成空间级的相关性掩码,找到影响模型决策的重要性区域,再与经过模糊化操作的输入图像进行线性加权重新输入到网络中得到特征图的目标分数,从空间域和通道域实现对深度神经网络进行由粗到细的解释。实验结果表明,相较于其他方法该文提出的CF-CAM在忠实度和定位性能上具有显著提升。此外,该文将CF-CAM作为一种数据增强策略应用于鸟类细粒度分类任务,对困难样本进行学习,可以有效提高网络识别的准确率,进一步验证了CF-CAM算法的有效性和优越性。 相似文献
10.