|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | An improved empirical wavelet transform method for rolling bearing fault diagnosis显示文摘Empirical wavelet transform(EWT)based on the scale space method has been widely used in rolling bearing fault diagnosis.However,using the scale space method to divide the frequency band,the redundant components can easily be separated,causing the band to rupture and making it difficult to extract rolling bearing fault characteristic frequency effectively.This paper develops a method for optimizing the frequency band region based on the frequency domain feature parameter set.The frequency domain feature parameter set includes two characteristic parameters:mean and variance.After adaptively dividing the frequency band by the scale space method,the mean and variance of each band are calculated.Sub-bands with mean and variance less than the main frequency band are combined with surrounding bands for subsequent analysis.An adaptive empirical wavelet filter on each frequency band is established to obtain the corresponding empirical mode.The margin factor sensitive to the shock pulse signal is introduced into the screening of empirical modes.The empirical mode with the largest margin factor is selected to envelope spectrum analysis.Simulation and experiment data show this method avoids over-segmentation and redundancy and can extract the fault characteristic frequency easier compared with only scale space methods. | HUANG HaiRun LI Ke SU WenSheng BAI JianYi XUE ZhiGang ZHOU Lang SU Lei PECHT Michael | 2020 | Science China(Technological Sciences)2020，63，11： | 10 |
| 2 | Discrete memristive neuron model and its interspike interval-encoded application in image encryption显示文摘Bursting is a diverse and common phenomenon in neuronal activation patterns and it indicates that fast action voltage spiking periods are followed by resting periods.The interspike interval(ISI)is the time between successive action voltage spikes of neuron and it is a key indicator used to characterize the bursting.Recently,a three-dimensional memristive Hindmarsh-Rose(mHR)neuron model was constructed to generate hidden chaotic bursting.However,the properties of the discrete mHR neuron model have not been investigated,yet.In this article,we first construct a discrete mHR neuron model and then acquire different hidden chaotic bursting sequences under four typical sets of parameters.To make these sequences more suitable for the application,we further encode these hidden chaotic sequences using their ISIs and the performance comparative results show that the ISI-encoded chaotic sequences have much more complex chaos properties than the original sequences.In addition,we apply these ISI-encoded chaotic sequences to the application of image encryption.The image encryption scheme has a symmetric key structure and contains plain-text permutation and bidirectional diffusion processes.Experimental results and security analyses prove that it has excellent robustness against various possible attacks. | BAO Han HUA ZhongYun LIU WenBo BAO BoCheng | 2021 | Science China(Technological Sciences)2021，64，10： | 2 |
| 3 | 稀疏重构SAM芯片焊点检测方法研究显示文摘提出了基于稀疏表示的声扫描显微镜(Scanning acoustic microscope,SAM)图像超分辨率重构方法,以解决其空间检测分辨率受超声波频率和穿透深度的限制,原始SAM图像分辨率较低,不利于封装缺陷辨识等问题。通过字典设计训练和稀疏系数α求解获得了重构的高分辨率SAM图像,利用Levenberg-Marquardt算法改进BP神经网络(LM-BP),并用于倒装芯片焊点缺陷识别。与原始图像及双三次插值图像相比,稀疏重构图像的峰值信噪比明显增大,提高了SAM图像质量,减小了芯片焊点的错误识别数目,错误率降至2.76%。试验结果表明稀疏表示的SAM重构算法和LM-BP神经网络训练速度快、识别精度高,可用于高密度半导体封装缺陷的检测及可靠性评估。 | 陆向宁 刘凡 何贞志 廖广兰 史铁林 | 2023 | 机械工程学报2023，59，6： | 0 |
| 4 | 基于极限学习机的车间节能目标预测方法显示文摘针对车间的混合流水线调度问题(HFSP)存在智能算法寻优过程中节能目标即适应值评估代价高的问题,首先,通过分析车间节能模型建模的编码方式,提出一种基于矩阵编码机制的特征向量提取方法,引入核函数有利于极限学习机(ELM)求解节能目标。其次,对需要构建代理模型的改进多目标多元宇宙优化算法(IMOMVO)进行计算复杂度分析,建立了基于ELM的代理模型,设计数据驱动优化的车间节能目标算法框架。最后,基于均匀分布变量的拉丁超立方抽样,形成初始化样本,与BP算法进行预测性能验证和计算时间对比两个实验。实验结果显示,ELM算法的拟合优度为0.973 81,预测性能指标均优于BP算法。单个适应值平均计算时间为5.4×10^-4s,仅为真实求解的18.5%。说明ELM在车间节能目标预测问题具有良好的效果。 | 刘大铖 李少波 魏宏静 | 2020 | 贵州大学学报(自然科学版)2020，37，4： | 0 |
| 5 | 基于机器学习电子封装互连材料研究现状显示文摘随着“后摩尔时代”到来,先进封装技术使芯片性能不断提升,传统的材料开发制备以及封装互连研究难以适应时代需要。根据工业4.0的愿景,采用机器学习的算法在封装材料开发预测、封装工艺优化、焊点可靠性等方面进行高效预测工作,是未来发展的趋势之一。材料领域应用机器学习的过程,包括收集数据、数据预处理、调参优化三方面。封装领域中大多数预测问题可以归类为分类问题,传统监督学习应用最为广泛,常用算法有支持向量机、人工神经网络等。深度学习在封装问题的预测中大放异彩,如卷积神经网络模型等深度学习算法已经逐渐应该用于预测中,并且获得极佳的预测效果。尽管机器学习辅助电子封装研究尚处于萌芽状态,但仍然能够看到机器学习在电子封装领域具有巨大研究及实用价值。 | 张墅野 段晓康 罗克宇 许孙武 张志昊 陈捷狮 何鹏 | 2023 | 机械工程学报2023，59，22： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员