Use of a neural network to link AFM etch patterns to plasma parameters

Plasmas play a critical role in depositing thin films or etching fine patterns while manufacturing integrated circuits. A new model for plasma diagnosis is presented. This was accomplished by linking atomic force microscopy (AFM) to plasma parameters using a neural network. Experimental AFM data were collected during the etching of silicon oxynitride films in C₂F₆ inductively coupled plasma. Surface roughness of etched patterns was characterized by means of discrete wavelet transformation. This led to the construction of three vertical (type I), diagonal (type II), and horizontal (type III) wavelet coefficient-based models. The performance of diagnosis models was evaluated in terms of the prediction and recognition accuracies. Both accuracies were optimized as a function of the number of hidden neurons. Comparisons revealed that the type I model yielded the largest recognition and the smallest prediction error. This was demonstrated even under stricter monitoring conditions. More improved diagnosis is expected by enhancing AFM resolution.     

模糊数学在电视机故障诊断中的应用

本文介绍了应用模糊数学方法进行计算机辅助诊断电视机的故障原因,这种方法有效性在于建立模型的因果关系正确性和自动搜索范围及速度。     

Current basis for the ultrastructural clinical diagnosis of tumors: A review

The ultrastructural diagnosis of tumors requires a careful analysis that should be done in an orderly fashion. It requires precise planning from the time of specimen collection to the selection of the area to be examined. Pictures must be taken systematically and every micrograph should allow to answer whether the number of cells photographed is adequate; whether mitoses are present, what is the pattern of the tumor; what is the appearance of the cell membrane; whether the cells are joined by junctional complexes; whether free surfaces possess microvilli or cilia; what organelles are present and how they are distributed; whether there are secretory granules, melanosomes, or other cytoplasmic elements. Nuclear and nucleolar size and shape have to be taken into consideration. The composition of the interstitial extracellular matrix is important in certain types of tumors. Although these questions are not the only ones to be addressed, their use in a logical fashion is helpful when it concerns the ultrastructural diagnosis.     

输血后丙型肝炎病毒NS4区抗体的动态变化及其诊断意义

利用重组NS4抗原及合成肽5－11抗原分别检测5例输血后丙型肝炎患者的系列血清136份，并与HCV总抗体及HCVRNA检测结果比较，发现抗NS4抗体的动态变化虽然有两种模式，但在多数情况下为一种模式，即抗体阳转后很少转阴，并且其动态变化基本与总抗体相似。同时发现2例病人抗NS4抗体阳转时间早于总抗体，因此，推测HCV检测试剂中增加NS4抗原可能对提高HCV诊断试剂的灵敏度有一定意义。     

一种汽车用液压电磁阀控制电路设计

液压电磁阀是现代汽车液压控制系统的重要部件,通过对电磁阀的精确驱动控制,全面的故障诊断,确保液压系统的安全可靠运行。本电路设计具有灵活性、安全性、可扩展性等特点,可以方便实现对不同类型、不同电气参数、不同数量的液压电磁阀的控制,对推动机电液一体化技术在汽车上的应用具有积极意义。     

2D12型往复式压缩机气阀故障诊断的时频分析方法

分析了 2D12型往复式压缩机气阀的工作过程 ,针对气阀工作特点提取吸气阀与排气阀均关闭时刻的压缩阶段的局域声音信号 ,应用广义双线性时频法诊断气阀故障 ,给出了正常气阀和故障气阀的时频分析图     

深度学习在主动脉中膜变性病理图像分类中的应用

胸主动脉瘤和夹层（TAAD）是严重的心血管疾病之一,而中膜变性（MD）的组织学改变对疾病的诊断及早期干预具有重要的临床意义。针对病理图像的高度复杂性使得MD的诊断过程耗时费力且一致性差的问题,提出了一种基于深度学习的病理图像分类方法,并将其应用于四种MD病变类型以进行性能验证。该方法使用了一种改进的基于GoogLeNet的卷积神经网络模型,首先采用迁移学习来将先验知识应用于TAAD病理图像的表达,然后使用Focal loss和L2正则化来解决数据不平衡问题,从而进一步优化模型性能。实验结果表明,所提模型的平均四分类准确率达到98.78%,表现出较好的泛化性能。可见所提方法可以有效地提升病理学家的诊断效率。     

汽车电控发动机故障诊断专家系统在高职教学中的应用

随着电子控制技术在汽车上的广泛应用,电控发动机的检修日益成为汽车维修的重点和难点,同时,发动机电控系统检修也是高职高专汽车类专业重要的专业课之一。对模糊理论和BP神经网络在故障诊断方面的应用进行了探讨,以BP神经网络为主,采用模糊理论作为有益补充设计了电控发动机故障诊断专家系统,给学生创造接近实际的虚拟学习环境,增强学生的学习积极性,取得了良好的教学效果。     

微机联锁远程故障诊断系统的研究

作者论述了微机联锁远程故障诊断系统的构成和实现,提出现阶段信号维修改革的任务、组织管理及实施远程故障诊断系统的关键技术。     

基于贝叶斯网络的复杂装备D-S诊断

利用D-S证据理论进行故障诊断时,首先需要确定出故障识别框架和基本置信分配.但在导弹武器等复杂装备中,存在着故障原因与故障征兆之间关联关系不确定的现象,增加了求取故障识别框架和基本置信分配的困难.文中提出了利用贝叶斯网络理论来解决难题的方法,使D-S证据理论能在复杂装备故障诊断中得到应用.文中首先给出了诊断贝叶斯网络的适应度函数,并利用改进的粒子群优化算法求解了诊断贝叶斯网络.然后分析了诊断贝叶斯网络中故障原因与故障征兆间的关联关系,给出了征兆集合、异因征兆集合概念.提出了求解故障识别框架和基本置信分配的方法.最后,通过实例的D-S诊断,验证了所提方法的正确性.