基于增强型对抗学习的无参考图像质量评价算法

为了提高无参考图像质量评价(NR-IQA)方法的性能,参考先进的深度生成对抗网络(GAN)研究成果,提出一种基于增强型对抗学习的无参考图像质量评价算法,即通过改进损失函数、网络模型结构来增强对抗学习强度,输出更可靠的模拟"参考图",进而可以像全参考图像质量评价(FR-IQA)方法一样模拟人的视觉比较过程。首先,利用数据集中失真的图像和未失真的原图像作为输入,从而基于增强对抗学习来训练网络模型;然后,利用该模型输出待测图像的模拟仿真图,提取仿真图的深度卷积特征;最后,将仿真图和待测失真图的卷积特征相融合,并输入到训练好的图像质量评价回归网络,输出图像的评测分数。在LIVE、TID2008和TID2013数据集上完成实验。实验结果表明,所提算法在图像质量上的总体客观评价性能优于当前的主流算法,与人的主观评价表现出的性能相一致。     

Nonlinear fault detection based on locally linear embedding

In this paper, a new nonlinear fault detection technique based on locally linear embedding （LLE） is developed. LLE can efficiently compute the low-dimensional embedding of the data with the local neighborhood structure information preserved. In this method, a data-dependent kernel matrix which can reflect the nonlinear data structure is defined. Based on the kernel matrix, the Nystrrm formula makes the mapping extended to the testing data possible. With the kernel view of the LLE, two monitoring statistics are constructed. Together with the out of sample extensions, LLE is used for nonlinear fault detection. Simulation cases were studied to demonstrate the performance of the proposed method.     

一类基于神经网络非线性观测器的鲁棒故障检测和诊断

利用神经网络的非线性建模能力，对一类具有建模不确定项的非线性系统提出一种基于观测器的故障检测和诊断的方法。设计的观测器不仅能实现故障检测，而旦应用神经网络设计的故障估计器能在线估计系统中的故障向量。通过分析验证了该方法对系统中的建模误差和外部扰动具有良好的鲁棒性。仿真结果表明所提出的方法是有效的。     

Quality relevant nonlinear batch process performance monitoring using a kernel based multiway non-Gaussian latent subspace projection approach

Multiway kernel partial least squares method (MKPLS) has recently been developed for monitoring the operational performance of nonlinear batch or semi-batch processes. It has strong capability to handle batch trajectories and nonlinear process dynamics, which cannot be effectively dealt with by traditional multiway partial least squares (MPLS) technique. However, MKPLS method may not be effective in capturing significant non-Gaussian features of batch processes because only the second-order statistics instead of higher-order statistics are taken into account in the underlying model. On the other hand, multiway kernel independent component analysis (MKICA) has been proposed for nonlinear batch process monitoring and fault detection. Different from MKPLS, MKICA can extract not only nonlinear but also non-Gaussian features through maximizing the higher-order statistic of negentropy instead of second-order statistic of covariance within the high-dimensional kernel space. Nevertheless, MKICA based process monitoring approaches may not be well suited in many batch processes because only process measurement variables are utilized while quality variables are not considered in the multivariate models. In this paper, a novel multiway kernel based quality relevant non-Gaussian latent subspace projection (MKQNGLSP) approach is proposed in order to monitor the operational performance of batch processes with nonlinear and non-Gaussian dynamics by combining measurement and quality variables. First, both process measurement and quality variables are projected onto high-dimensional nonlinear kernel feature spaces, respectively. Then, the multidimensional latent directions within kernel feature subspaces corresponding to measurement and quality variables are concurrently searched for so that the maximized mutual information between the measurement and quality spaces is obtained. The I² and SPE monitoring indices within the extracted latent subspaces are further defined to capture batch process faults resulting in abnormal product quality. The proposed MKQNGLSP method is applied to a fed-batch penicillin fermentation process and the operational performance monitoring results demonstrate the superiority of the developed method as apposed to the MKPLS based process monitoring approach.     

基于对抗学习和全局知识信息的关系检测技术研究

针对现有的知识库关系检测任务对于一些不可见关系无法做到准确的向量表示而出现词汇溢出的问题,提出了基于对抗学习和全局知识信息的关系检测模型。该模型使用对抗学习对知识库关系表示模型进行特征强化,使用TransH(translating on hyperplanes)模型提取全局知识信息,同时通过联合训练,将全局知识信息融合进关系表示模型中,进一步提升关系模型的表示能力。实验结果表明,提出的融合模型对于关系检测效果有一定的提升,并且缓解了词汇溢出的问题。     

一种新的基于迭代学习的故障检测和估计算法

将迭代学习策略应用到故障诊断中,提出一种新的故障检测和估计算法.引入虚拟故障参数,采用基于PID的迭代学习策略调节其中的虚拟故障,使虚拟故障逼近系统中实际发生的故障,从而达到对系统故障诊断的目的.理论证明和仿真结果验证了新算法的可行性和有效性.     

基于动态高效潜结构投影的质量相关故障检测

高效潜结构投影(EPLS)算法是一种反映过程变量与质量变量相关关系的多变量统计分析方法,在质量相关故障检测中具有良好的检测效果.然而EPLS算法是一种静态检测模型,不能反映实际工业过程或装备测试中的动态特性,对动态过程中质量相关故障的检测率较低.为此,本文提出了一种基于自回归移动平均模型(ARMAX)的动态高效潜结构投...     

基于随机投影与集成学习的离群点检测算法

针对传统基于相似度的离群点检测算法在高维不均衡数据集上效果不够理想的问题,提出一种新颖的基于随机投影与集成学习的离群点检测（ensemble learning and random projection-based outlier detection,EROD）框架。算法首先集成多个随机投影方法对高维数据进行降维,提升数据多样性;然后集成多个不同的传统离群点检测器构建异质集成模型,增加算法鲁棒性;最后使用异质模型对降维后的数据进行训练,训练后的模型经过两次优化组合以降低泛化误差,输出最终的对象离群值,离群值高的对象被算法判定为离群点。分别在四个不同领域的高维不均衡真实数据集上进行对比实验,结果表明该算法与传统离群点检测算法和基于集成学习的离群点检测算法相比,在AUC和precision@n值上平均提高了3.6%和14.45%,证明EROD算法具有处理高维不均衡数据异常的优势。     

基于双向循环生成对抗网络的无线传感网入侵检测方法

针对无线传感器网络（WSN）入侵检测方法在离散高维特征的不平衡数据集上检测精度低和泛化能力差的问题,提出一种基于双向循环生成对抗网络的WSN入侵检测方法 BiCirGAN。首先,引入对抗学习异常检测（ALAD）通过潜在空间合理地表示高维、离散的原始特征,提高对原始特征的可理解性。其次,采用双向循环对抗的结构确保真实空间和潜在空间双向循环的一致性,从而保证生成对抗网络（GAN）训练的稳定性,并提高异常检测的性能。同时,引入Wasserstein距离和谱归一化优化方法改进GAN的目标函数,以进一步解决GAN的模式崩坏与生成器缺乏多样性的问题。最后,由于入侵攻击数据的统计属性随时间以不可预见的方式变化,建立带有Dropout操作的全连接层网络对异常检测结果进行优化。实验结果表明,在KDD99、UNSW-NB15和WSN＿DS数据集上,相较于AnoGAN、BiGAN、MAD-GAN以及ALAD方法,BiCirGAN在检测精确度上提高了3.9%～33.0%,且平均推断速度是ALAD方法的4.67倍。     

基于深度学习的变压器在线故障检测

针对供配电网络中变压器故障检测和预报的不足,提出基于深度学习的变压器故障检测方法,详细介绍了变压器监测数据预处理方法及步骤,给出了深度学习网络的具体结构和学习过程,深度学习结果表明该故障检测方法的有效性和实用性。     

Online fault detection and isolation of nonlinear systems based on neurofuzzy networks

An online fault detection and isolation (FDI) technique for nonlinear systems based on neurofuzzy networks (NFN) is proposed in this paper. Two NFNs are used. The first one trained by data obtained under normal operating condition models the system and the second one trained online models the residuals. Fuzzy rules that are activated under fault free and faulty conditions are extracted from the second NFN and stored in the symptom vectors using a binary code. A fault database is then formed from these symptom vectors. When applying the proposed FDI technique, the NFN that models the residuals is updated recursively online, from which the symptom vector is obtained. By comparing this symptom vector with those in the fault database, faults are isolated. Further, the fuzzy rules obtained from the symptom vector can also provide linguistic information to experienced operators for identifying the faults. The implementation and performance of the proposed FDI technique is illustrated by simulation examples involving a two-tank water level control system under faulty conditions.     

High-gain observation with disturbance attenuation and application to robust fault detection

One approach to the problem of residual generation in a purpose of fault detection is to use an observer. One particular difficulty is to distinguish between faults and disturbances. Various observers have already been inspected in that direction, generally based on exact decoupling w.r.t. unknown disturbances. Here the use of high-gain observer techniques is inspected, with a purpose of attenuation of disturbances rather than exact decoupling: conditions allowing some “robust partial estimation” are first presented, and their possible use in fault detection is then discussed.     

Research on fault diagnosis method of MS-CNN rolling bearing based on local central moment discrepancy

Transfer learning is an excellent approach to deal with the problem that the target domain label can not be adequately obtained when rolling bearing cross-condition fault detection. A transfer learning fault diagnosis method of multi-scale CNN rolling bearings based on local central moment discrepancy is presented in this research. The method maps bearing vibration data to a shared space by building a shared multi-scale feature extraction structure and fully connected layers. The source domain label and target domain pseudo-label are used to divide the category subspace in the shared space. And then the local central moment discrepancy is used to match source and target domain in the category subspace to realize fault knowledge transfer under different conditions. The experimental findings reveal that multi-scale CNN migration diagnosis based on local central moment discrepancy has superior accuracy and stability in diverse diagnostic tasks when compared to classic transfer learning approaches.     

Fault detection based on one-class deep learning for manufacturing applications limited to an imbalanced database

Despite extensive studies for the industrial applications of deep learning, its actual usage in manufacturing sites has been extremely restrained by the difficulty in obtaining sufficient industrial data, especially for production failure cases. In this study, we introduced a fault-detection module based on one-class deep learning for imbalanced industrial time-series data, which consists of three submodules, namely, time-series prediction based on deep learning, residual calculation, and one-class classification using one-class support vector machine and isolation forest. Four different networks were used for the time-series prediction: multilayer perception (MLP), residual network (ResNet), long–short-term memory (LSTM), and ResNet–LSTM, each trained with the one-class data having only the production success cases. We adopted the residuals of the deep-learning prediction as an elaborated feature for the construction of the one-class classification. We also tested the fault-detection module with the actual mass production data of a die-casting process. By adopting the features elaborated by the deep-learning time-series prediction, we showed that the total accuracy of the one-class classification significantly improved from 90.0% to 96.0%. Especially for its capability to detect production failures, the accuracy improved from 84.0% to 96.0%. The area under the receiver operating characteristics (AUROC) also improved from 87.56% to 98.96%. ResNet showed the best performance for detecting production failures, whereas ResNet–LSTM produced better results for ensuring the production success. Our results suggest that the one-class deep learning is a promising approach for extracting important features from time-series data to realize a one-class fault-detection module.     

Relaxed fault detection and isolation: An application to a nonlinear case study

Given a number of possibly concurrent faults (and disturbances) that may affect a nonlinear dynamic system, it may not be possible to solve the standard fault detection and isolation (FDI) problem, i.e., to detect and isolate each single fault from all other, possibly concurrent faults and disturbances, due to the violation of the available necessary conditions of geometric nature. Motivated by a robotic application where this negative situation structurally occurs, we propose some relaxed formulations of the FDI problem and show how necessary and sufficient conditions for their solution can be derived from those available for standard FDI. The design of a hybrid residual generator follows directly from the fulfillment of the corresponding solvability conditions. In the considered nonlinear case study, a robotic system affected by possible actuator and/or force sensor faults, we detail the application of these results and present experimental tests for validation.     

基于多阶段生成对抗网络的单幅图像阴影去除方法

传统的深度学习阴影去除方法常常会改变非阴影区域的像素且无法得到边界过渡自然的阴影去除结果。为了解决该问题,基于生成对抗网络（GAN）提出一种新颖的多阶段阴影去除框架。首先,多任务驱动的生成器分别通过阴影检测子网和蒙版生成子网为输入图像生成相应的阴影掩膜和阴影蒙版;其次,在阴影掩膜和阴影蒙版的引导下,分别设计全影模块和半影模块,分阶段去除图像中不同类型的阴影;然后,以最小二乘损失为主导构建一种新的组合损失函数以得到更好的结果。与最新的深度学习阴影去除方法相比,在筛选数据集上,所提方法的平衡误差率（BER）减小约4.39%,结构相似性（SSIM）提高约0.44%,像素均方根误差（RMSE）减小约13.32%。实验结果表明该方法得到的阴影去除结果边界过渡更加平滑。     

基于深度对齐网络的生成对抗网络伪造人脸检测

针对现有的生成对抗网络（GAN）伪造人脸图像检测方法在有角度及遮挡情况下存在的真实人脸误判问题,提出了一种基于深度对齐网络（DAN）的GAN伪造人脸图像检测方法。首先,基于DAN设计面部关键点提取网络,以提取真伪人脸关键点位置;然后,采用主成分分析（PCA）方法将每一组关键点映射到三维空间,从而减少冗余信息以及降低特征维度;最后,利用支持向量机（SVM）五折交叉验证对特征进行分类,并计算准确率。实验结果表明,该方法通过提高面部关键点定位准确度改善了由于定位误差引起的面部不协调问题,进而降低了真实人脸误判率。与VGG19、XceptionNet和Dlib-SVM方法相比,正脸情况下,该方法的ROC下面积（AUC）值提高了4.48到32.96个百分点,平均精度（AP）提高了4.26到33.12个百分点;有角度及遮挡人脸情况下,该方法的AUC值提高了10.56到30.75个百分点,AP提高了7.42到42.45个百分点。     

基于度量学习的行人检测算法

针对拥挤场景下行人漏检率较高的问题,设计了新的类平衡策略.其次,采用度量学习方法改进目前的行人语义提取效果,并设计了新的距离度量方法.最后,结合提取的行人语义信息设计了新的非极大值抑制算法.在行人检测数据集CityPersons和CrowdHuman上,与目前的行人检测器进行对比,效果优于目前最优无锚框的行人检测器,同时也证明了度量学习方法在行人检测中的有效性.     

基于机器学习方法的直流电弧故障检测

为了解决传统分析方法在直流供电系统中电弧故障检测的精确度不足及过程繁琐的问题,将直流电弧故障检测归为二分类问题,引入机器学习方法,通过直流电弧实验得到正常状态和电弧状态的数据,从时域中提取电流均值等4个特征,从频域中提取高频分量标准差等3个特征.利用提取到的特征对支持向量机(SVM)进行训练,利用求解得到的模型对测试数据集进行分类,分类准确率为94.483%.结果证明:所提方法能有效检测直流电弧故障,提高故障检测精度,且步骤精简,易于推广.