Deep-SBFL:基于频谱的深度神经网络缺陷定位方法

深度神经网络已经在自动驾驶和智能医疗等领域取得了广泛的应用.与传统软件一样,深度神经网络也不可避免地包含缺陷,如果做出错误决定,可能会造成严重后果.因此,深度神经网络的质量保障受到了广泛关注.然而,深度神经网络与传统软件存在较大差异,传统软件质量保障方法无法直接应用于深度神经网络,需要设计有针对性的质量保障方法.软件缺陷定位是保障软件质量的重要方法之一,基于频谱的缺陷定位方法在传统软件的缺陷定位中取得了很好的效果,但无法直接应用于深度神经网络.在传统软件缺陷定位方法的基础上提出了一种基于频谱的深度神经网络缺陷定位方法 Deep-SBFL.该方法首先通过收集深度神经网络的神经元输出信息和预测结果作为频谱信息;然后将频谱信息进行处理作为贡献信息,以用于量化神经元对预测结果所做的贡献;最后提出了针对深度神经网络缺陷定位的怀疑度公式,基于贡献信息计算深度神经网络中神经元的怀疑度并进行排序,以找出最有可能存在缺陷的神经元.为验证该方法的有效性,以E_Inspect@n (结果排序列表前n个位置内成功定位的缺陷数)和EXAM (在找到缺陷元素之前必须检查元素的百分比)作为评测指...     

基于粗糙集的多层容错神经网络故障诊断

针对神经网络故障诊断问题中输入属性维数多和数据量庞大的情况,首先利用粗糙集理论对原始数据进行约简,并按照一定的原则选取多个约简;然后对所得到的多个约简分别构建子神经网络,将多个子网络合成统一的容错网络。结合实例应用取得了令人满意的结果,并为高可靠性设备的故障诊断提供了新的思路。     

基于粗糙集-神经网络集成的故障诊断

综合粗糙集和神经网络的优点,提出一种基于粗糙集-神经网络集成的智能故障诊断模型.在数据采集和预处理的基础上,利用粗糙集(RS)理论对原始故障诊断样本进行离散化处理,并根据条件属性(集)对决策属性的正域的大小来选择属性,提取出对诊断故障贡献最大的最小故障特征子集,从而确定神经网络的拓扑结构;通过网络训练建立故障特征与故障之间的映射关系,采用神经网络集成的方法实现故障的诊断.通过热电厂发电机组的故障诊断实例,表明了这种故障诊断方法的工程有效性.     

Tracing the geometry of an active fault using remote sensing and digital elevation model: Ganos segment,North Anatolian Fault zone,Turkey

The surface expression of the North Anatolian Fault between the Sea of Marmara and the Gulf of Saros, Turkey was determined using Landsat 5 Thematic Mapper (TM) and Système Probatoire de l'Observation de la Terre (SPOT) high resolution visible (HRV) Panchromatic images and a digital elevation model (DEM) based on 1?:?25?000 scale topographic maps. The remotely sensed data had been enhanced to reveal linear details. Between Gaziköy (Marmara Shore) and Saros two small releasing steps were identified, one to the east of Mürseli village and the other between Mürseli and Yayaköy villages. Hitherto, a single straight fault line had been drawn between Gaziköy and Saros on the geological map of this region. Between the Yeniköy and Sofular villages, the fault strike bends ～7° further to the SE. The course of the fault between the Sea of Marmara and the Gulf of Saros is, therefore, more complicated than formerly believed and this is probably because the fault has utilized the pre-existing structures of the Neo-Tethyan subduction–accretion complex.     

基于RSPNN的制粉系统故障诊断

针对发电厂制粉系统故障与征兆对应关系复杂及过程信息的不确定性及传统BP神经网络故障诊断的缺点,提出了基于粗糙集概率神经网络(RSPNN)的制粉系统故障诊断方法,以改善传统BP神经网络初始值敏感、易使学习过程陷入局部极小值以及样本数据过大时训练速度慢等问题。首先采用自组织映射神经网络(SOMNN)对连续样本数据进行离散化;再利用基于区分矩阵的HORAFA算法对离散化样本数据进行RS属性约简,并将约简结果作为概率神经网络(PNN)的输入;最后利用PNN作为诊断决策分类器,输出故障模式,并进行了仿真研究。仿真结果表明,该方法不仅优化神经网络的拓扑结构,降低神经网络的训练时间,而且能准确、快速地诊断制粉系统故障类型,同时对发电厂制粉系统及其相关设备的在线故障诊断问题有一定启发性。     

一种基于RBF神经网络的转台分系统故障诊断方法

针对三轴精密测试转台各分系统故障诊断的需要,提出了一种基于径向基函数(RBF)神经网络的局部故障诊断方法。首先,给出了相应的RBF神经网络的结构,以及一种基于递归最小二乘法的改进学习算法;然后,将其应用到转台控制分系统的局部故障诊断中。根据控制分系统的常见故障及其特征信息,建立起基于RBF神经网络的故障诊断模型;最后,仿真实验结果验证了该方法的有效性。     

An artificial neural network (p, d, q) model for timeseries forecasting

Artificial neural networks (ANNs) are flexible computing frameworks and universal approximators that can be applied to a wide range of time series forecasting problems with a high degree of accuracy. However, despite all advantages cited for artificial neural networks, their performance for some real time series is not satisfactory. Improving forecasting especially time series forecasting accuracy is an important yet often difficult task facing forecasters. Both theoretical and empirical findings have indicated that integration of different models can be an effective way of improving upon their predictive performance, especially when the models in the ensemble are quite different. In this paper, a novel hybrid model of artificial neural networks is proposed using auto-regressive integrated moving average (ARIMA) models in order to yield a more accurate forecasting model than artificial neural networks. The empirical results with three well-known real data sets indicate that the proposed model can be an effective way to improve forecasting accuracy achieved by artificial neural networks. Therefore, it can be used as an appropriate alternative model for forecasting task, especially when higher forecasting accuracy is needed.     

An unsupervised artificial neural network versus a rule-based approach for fault detection and identification in an automated assembly machine

Artificial neural networks (ANNs) are suitable for fault detection and identification (FDI) applications because of their pattern recognition abilities. In this study, an unsupervised ANN based on Adaptive Resonance Theory (ART) is tested for FDI on an automated O-ring assembly machine testbed, and its performance and practicality are compared to a conventional rule-based method. Three greyscale sensors and two redundant limit switches are used as cost-effective sensors to monitor the machine’s assembly process. Sensor data are collected while the machine is operated under normal condition, as well as 10 fault conditions. Features are selected from the raw sensor data, and data sets are created for training and testing the ANN. The performance of the ANN for detecting and identifying known, unknown and multiple faults is evaluated; the performance is compared to a conventional rule-based method using the same data sets. Results show that the ART ANN is able to achieve excellent fault detection performance with minimal modeling requirements; however, the performance depends on careful tuning of its vigilance parameter. Although the rule-based system requires more effort to set up, it is judged to be more useful when unknown or multiple faults are present. The ART network creates new outputs for unknown and multiple fault conditions, but it does not give any more information as to what the new fault is. By contrast, the rule-based method is able to generate symptoms that clearly identify the unknown and multiple fault conditions. Thus, the rule-based method is judged to be the most feasible method for FDI applications.     

Differential evolution trained wavelet neural networks: Application to bankruptcy prediction in banks

In this study, differential evolution algorithm (DE) is proposed to train a wavelet neural network (WNN). The resulting network is named as differential evolution trained wavelet neural network (DEWNN). The efficacy of DEWNN is tested on bankruptcy prediction datasets viz. US banks, Turkish banks and Spanish banks. Further, its efficacy is also tested on benchmark datasets such as Iris, Wine and Wisconsin Breast Cancer. Moreover, Garson’s algorithm for feature selection in multi layer perceptron is adapted in the case of DEWNN. The performance of DEWNN is compared with that of threshold accepting trained wavelet neural network (TAWNN) [Vinay Kumar, K., Ravi, V., Mahil Carr, & Raj Kiran, N. (2008). Software cost estimation using wavelet neural networks. Journal of Systems and Software] and the original wavelet neural network (WNN) in the case of all data sets without feature selection and also in the case of four data sets where feature selection was performed. The whole experimentation is conducted using 10-fold cross validation method. Results show that soft computing hybrids viz., DEWNN and TAWNN outperformed the original WNN in terms of accuracy and sensitivity across all problems. Furthermore, DEWNN outscored TAWNN in terms of accuracy and sensitivity across all problems except Turkish banks dataset.     

Forecasting of Software Reliability Using Neighborhood Fuzzy Particle Swarm Optimization Based Novel Neural Network

This paper proposes an artificial neural network (ANN) based software reliability model trained by novel particle swarm optimization (PSO) algorithm for enhanced forecasting of the reliability of software. The proposed ANN is developed considering the fault generation phenomenon during software testing with the fault complexity of different levels. We demonstrate the proposed model considering three types of faults residing in the software. We propose a neighborhood based fuzzy PSO algorithm for competent learning of the proposed ANN using software failure data. Fitting and prediction performances of the neighborhood fuzzy PSO based proposed neural network model are compared with the standard PSO based proposed neural network model and existing ANN based software reliability models in the literature through three real software failure data sets. We also compare the performance of the proposed PSO algorithm with the standard PSO algorithm through learning of the proposed ANN. Statistical analysis shows that the neighborhood fuzzy PSO based proposed neural network model has comparatively better fitting and predictive ability than the standard PSO based proposed neural network model and other ANN based software reliability models. Faster release of software is achievable by applying the proposed PSO based neural network model during the testing period.      

Coseismic deformation and source model of the 12 November 2017 MW 7.3 Kermanshah Earthquake (Iran–Iraq border) investigated through DInSAR measurements

A large earthquake with a magnitude of M_W 7.3 struck the border of Iran and Iraq at the province of Kermanshah, Iran. In our study, coseismic deformation and source model of the 12 November 2017 Kermanshah Earthquake are investigated using ALOS-2 ScanSAR and Sentinel-1A/B TOPSAR Differential Interferometric Synthetic Aperture Radar (DInSAR) techniques. Geodetic inversion has been performed to constrain source parameters and invert slip distribution on the fault plane. The optimised source model from joint inversion shows a blind reverse fault with a relatively large right-lateral component, striking 353.5° NNW-SSE and dipping 16.3° NE. The maximum slip is up to 3.8 m at 12–14 km depth and the inferred seismic moment is 1.01 × 10²⁰ Nm, corresponding to M_W 7.3, consistent with seismological solutions. The high-resolution optical images from SuperView-1 satellite suggest that most of the linear surface features mapped by DInSAR measurements are landslides or surface cracks triggered by the earthquake. Coulomb stress changes on the source fault indicating consistency between aftershock distribution and high loaded stress zones. Based on the stress change on neighbouring active faults around this area, the Kermanshah Earthquake has brought two segments of the Zagros Mountain Front Fault (MFF), MFF-1 and MFF-2, 0.5–3.1 MPa and 0.5–1.96 MPa closer to failure, respectively, suggesting the risk of future earthquakes. Recent major aftershocks (M_W≥ 5.0) could probably ease the seismic hazard on MFF-2, but the risk of earthquakes on MFF-2 is still increasing.     

A novel hybrid classification model of artificial neural networks and multiple linear regression models

The classification problem of assigning several observations into different disjoint groups plays an important role in business decision making and many other areas. Developing more accurate and widely applicable classification models has significant implications in these areas. It is the reason that despite of the numerous classification models available, the research for improving the effectiveness of these models has never stopped. Combining several models or using hybrid models has become a common practice in order to overcome the deficiencies of single models and can be an effective way of improving upon their predictive performance, especially when the models in combination are quite different. In this paper, a novel hybridization of artificial neural networks (ANNs) is proposed using multiple linear regression models in order to yield more general and more accurate model than traditional artificial neural networks for solving classification problems. Empirical results indicate that the proposed hybrid model exhibits effectively improved classification accuracy in comparison with traditional artificial neural networks and also some other classification models such as linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), K-nearest neighbor (KNN), and support vector machines (SVMs) using benchmark and real-world application data sets. These data sets vary in the number of classes (two versus multiple) and the source of the data (synthetic versus real-world). Therefore, it can be applied as an appropriate alternate approach for solving classification problems, specifically when higher forecasting accuracy is needed.     

贝叶斯正则化神经网络预测金属晶体结合能的研究

采用贝叶斯正则化神经网络(BRNN)对61种金属晶体结合能进行了预测。对网络结构、训练集、预测集以及学习次数进行了优化，并用独立预测样本对贝叶斯正则化神经网络作了检验。预测结果表明，在推广能力方面，贝叶斯正则化神经网络优于熟知的反向传播(BP)神经网络和多元线性回归方法(MLR)。它可望成为元素和化合物构效关系研究的辅助手段。     

基于BP神经网络的天然与非天然地震识别

基于BP神经网络算法,选择山东地区天然与非天然地震事件为研究对象,选取P波初动方向、P波初动振幅与S波最大振幅比以及P波最大振幅与S波最大振幅比作为神经网络输入元,构建合理的网络模型,搜集大量的地震样本数据进行训练,实现对山东地区天然与非天然地震事件的识别。震例预测结果显示,P波初动方向和振幅比可以作为识别天然地震和非天然地震主要依据,BP神经网络方法具有对地震事件类型识别的可行性。     

基于梯度结构的图神经网络对抗攻击

图神经网络在半监督节点分类任务中取得了显著的性能. 研究表明, 图神经网络容易受到干扰, 因此目前已有研究涉及图神经网络的对抗鲁棒性. 然而, 基于梯度的攻击不能保证最优的扰动. 提出了一种基于梯度和结构的对抗性攻击方法, 增强了基于梯度的扰动. 该方法首先利用训练损失的一阶优化生成候选扰动集, 然后对候选集进行相似性评估, 根据评估结果排序并选择固定预算的修改以实现攻击. 通过在5个数据集上进行半监督节点分类任务来评估所提出的攻击方法. 实验结果表明, 在仅执行少量扰动的情况下, 节点分类精度显著下降, 明显优于现有攻击方法.     

基于集成多神经网络在模拟电路故障诊断中的研究

以进行模拟电路故障诊断为主要目的,针对单神经网络故障字典法在进行复杂电路系统故障诊断时,对多故障和多任务诊断的不足之处,讨论了基于多故障的神经网络集成技术,采用集成多神经网络来提高诊断速度和精度,提出了集成多神经网络故障字典法来解决多故障任务,对基于层次分类模型的多重结构神经网络进行了研究,给出了两种对故障定位的统一融合算法,克服了采用单神经网络多故障时学习速度慢,出现新故障的网络要重新进行学习等缺点.并给出了应用实例.     

Fault Diagnosis Using Wavelet Neural Networks

Wavelet neural networks are a class of neural networks consisting of wavelets. This paper presents a novel universal tool for fault diagnosis and algorithms for wavelet neural network construction are proposed. Using the model of wavelet neural networks, we can not only extract the features of system but also predict the development of the fault.     

基于RNN-LSTM的磨矿系统故障诊断技术

目前磨矿系统故障诊断多为人为判断，效率低、准确率低、成本高且容易造成人员伤亡.传统方法对高维度和时间相关性较大的样本数据集分类能力较差，针对以上问题，提出一种基于RNN-LSTM（Recurrent Neural Network-Long Short-Term Memory）的深度学习方法，实现磨矿系统故障的智能化诊断.该方法通过将数据集"分批处理"分别输入到LSTM单元网络中，提取数据集在时间维度上的相关性，并比较分析前后时刻的输入特征向量实现对故障分类.通过分别对RNN-LSTM深度学习网络与基于自编码分类方法进行实验对比验证，得出结论：在时间相关性较强的高维度数据集中基于RNN-LSTM深度方法辨识效果明显优于基于自编码方法的分类器，最终网络对于故障诊断的错误率低至3%.     

增强前馈神经网络对多点断路故障的容错能力

神经网络在发生多点断路故障时,网络中的多个隐层神经元及其相关的连接权同时失效。该文针对陷层神经元可以动态增加的一类前馈神经网络,提出了一种三阶段方法T3。T3先对网络进行一次训练,然后用验证集进行测试以确定网络的故障曲线拐点,在此基础上通过第二次训练自适应地增加冗余隐层神经元。实验表明,T3使用较小的冗余量就可以显著改善网络对多点断路故障的容错性,在网络的容错能力和结构复杂度之间较好地达成了折衷。     

图像理解中的卷积神经网络

近年来,卷积神经网络（Convolutional neural networks,CNN）已在图像理解领域得到了广泛的应用,引起了研究者的关注. 特别是随着大规模图像数据的产生以及计算机硬件（特别是GPU）的飞速发展,卷积神经网络以及其改进方法在图像理解中取得了突破性的成果,引发了研究的热潮. 本文综述了卷积神经网络在图像理解中的研究进展与典型应用. 首先,阐述卷积神经网络的基础理论;然后,阐述其在图像理解的具体方面,如图像分类与物体检测、人脸识别和场景的语义分割等的研究进展与应用.