Position computation models for high-speed train based on support vector machine approach

High-accuracy positioning is not only an essential issue for efficient running of high-speed train (HST), but also an important guarantee for the safe operation of high-speed train. Positioning error is zero when the train is passing through a balise. However, positioning error between adjacent balises is going up as the train is moving away from the previous balise. Although average speed method (ASM) is commonly used to compute the position of train in engineering, its positioning error is somewhat large by analyzing the field data. In this paper, we firstly establish a mathematical model for computing position of HST after analyzing wireless message from the train control system. Then, we propose three position computation models based on least square method (LSM), support vector machine (SVM) and least square support vector machine (LSSVM). Finally, the proposed models are trained and tested by the field data collected in Wuhan-Guangzhou high-speed railway. The results show that: (1) compared with ASM, the three models proposed are capable of reducing positioning error; (2) compared with ASM, the percentage error of LSM model is reduced by 50.2% in training and 53.9% in testing; (3) compared with LSM model, the percentage error of SVM model is further reduced by 38.8% in training and 14.3% in testing; (4) although LSSVM model performs almost the same with SVM model, LSSVM model has advantages over SVM model in terms of running time. We also put forward some online learning methods to update the parameters in the three models and better positioning accuracy is obtained. With the three position computation models we proposed, we can improve the positioning accuracy for HST and potentially reduce the number of balises to achieve the same positioning accuracy.     

A generalized method for the inherent energy performance modeling of machine tools

Machine tools (MTs), as the key equipment of manufacturing systems, have enormous quantities and consume a great amount of energy. However, the diversity of both machines and their energy consumption properties make it difficult to transfer the energy-saving knowledge and services among different MT. To facilitate the initialization configuration of energy-saving services, the inherent energy performance (IEP) is investigated to describe the differences in energy consumption among MTs, and a generalized method for modeling the IEP of MT and its electrical subsystems is proposed. Three key enablers, including generalized experimental design rules, automatic coding, and data processing algorithms, are presented and integrated into a supporting system to reduce the modeling efforts and knowledge requirements. Case studies of an offline manufacturing scenario and an Internet of Things (IoT)-enabled manufacturing scenario were carried out to verify the effectiveness and convenience of the proposed method. The results show that the proposed method can provide essential modeling support for large-scale energy-saving service configurations and energy-efficient MT development.     

Classification and regression models of audio and vibration signals for machine state monitoring in precision machining systems

We present a data-driven method for monitoring machine status in manufacturing processes. Audio and vibration data from precision machining are used for inference in two operating scenarios: (a) variable machine health states (anomaly detection); and (b) settings of machine operation (state estimation). Audio and vibration signals are first processed through Fast Fourier Transform and Principal Component Analysis to extract transformed and informative features. These features are then used in the training of classification and regression models for machine state monitoring. Specifically, three classifiers (K-nearest neighbors, convolutional neural networks and support vector machines) and two regressors (support vector regression and neural network regression) were explored, in terms of their accuracy in machine state prediction. It is shown that the audio and vibration signals are sufficiently rich in information about the machine that 100% state classification accuracy could be accomplished. Data fusion was also explored, showing overall superior accuracy of data-driven regression models.     

基于特征选择和深度信念网络的文本情感分类算法

由于人类语言的复杂性，文本情感分类算法大多都存在因为冗余而造成的词汇量过大的问题。深度信念网络（DBN）通过学习输入语料中的有用信息以及它的几个隐藏层来解决这个问题。然而对于大型应用程序来说，DBN是一个耗时且计算代价昂贵的算法。针对这个问题，提出了一种半监督的情感分类算法，即基于特征选择和深度信念网络的文本情感分类算法（FSDBN）。首先使用特征选择方法（文档频率（DF）、信息增益（IG）、卡方统计（CHI）、互信息（MI））过滤掉一些不相关的特征从而使词汇表的复杂性降低；然后将特征选择的结果输入到DBN中，使得DBN的学习阶段更加高效。将所提算法应用到中文以及维吾尔语中，实验结果表明在酒店评论数据集上，FSDBN在准确率方面比DBN提高了1.6%，在训练时间上比DBN缩短一半。     

A taxonomy and survey of attacks against machine learning

The majority of machine learning methodologies operate with the assumption that their environment is benign. However, this assumption does not always hold, as it is often advantageous to adversaries to maliciously modify the training (poisoning attacks) or test data (evasion attacks). Such attacks can be catastrophic given the growth and the penetration of machine learning applications in society. Therefore, there is a need to secure machine learning enabling the safe adoption of it in adversarial cases, such as spam filtering, malware detection, and biometric recognition. This paper presents a taxonomy and survey of attacks against systems that use machine learning. It organizes the body of knowledge in adversarial machine learning so as to identify the aspects where researchers from different fields can contribute to. The taxonomy identifies attacks which share key characteristics and as such can potentially be addressed by the same defence approaches. Thus, the proposed taxonomy makes it easier to understand the existing attack landscape towards developing defence mechanisms, which are not investigated in this survey. The taxonomy is also leveraged to identify open problems that can lead to new research areas within the field of adversarial machine learning.     

Stacked autoencoder based deep random vector functional link neural network for classification

Extreme learning machine (ELM), which can be viewed as a variant of Random Vector Functional Link (RVFL) network without the input–output direct connections, has been extensively used to create multi-layer (deep) neural networks. Such networks employ randomization based autoencoders (AE) for unsupervised feature extraction followed by an ELM classifier for final decision making. Each randomization based AE acts as an independent feature extractor and a deep network is obtained by stacking several such AEs. Inspired by the better performance of RVFL over ELM, in this paper, we propose several deep RVFL variants by utilizing the framework of stacked autoencoders. Specifically, we introduce direct connections (feature reuse) from preceding layers to the fore layers of the network as in the original RVFL network. Such connections help to regularize the randomization and also reduce the model complexity. Furthermore, we also introduce denoising criterion, recovering clean inputs from their corrupted versions, in the autoencoders to achieve better higher level representations than the ordinary autoencoders. Extensive experiments on several classification datasets show that our proposed deep networks achieve overall better and faster generalization than the other relevant state-of-the-art deep neural networks.     

结合优化支持向量机与K-means++的工控系统入侵检测方法

针对工业控制系统传统单一检测算法模型对不同攻击类型检测率和检测速度不佳的问题，提出一种优化支持向量机和K-means++算法结合的入侵检测模型。首先利用主成分分析法（PCA）对原始数据集进行预处理，消除其相关性；其次在粒子群优化（PSO）算法的基础上加入自适应变异过程避免在训练的过程中陷入局部最优解；然后利用自适应变异粒子群优化（AMPSO）算法优化支持向量机的核函数和惩罚参数；最后利用密度中心法改进K-means算法与优化后的支持向量机组合成入侵检测模型，从而实现工业控制系统的异常检测。实验结果表明，所提方法在检测速度和对各类攻击的检测率上得到明显提升。     

基于DRGAN和支持向量机的合成孔径雷达图像目标识别

为解决SAR图像目标识别中样本缺乏和方位角敏感问题,提出了一种基于DRGAN和SVM的SAR图像目标识别算法。首先,采用多尺度分形特征对SAR图像进行增强,经过分割得到目标二值图像,基于Hu二阶矩估计目标的方位角。然后对估计得到的目标方位角进行量化编码,结合原始图像作为输入,对设计的DRGAN模型参数进行训练与优化。由于DRGAN中的深度生成模型将目标姿态与外观表示进行解耦设计,故可利用该模型将SAR图像样本变换到同一方位角区间。基于变换后的训练样本分别提取归一化灰度特征,利用SVM训练分类器。采用MSTAR数据集在多个不同操作条件下对提出的算法进行测试,实验结果表明,在带变体的标准操作条件下,能够达到97.97%的分类精度,优于部分基于CNN模型的分类精度,在4种扩展操作条件下的分类精度分别为97.83%,91.77%,97.11%和97.04%,均优于传统方法的分类精度。在SAR图像目标方位角估计存在一定误差的情况下,训练得到的GAN模型作为SAR图像目标旋转估计器,能够使得在不进行复杂样本预处理的前提下,仍然取得较高的SAR图像目标识别精度。     

一种新型自动化蓝莓采摘机的设计

针对小浆果种植领域中蓝莓收获难的问题,基于计算机软件和可编程序控制器,设计了一种新型自动化蓝莓采摘机。介绍了这一蓝莓采摘机的结构与工作原理,分析了控制方案,制作了样机,并进行了采摘试验。应用这一新型自动化蓝莓采摘机,可以满足蓝莓收获效率高、破损率低等要求。     

基于改进PSO-SVM的钻井液侵入储层深度预测

开展钻井液侵入储层深度预测,对于测井评价以及提高油井产能具有一定的现实意义。在分析钻井液侵入储层的机理和特征的基础上,提出了钻井液侵入储层的影响因素指标体系,建立了改进PSO-SVM的钻井液侵入储层深度预测模型,以塔里木塔中35口井为例进行了实证分析,并与传统BP神经网络和SVM模型预测结果进行了对比。研究结果表明:侵入深度与泥饼的渗透率、钻井液与储层压差以及侵入时间正相关,与储层孔隙度和钻井液粘度负相关,改进的PSO-SVM模型预测结果误差小,准确率高,能够用于钻井液侵入储层深度预测,具有广泛的应用前景。