绝缘子污秽等级的高光谱特征优化识别技术研究

为解决传统污秽检测方法对输电线路绝缘子污闪防治的局限性,采用非接触式、高分辨率的高光谱成像技术对污秽在线检测技术进行研究。为有效提取反应污秽度的光谱特征并削弱冗余与干扰信息的影响,提出一种小波包能量谱特征优化的绝缘子污秽等级识别技术。首先,对不同污秽度的绝缘子样品的光谱图像进行背景分割,提取均匀覆污区像素点光谱均值曲线;其次,对不同图像的光强均匀度差异、环境噪声进行预处理,并通过导数变换提升不同污秽等级间的可区分性。再次,对预处理后的谱线进行小波能量谱特征提取。最后,基于所提特征建立基于支持向量机(support vector machines, SVM)的污秽等级识别模型。实验结果表明,相比于采用全波段数据或PCA特征数据作为输入,基于小波能量谱特征建立的支持向量机(SVM)污秽等级识别模型对样品识别准确率达到99.8%。#$NL关键词：关高光谱成像;绝缘子污秽等级;小波包能量谱;支持向量机#$NL中图分类号：TM933     

基于改进样板机法的风光互补新能源电站容量配比优化

风光等新能源电站出力具有间歇性和波动性,合理的风光容量配比可以充分实现二者的互补。不准确的电站理论功率计算会影响风光真实特征的提取,进而导致较大的容量配比误差。本文在新能源电站理论功率计算所通常采用的样板机法基础上进行改进,首先对异常数据进行识别及重构,然后识别新能源电站中的异常样板机并更新样板机集合,进一步根据非样板机的实际运行情况选取动态信息窗,利用动态信息窗内样板机和非样板机的实测功率,对非样板机分组并动态识别每组非样板机的比例系数,从而计算新能源电站的理论功率。基于多年历史理论功率对新能源电站进行特征分析,模拟随机出力场景,并进行场景筛选,建立了基于源荷不匹配风险的风光配比优化方法。通过算例验证了改进样板机法的准确性,利用该方法得到西北某地区电网风、光伏电站多年的改进理论功率数据,并优化得到该地区的风光最优配比。     

直线永磁磁通切换电机推力脉动抑制方法分析

直线永磁磁通切换(LFSPM)电机功率密度高,定子结构简单可靠,在长行程直线牵引场合具有很大的应用潜力。然而很多应用场合在要求直线电机具有高推力密度的同时也能够呈现较低的推力脉动。因此研究推力脉动的产生机理以及抑制方法是提升LFSPM电机应用潜力的重要手段。利用有限元软件对抑制不同成因推力脉动的方法以及这些方法的组合效果进行仿真比较,比较过程中,给出了一种的步进(错齿)位移选择方法,基于该方法,各结构在减小推力脉动的同时可以有效兼顾电机的输出推力平均值。最后对比较结果进行了分析和总结,得出有价值的结论。     

计及开关频率分区优化的PMSM三电平模型预测控制

为降低传统三电平供电的永磁同步电机模型预测控制系统的开关频率,提出一种计及开关频率分区优化的PMSM三电平模型预测控制方法。首先,将空间电压矢量平面划分为12个扇区,并且根据上一时刻电压矢量所在位置,选择相邻扇区的电压矢量作为备选矢量。其次,加入对开关状态切换的限制,每个开关周期仅允许一相开关状态发生连续性跳变,减少备选矢量数量的同时能够有效降低开关频率。进一步,利用具有相同空间位置但对中点电位影响相反的正负冗余小矢量来平衡中点电位。最后,通过仿真和实验,验证该控制策略的有效性。     

基于深度学习和灰度纹理特征的铁路接触网绝缘子状态检测

铁路接触网绝缘子状态检测对铁路行车安全有着 重大的意义,为解决目前人工对绝缘 子图像检测结果的不确定性,提出一种深度学习结合灰度纹理特征的检测方法。首先使用 Faster R-CNN (faster region-based convolutional neural network)目标检测算法对图像中绝缘子精确识别,再通过灰度共生矩阵对绝缘子纹理 特征进行分析提取,之后结合支持向量机将绝缘子分为正常绝缘子和异常绝缘子,实验数 据结果证明使用能量、熵、相关度3种纹理特征进行绝缘子状态分类时对实验数据中的正 常状态绝缘子的分类精度可达100%,异常状态绝缘子的分类精度达97.5%,最后依据绝缘 子图像灰度分布的周期性特点,利用灰度积分投影将异常绝缘子分为破损绝缘子和夹杂异 物绝缘子。实验结果表明所提方法可以有效对绝缘子状态进行检测分类。     

CBAC4C: conflict-based VM isolation control for cloud computing

For businesses to benefit from the many opportunities of cloud computing, they must first address a number of security challenges, such as the potential leakage of confidential data to unintended third parties. An inter-VM (where VM is virtual machine) attack, also known as cross-VM attack, is one threat through which cloud-hosted confidential data could be leaked to unintended third parties. An inter-VM attack exploits vulnerabilities between co-resident guest VMs that share the same cloud infrastructure. In an attempt to stop such an attack, this paper uses the principles of logical analysis to model a solution that provides physical separation of VMs belonging to conflicting tenants based on their levels of conflict. The derived mathematical model is founded on scientific principles and implemented using four conflict-aware VM placement algorithms. The resultant algorithms consider a tenant's risk appetite and cost implications. The model offers guidance to VM placement and is validated using a proof of concept. A cloud simulation tool was used to test and evaluate the effectiveness and efficiency of the model. The findings reflect that the introduction of the proposed model introduced a time lag in the time it took to place VM instances. On top of this, it was also discovered that the number and size of the VM instances has an effect on the VM placement performance. The findings further illustrate that the conflict tolerance level of a VM has a direct impact on the time it took to place.     

Mammogram Learning System for Breast Cancer Diagnosis Using Deep Learning SVM

The most common form of cancer for women is breast cancer. Recent advances in medical imaging technologies increase the use of digital mammograms to diagnose breast cancer. Thus, an automated computerized system with high accuracy is needed. In this study, an efficient Deep Learning Architecture (DLA) with a Support Vector Machine (SVM) is designed for breast cancer diagnosis. It combines the ideas from DLA with SVM. The state-of-the-art Visual Geometric Group (VGG) architecture with 16 layers is employed in this study as it uses the small size of 3 × 3 convolution filters that reduces system complexity. The softmax layer in VGG assumes that the training samples belong to exactly only one class, which is not valid in a real situation, such as in medical image diagnosis. To overcome this situation, SVM is employed instead of the softmax layer in VGG. Data augmentation is also employed as DLA usually requires a large number of samples. VGG model with different SVM kernels is built to classify the mammograms. Results show that the VGG-SVM model has good potential for the classification of Mammographic Image Analysis Society (MIAS) database images with an accuracy of 98.67%, sensitivity of 99.32%, and specificity of 98.34%.     

Cost Effective Optimal Task Scheduling Model in Hybrid Cloud Environment

In today’s world, Cloud Computing (CC) enables the users to access computing resources and services over cloud without any need to own the infrastructure. Cloud Computing is a concept in which a network of devices, located in remote locations, is integrated to perform operations like data collection, processing, data profiling and data storage. In this context, resource allocation and task scheduling are important processes which must be managed based on the requirements of a user. In order to allocate the resources effectively, hybrid cloud is employed since it is a capable solution to process large-scale consumer applications in a pay-by-use manner. Hence, the model is to be designed as a profit-driven framework to reduce cost and make span. With this motivation, the current research work develops a Cost-Effective Optimal Task Scheduling Model (CEOTS). A novel algorithm called Target-based Cost Derivation (TCD) model is used in the proposed work for hybrid clouds. Moreover, the algorithm works on the basis of multi-intentional task completion process with optimal resource allocation. The model was successfully simulated to validate its effectiveness based on factors such as processing time, make span and efficient utilization of virtual machines. The results infer that the proposed model outperformed the existing works and can be relied in future for real-time applications.     

Defect Prediction Using Akaike and Bayesian Information Criterion

Data available in software engineering for many applications contains variability and it is not possible to say which variable helps in the process of the prediction. Most of the work present in software defect prediction is focused on the selection of best prediction techniques. For this purpose, deep learning and ensemble models have shown promising results. In contrast, there are very few researches that deals with cleaning the training data and selection of best parameter values from the data. Sometimes data available for training the models have high variability and this variability may cause a decrease in model accuracy. To deal with this problem we used the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) for selection of the best variables to train the model. A simple ANN model with one input, one output and two hidden layers was used for the training instead of a very deep and complex model. AIC and BIC values are calculated and combination for minimum AIC and BIC values to be selected for the best model. At first, variables were narrowed down to a smaller number using correlation values. Then subsets for all the possible variable combinations were formed. In the end, an artificial neural network (ANN) model was trained for each subset and the best model was selected on the basis of the smallest AIC and BIC value. It was found that combination of only two variables’ ns and entropy are best for software defect prediction as it gives minimum AIC and BIC values. While, nm and npt is the worst combination and gives maximum AIC and BIC values.     

Data augmentation for low-resource languages NMT guided by constrained sampling

Data augmentation (DA) is a ubiquitous approach for several text generation tasks. Intuitively, in the machine translation paradigm, especially in low-resource languages scenario, many DA methods have appeared. The most commonly used methods are building pseudocorpus by randomly sampling, omitting, or replacing some words in the text. However, previous approaches hardly guarantee the quality of augmented data. In this study, we try to augment the corpus by introducing a constrained sampling method. Additionally, we also build the evaluation framework to select higher quality data after augmentation. Namely, we use the discriminator submodel to mitigate syntactic and semantic errors to some extent. Experimental results show that our augmentation method consistently outperforms all the previous state-of-the-art methods on both small and large-scale corpora in eight language pairs from four corpora by 2.38–4.18 bilingual evaluation understudy points.