Monitoring combustion systems using HMM probabilistic reasoning in dynamic flame images

In this paper, a novel method of on-line flame detection in video is proposed. Processing the data generated by an ordinary camera monitoring scene, it aims to early detect the current state of the combustion system and prevent the system from further degradation and occurrence of failure. Due to the dynamic change of the combustion system, the turbulent flame flicker produces images with different spatial and high temporal resolutions. The proposed method consists of hidden Markov model (HMM) and multiway principal component analysis (MPCA). MPCA is used to extract the cross-correlation among spatial relationships in the low dimensional space while HMM constructs the temporal behavior of the sequential observation. Although the prior process knowledge may not be available in the operation processes, the probability distribution of the normal status can be trained by the images collected from the normal operation processes. Subsequently, monitoring of a new observed image is achieved by a recursive Viterbi algorithm which can find the transition state sequence from series of observed image data. The proposed method, like the philosophy of traditional statistical process control, can generate simple probability monitoring charts to track the progress of the current transition state sequence and monitor the occurrence of the observable upsets. The advantages of the proposed method, data from the monitoring practice in the real combustion systems, are presented to help readers delve into the matter.     

基于紫外成像的电晕放电检测技术改进研究

针对当前电晕放电检测存在检测精度低的问题,提出一种基于紫外成像的电晕放电检测方法。该方法将基于紫外成像的电晕放电检测效果的干扰因素划分为距离、增益、温度和湿度检测等;构建了紫外线成像电晕放电检测体系,采用物镜系统解决电晕放电目标成像问题,利用光传感器将电晕放电图像进行光电转换,结合紫外滤光片滤除日盲区之外的光波,采用控制分析体系将光的强度量进行统计并分析,同时控制整个检测体系。利用图像显示体系将电晕放电图像信息融合、显示。在紫外线成像电晕放电检测体系基础上,分别对检测干扰因素逐一处理,设计基于紫外成像的电晕放电检测流程图,完成电晕放电高质量检测。实验表明,文章所提方法不仅有效处理了电晕放电检测干扰因素,还提高了放电源检测的检测精度。     

基于图像处理技术的输电线路走廊烟火识别研究

识别输电线路走廊烟火对保障输电线路安全运行有重要意义。采用图像处理技术,在分析输电走廊环境下烟火特征的基础上,首先对监控视频图像数据进行采样、灰度化、滤波和图像增强等预处理,然后选择混合高斯法构建图像背景模型,并确定前景目标图像区域,最后将前景图像和烟火做特征比较,以排除干扰物影响并得到烟火目标。仿真分析空旷野外和居民村落两种环境,验证了该方法识别烟火的有效程度。     

Automatic saw-mark detection in multicrystalline solar wafer images

This paper presents a method of automatic defect inspection for the photovoltaic industry, with a special focus on multicrystalline solar wafers. It presents a machine vision-based scheme to automatically detect saw-mark defects in solar wafer surfaces. A saw-mark defect is a severe flaw that occurs when a silicon ingot is cut into wafers. Early detection of saw-mark defects in the wafer cutting process can reduce material waste and improve production yields. A multicrystalline solar wafer surface presents random shapes, sizes, and orientations of crystal grains in the surface, making the automatic detection of saw-mark defects extremely difficult. The proposed saw-mark detection scheme involves two main procedures: (1) Fourier image reconstruction to remove the multi-grain background of a solar wafer image and (2) a line detection process in the reconstructed image to locate saw-marks. The Fourier transform (FT) is used to eliminate crystal grain patterns and results in a non-textured surface in the reconstructed image. Since a saw-mark is presented horizontally in the sliced wafer, vertical scan lines in the reconstructed image are individually evaluated by a line detection process. A pixel far away from the line sought can then be effectively identified as a defect point. Experimental results show that the proposed method can effectively detect various saw-mark defects, specifically black lines, white lines, and impurities in multicrystalline solar wafers.     

基于视频图像处理技术的火焰识别算法

针对现有的真火训练评估系统难以直接准确评价出实际灭火效果,以及部分真火训练环境比较空旷的特点,提出了一种基于视频图像处理技术识别火焰的算法,该算法能够直接检测出火焰大小,从而实现对训练效果更加直观准确的评估。通过对比分析的方法,确定采用YCbCr彩色模型算法,该算法拥有其自身独特的优点,即减轻改变照明度造成的负面影响并且提高火焰检测的准确率。为提高评估的准确性,针对火灾检测技术的难点和关键点提出了基于视频图像的解决办法,即先对火焰的圆形度、火焰边界的粗糙度和尖角个数分别进行特征提取,再进行动态特征提取,完成数据归一化处理。采取训练集对分类器展开训练,再根据训练集得到的模型对测试集进行预测,最终获取分类识别率。以某实际真火训练的1000张图片为样本,对模型中的火焰图像处理效果进行验证。实验结果表明:基于饱和度的规则和颜色模型能够更准确地体现火焰行为,在不利的照明背景条件下其鲁棒性更强。对比激光束检测、超声波检测和水量检测等方法,基于图像的火焰识别评估方式具有不易变形、定量直观、准确性高及操作方便等优点。     

基于图像处理的光伏局部阴影区域分割方法

为解决真实场景中光伏组件局部阴影区域分割受环境干扰的问题,该文提出基于图像处理的光伏局部阴影区域分割方法。根据光伏组件特征用改进的阈值分割方法提取组件边框,采用Hough变换检测组件边框线段,计算包围边框线段端点的最小凸包作为感兴趣区域。在感兴趣区域内再次使用改进的阈值分割方法提取局部阴影区域。用模拟仿真图像和真实场景图像验证该文所提方法的有效性,实验结果表明：所提方法可准确分割出光伏组件上的局部阴影区域且保持了遮挡区域的细节信息。与多阈值分割方法和基于Canny边缘检测的分割方法相比,该文所提方法的错误分类误差更小,类别像素准确率更高。     

图像稠密坐标网格的去除

针对工程图为脉冲边滑曲线图且稠密坐标网格使得边缘提取成功率较低,提出了基于Hough变换与基于线检测对水轮机特性曲线稠密坐标网格去除方法.利用Canny算法对图像进行边缘检测,采用Hough变换检测图像中直线,并分析了实值标量取值对检测结果的影响,从原图像中减去提取的直线即为被去除坐标网格的图像.实验结果表明,基于线检测模板对去除稠密坐标网格具有较好的效果.     

基于最优多尺度集的形态学红外图像缺陷检测研究

图像增强在风力机叶片红外缺陷检测中具有重要应用,该文针对红外检测中图像清晰度低的问题,提出一种自适应迭代阈值加权的差分顶帽变换红外图像增强方法（AITWD top-hat）。首先,通过多尺度白顶帽与黑顶帽变换获得多组顶帽信息。其次,为获得更多的图像细节特征,采用对比度改善系数（contrast improvement ratio, CIR）作为阈值数,提出一种迭代阈值的图像加权方法,并对获得的最优尺度集图像进行融合,得到增强图像。最后,将该文所提方法与现有方法从定性和定量2个方面进行对比和验证。实验结果表明,该文方法不仅具有较好的图像细节增强效果,而且有效克服了传统形态学图像增强时稳定性差的问题,PSNR值和SSIM值分别提高了9.01%和9.07%。     

基于迁移学习的光伏组件鸟粪覆盖检测

该文基于无人机光伏组件可见光图像采集,提出一种基于迁移学习的光伏组件鸟粪检测方法。方法首先基于掩膜区域卷积神经网络（Mask-RCNN）对光伏组件边界进行框选,再利用迁移学习策略,构建光伏组件鸟粪缺陷检测模型,实现鸟粪智能检测。利用上述方法,实现光伏组件鸟粪覆盖检测准确率为96.75%。     

基于深度卷积神经网络的光伏组件热斑检测

热斑效应是造成光伏组件损坏的主要原因之一,提早发现光伏组件热斑效应并及时解决,可有效减少损失。该文针对热斑效应问题提出改进的Faster R-CNN红外热斑图像检测方法,该方法是基于SpotFPN多尺度特征学习模块,将SpotFPN应用在二阶段目标检测网络中提高了模型的检测精度,改善热斑的识别准确率。同时为解决数据集不足出现的过拟合问题,利用数据增强技术有效扩充红外热斑数据集。检测模型采用预训练权重对热斑数据集进行学习,其平均精度较改进前提高了约3%。     

基于显著性检测的风机叶片裂纹提取研究

在基于数字图像处理技术的风力发电机桨叶故障巡检过程中,为了提取桨叶上的裂纹,结合裂纹与桨叶对比度的特点,文章提出了一种基于显著性区域检测与连通域原理相结合的裂纹提取方法。利用灰度化和基于引导滤波器的图像增强算法对图像进行预处理。然后转变处理前景目标区域的方法,转而分析背景区域的特点,达到把背景区域去除的效果,并与连通域检测标记相结合,最终把裂纹区域提取出来。实验结果表明,基于显著性区域检测的方法能够把风机桨叶上的裂纹较为完整地标记出来。     

Bird detection and species classification with time‐lapse images around a wind farm: Dataset construction and evaluation

Collisions of birds, especially endangered species, with wind turbines is a major environmental concern. Automatic bird monitoring can be of aid in resolving the issue, particularly in environmental risk assessments and real‐time collision avoidance. For automatic recognition of birds in images, a clean, detailed, and realistic dataset to learn features and classifiers is crucial for any machine‐learning‐based method. Here, we constructed a bird image dataset that is derived from the actual environment of a wind farm and that is useful for examining realistic challenges in bird recognition in practice. It consists of high‐resolution images covering a wide monitoring area around a turbine. The birds captured in these images are at relatively low resolution and are hierarchically labeled by experts for fine‐grained species classification. We conducted evaluations of state‐of‐the‐art image recognition methods by using this dataset. The evaluations revealed that a deep‐learning‐based method and a simpler traditional learning method were almost equally successful at detection, while the former captures more generalized features. The most promising results were provided by the deep‐learning‐based method in classification. The best methods in our experiments recorded a 0.98 true positive rate for bird detection at a false positive rate of 0.05 and a 0.85 true positive rate for species classification at a false positive rate of 0.1.     

基于图像识别的高温熔渣流量测量方法研究

熔渣流量是影响高温熔渣离心粒化效果的关键因素。为了测量熔渣流量，提出了一种流量测量方法，利用高速摄影机拍摄液柱流动图像，采用边缘检测、阈值分割等方法，获得液柱直径及流速，计算流量，采用常温甘油溶液作为测试对象进行实验，研究该方法测量精度的影响因素。结果显示，长曝光时间、合理的拍摄帧率、短物距有利于提高测量精度；同时，感兴趣区域（ROI）的位置为距管口2~5倍管径处、ROI长度为1/3~2/3倍管径时，测量精度较高。通过对实际高温熔渣流动过程的检测，其测量误差约9.0%，证明了该方法的实用性。     

Infrared Tracking From Morphological Image Processing Tools—Application to Heat Transfer Characterization in Granular Media

The main objective of this study is to track some moving objects in an infrared image sequence and retrieve their temperature field. We first implement some tracking techniques commonly used in the computer vision field. Unfortunately, trackers of this kind are not directly well suited for thermal frame monitoring, since heat diffusion yields both light level changes and three-dimensional effects. The proposed algorithms are thus adapted and then tested for velocity measurements in real experiments involving the stochastic moving of spheres on an oscillating plate. Circle Hough transforms, digital image correlation, and mathematical morphology tools are considered and their efficiency is evaluated. The labeling of the moving objects is then implemented in order to retrieve and discriminate the thermal field of each target within a lagrangian approach. Heat transfer between the moving object and the surrounding medium can then be characterized: In the case presented herein, the convective heat transfer coefficient is estimated.     

一种新的风电场复杂下垫面场地校准模型

针对现有方法模型简单、气象因素考虑不足和缺少异常数据检测等问题,提出一种基于相对密度离群因子（relative RDOF）异常检测和支持向量回归（SVR）组合的场地校准模型。首先,通过RDOF与四分位法对测风塔与风电机组距下垫面相同高度风速间的风速比进行异常数据检测,间接实现风速预处理;同时,根据测风塔与风电机组相同高度风切变序列的相关性,动态引入大气稳定度等级,与多种常规自然因素共同作为模型输入。然后,建立SVR场地校准模型修正风速。最后,通过算例验证,并与国际电工委员会（IEC）标准方法进行对比。结果表明,该方法可有效提升场地校准模型的风速修正精度。     

A novel anti islanding detection method for grid connected fuel cell power generation systems

Renewable energy sources have been developed rapidly all around the world, and one of these green energy sources is hydrogen energy. The fuel cell systems have become prominent in renewable energy sources because of its minimal dimensions and energy conversion method. There have been developed, some applications, especially in domestic and automotive areas, and fuel cell systems are also have been started to use in grid connected systems. Fuel cell systems must have some electrical connection standards while they connected to an electrical grid. One of these electrical conditions and may be the most important one is unplanned islanding condition. Islanding is a very dangerous situation because it can damage to the fuel cell and related electrical systems and also working people have been at risk in islanding situation on the grid. In this study, a novel islanding detection method was introduced for grid connected fuel cell systems. 0.5 kW solid oxide fuel cell (SOFC) system used in developed experimental system and a novel anti islanding detection method was researched by using an effective method. The proposed method was also developed by using Matlab Simulink and its useful tools. The developed islanding detection method is robust, reliable and has a fast response time, according to present methods. The results confirm the suggested conditions, and it can be seen in this method, it can also be adapted easily to the grid connected fuel cell systems.     

电网正、负序电压基波分量的提取算法在并网型逆变器控制中的应用

提出一种在电网三相电压不平衡的情况下,通过三角函数运算的方法解耦不平衡电网基波正、负序电压分量。由于电网负序电压分量送入锁相环后会造成锁相环的震荡,导致锁相环输出误差大无法精确锁相。根据正序分量与负序分量之间的相角关系,推导出提取正序分量的公式,把该公式进行数字化处理,将其应用于dq数字锁相环,提高了dq锁相环在电网三相电压不平衡条件下的有效性和准确性。提出了网压前馈对逆变器控制的重要性,分析了利用正序电压瞬时值进行前馈控制的优点。并通过实验验证了公式的有效性和控制的稳定性、可靠性。     

Optical analysis of solar facility heliostats

Useful insights into significant operating parameters are gained by treating the mirror collector system of a central receiver solar power station as an optical system, no matter how large. We at Sandia Laboratories have developed a method for estimating the approximate size of the solar image cast by individual heliostats. As a consequence from the Coddington equations, we have developed a simple analysis of astigmatism, the major aberration which describes optical off-axis image behavior of a focusing collector system. These predictive equations agree well with experiments performed with a spherical mirror over a range of angles of incidence exceeding 60°. Other than flat mirros, several heliostat configurations were proposed and explored and were found amenable to the same analysis. All desings which attempt to superpose the reflected energy from several mirros mounted and tracked together on the same frame are subject to the same simple rules. Although this analysis is not rigorous, it does give useful insight into the cause of the large image spread found under some operating conditions and it does indicate methods for optimizing the system. Conclusions reached are explicitly related to the off-axis performance of the mirror collectors of a central receiver solar power station.     

基于模式识别的燃气管网泄漏检测技术

管道泄漏检测是确保燃气管道安全运行的重要技术手段,然而由于城市燃气管网是错综复杂的环状管网,目前尚没有一种方法能较好的对城市管网进行安全监控和泄漏检测。模式识别技术是燃气管网泄漏检测技术发展的方向,针对环状管网应用模式识别技术中存在特征向量维度大的问题,提出了管网分区检测的思想,并阐述了其在管网泄漏检测中的具体步骤。最后通过一实例验证了模式识别法用于城市环状燃气管网泄漏检测的可行性。     

主元分析方法在火电厂锅炉过程故障检测中的应用

研究了将主元分析方法(PCA)应用到火电厂锅炉控制过程的故障检测的若干问题，提出了适合于此类过程的改进PCA建模方法和在线故障检测方法。对过程数据的处理、变量相关关系的发掘和主元检测模型的建立以及在线检测算法等方面进行了阐述。最后，基于现场采集的长期历史数据，给出了在锅炉过程传感器故障检测中的应用实例，试验结果表明了所提出的方法的可行性。图3表1参10