Spatial mask filtering algorithm for partial discharge pulse extraction of large generators

A spatial mask filter algorithm （SMF） for partial discharge （PD） pulse extraction is proposed in this paper. In this algorithm, firstly, a ＇Teager＇ operator is used to strengthen wavelet coefficient local energy; then direct multiplication of coefficients at two adjacent scales is used to detect singularity points of the signal and to obtain scale based spatial mask filter; finally, an ＇ AND＇ logic operator is used in different filters to obtain the last spatial mask filter. By multiplication of wavelet coefficients with the final mask filter and wavelet reconstruction process, partial discharge pulses are extracted. The results of digital simulation and practical experiment show that this method is superior to traditional wavelet shrinkage method （TWS）. This algorithm not only can increase the signal to noise ratio （SNR）, but also can preserve the encrgy and pulse amplitude.

Extraction of Partial Discharge Acoustic Signal by Wavelet Transform with Teager′s Energy Operator

To develop a measurement system for monitoring partial discharge （PD） without the effect of external interferences,an algorithm of PD signal extraction based on wavelet transform with Teager＇s energy operators was presented. Acoustic signal generated by PD was selected to remove excessive interfering signals and electromagnetic interferences. Acoustic signals were collected and decomposed into I0 levels by wavelet transform into approximation and detail components. “Daubechies 25” was proved to be the most suitable mother wavelet for the extraction of PD acoustic signals. Compared with conventional wavelet denoising method, Teager＇s energy operators were adopted to the PD signal reconstruction and the signal to noise ratio was in creased by 20%-25% inthe experiment,without lost in energy and pulse amplitude.

Partial Formalization: An Approach for Critical Analysis of Definitions and Methods Used in Bulk Extraction-Based Molecular Microbial Ecology

Partial formalization, which involves the development of deductive connections among statements, can be used to examine assumptions, definitions and related methodologies that are used in science. This approach has been applied to the study of nucleic acids recovered from natural microbial assemblages (NMA) by the use of bulk extraction. Six pools of bulk-extractable nucleic acids (BENA) are suggested to be present in a NMA: (pool 1) inactive microbes (abiotic-limited);(pool 2) inactive microbes (abiotic permissive, biotic-limited);(pool 3) dormant microbes (abiotic permissive, biotic-limited, but can become biotic permissive);(pool 4) in situ active microbes (the microbial community);(pool 5) viruses (virocells/virions/cryptic viral genomes);and (pool 6) extracellular nucleic acids including extracellular DNA (eDNA). Definitions for cells, the microbial community (in situ active cells), the rare biosphere, dormant cells (the microbial seed bank), viruses (virocells/virions/cryptic viral genomic), and diversity are presented, together with methodology suggested to allow their study. The word diversity will require at least 4 definitions, each involving a different methodology. These suggested definitions and methodologies should make it possible to make further advances in bulk extraction-based molecular microbial ecology.

Chromatic processing for feature extraction of PD-induced UHF signals in GIS

Partial discharge(PD) detection is an effective means of discovering insulation faults in gas-insulated switchgear(GIS). One of the most extensively used methods in PD detection has historically been the ultrahigh frequency(UHF) method. This study evaluates the chromatic processing methodology and its key factors for feature extraction of UHF signals in GIS. Three types of artificial defects are installed in the GIS tank at 0°, 90°, and 180°, respectively. The features of the UHF signals are extracted in the chromatic space, and PD discrimination of the defects is achieved. The influences of processors are studied before the feature selections are suggested. The time-stepping method is proposed to determine the rules of UHF signal frequency characteristics that vary with time. Finally, the process and options of the chromatics-inspired methodology are summarized.

Selective ensemble modeling based on nonlinear frequency spectral feature extraction for predicting load parameter in ball mills

Strong mechanical vibration and acoustical signals of grinding process contain useful information related to load parameters in ball mills. It is a challenge to extract latent features and construct soft sensor model with high dimensional frequency spectra of these signals. This paper aims to develop a selective ensemble modeling approach based on nonlinear latent frequency spectral feature extraction for accurate measurement of material to ball volume ratio. Latent features are first extracted from different vibrations and acoustic spectral segments by kernel partial least squares. Algorithms of bootstrap and least squares support vector machines are employed to produce candidate sub-models using these latent features as inputs. Ensemble sub-models are selected based on genetic algorithm optimization toolbox. Partial least squares regression is used to combine these sub-models to eliminate collinearity among their prediction outputs. Results indicate that the proposed modeling approach has better prediction performance than previous ones.

融合Partial卷积与残差细化的遥感影像建筑物提取算法

由于高空间分辨率遥感图像中背景与建筑物对象的相似度高,导致网络难以兼顾不同大小的建筑物,建筑边界区域的像素与背景混淆,建筑边界很容易被漏检。为解决上述问题,提出融合Partial卷积与残差细化的遥感影像建筑物提取算法(UUNet)。以U-Net为基线网络,首先,改进编码器。在编码器前端加入两个Conv4×4,在最初扩大感受野,捕捉更多遥感影像特征信息,利用Partial卷积(PConv3×3)构造的PC模块,增强编码器提取多尺度建筑物特征的能力,用Conv2×2进行两倍下采样,减少建筑物特征信息丢失。其次,减少参数量。裁剪U-Net网络解码器三层结构为UUNet网络解码器。最后,增加改进的残差细化模块。在解码器输出端构造裁剪到三层结构的U型残差细化模块,对解码器输出的粗糙建筑物特征图进行进一步提纯,使建筑物边缘信息更加清晰,网络解码器与U型残差细化模块编码器进行跳跃连接,保留最初特征,将SimAM嵌入细化模块中,提高建筑物关注度,优化网络改善边界模糊,提升目标边界提取质量。在Satellite datasetⅡ(East Asia)数据集上进行消融实验,UUNet比U-Net的IoU_(Building)、IoU_(Background)、F1、OA和MIoU分别提高2.78个百分点、0.12个百分点、1.91个百分点、0.19个百分点、1.45个百分点,表明UUNet网络优于基线网络;在Satellite datasetⅡ(East Asia)数据集和WHU数据集上做对比实验,UUNet相较于现有的主流算法更优,能够显著地提升高分辨率遥感影像中建筑物提取的效果。

Application of PDE and Mathematical Morphology in the Extraction Validation of the Roads

The digital images generated by remote sensors often contain noises that are inherent in the process of imaging and transmission. The application of digital processing techniques greatly enhances the ability to extract information on surface targets from remote sensing data. When digital images are used with high spatial resolution, one of the problems emerging the high variability of targets presents in such images. From the computational point of view, the use of partial differential equations is favored by the large number of numerical methods showed in the literature. Many of the models are considered non-complex both from the mathematical and computational standpoints, due to the characteristics of explicit equations. This work uses techniques of the partial differential equations (PDE) and mathematical morphology to extract cartographic features in digital images of the remote sensing. The selected study area corresponds to an image containing part of the Mário Covas Ring Road, located in the metropolitan region of Sao Paulo (SP), Brazil. The results are promising and show the high potential of using mathematical morphology in the field of cartography.

Partial Purification and Characterization of Protease from Abrus precatorius Linn. (Fabaceae) from Cameroon

Crude enzyme extracts were prepared from leaves and stems of Linn. (Fabaceae) from Cameroon under optimized conditions. Proteolytic enzymes were precipitated with ammonium sulfate at 35% (w/v) saturation and assayed for enzyme activity. The effects of temperature, pH, incubation time and substrate specificity were studied. SDS-PAGE was used to determine molecular weight of precipitated protease. Results indicated that proteolytic activity of crude extract was 35.20 U/ml compared to 51.03 U/ml of partial purified extract. The optimum enzyme activity was found to be at 40°C, while 50% of activity was maintained at 60°C after 60 min incubation. Partial purified crude extract exhibited two optimum pH (2.75 and 9.0). The highest enzyme activity towards Bovine Serum Albumine (25.9 U/ml) was noted. SDS-PAGE gels exhibited molecular weight between 40 - 60 KDa. This result confirms that partial purified extract of A. precatorius contains proteases and could be a promising source for proteolytic enzyme extraction.

A Novel Aquila Optimizer Based PV Array Reconfiguration Scheme to Generate Maximum Energy under Partial Shading Condition

This paper develops a real-time PV arrays maximum power harvesting scheme under partial shading condition(PSC)by reconfiguring PV arrays using Aquila optimizer(AO).AO is based on the natural behaviors of Aquila in capturing prey,which can choose the best hunting mechanism ingeniously and quickly by balancing the local exploitation and global exploration via four hunting methods of Aquila:choosing the searching area through high soar with the vertical stoop,exploring in different searching spaces through contour flight with quick glide attack,exploiting in convergence searching space through low flight with slow attack,and swooping through walk and grabbing prey.In general,PV arrays reconfiguration is a problem of discrete optimization,thus a series of discrete operations are adopted in AO to enhance its optimization performance.Simulation results based on 10 cases under PSCs show that the mismatched power loss obtained by AO is the smallest compared with genetic algorithm,particle swarm optimization,ant colony algorithm,grasshopper optimization algorithm,and butterfly optimization algorithm,which reduced by 4.34%against butterfly optimization algorithm.

A New Feature Space for Partial Discharge Signal Separation Based on DWT Coefficient Variance

In this paper, a new feature space for PD （partial discharge） signal separation is presented. Three typical PD defects were experimentally reproduced in a laboratory for obtaining independent PD sources. Signals were acquired with a digital storage oscilloscope and then post-processed with DWT （discrete Wavelet transform） for de-noising. The new feature space for PD source separation was constructed with the variance of each Wavelet coefficient vector and was compared with an established feature space for PD source separation; based on the energy of DWT coefficient vectors. After a space reduction by mean of PCA （principal components analysis）, the separation capability among them was measured by comparing the final classification error after training a neural network Results showed that with this new feature space it is possible to separate different sources of PD signals. Later, the feature space proposed was used to separate two PD sources from a real equipment tested. Further analysis on the reduced feature space has shown the band location of PD signals information for separating purpose.

基于Transformer紧凑编码的局部近重复视频检测算法

针对现有局部近重复视频检测算法特征存储消耗大、整体查询效率低、提取特征时并未考虑近重复帧之间细微的语义差异等问题,文中提出了一种基于Transformer紧凑编码的局部近重复视频检测算法。首先,提出了一个基于Transformer的特征编码器,其学习了大量近重复帧之间细微的语义差异,可以在编码帧特征时对各个区域特征图引入自注意力机制,在有效降低帧特征维度的同时也提高了编码后特征的表示性。该特征编码器通过孪生网络训练得到,该网络不需要负样本就可以有效学习近重复帧之间的相似语义信息,因此无需沉重和困难的难负样本标注工作,使得训练过程更加简易和高效。其次,提出了一个基于视频自相似度矩阵的关键帧提取方法,可以从视频中提取丰富但不冗余的关键帧,从而使关键帧特征序列能够更全面地描述原视频内容,提升算法的性能,同时也大幅减少了存储和计算冗余关键帧带来的开销。最后,基于关键帧的低维紧凑编码特征,采用基于图网络的时间对齐算法,实现局部近重复视频片段的检测和定位。该算法在公开的局部近重复视频检测数据集VCDB上取得了优于现有算法的实验性能。

单边区间集模糊半概念的知识表示及提取方法

为了使“区间”形式加以表述的不确定信息的提取具有侧重性,需提取出对象(属性)集对应的属性(对象)区间集。本文在模糊形式背景中,通过引入2个阈值,将单边区间集与经典半概念结合,提取出属性(对象)集对应的对象(属性)区间集,从而提出区间集外延–集合内涵(集合外延–区间集内涵)(interval set extent-set intent(set extent-interval set intent),ISE-SI(SE-ISI))型单边区间集模糊半概念。全体ISE-SI(SE-ISI)型单边区间集模糊半概念构成格,并给出基于格搜寻全体ISE-SI(SE-ISI)型单边区间集模糊半概念的算法。通过与已有成果对比,显示出这2种知识表示形式的多方优势。本文所得结果在知识表示及提取方法上具有适用范围广、实际应用强等优点。

经典名方旋覆代赭汤乙酸乙酯部位指纹图谱及化学模式识别研究

目的:建立旋覆代赭汤乙酸乙酯部位UPLC指纹图谱,结合化学模式识别分析不同产地饮片对旋覆代赭汤质量的影响。方法:采用Waters UPLC BEH C_(18)(2.1 mm×150 mm,1.7μm)色谱柱,流动相为乙腈-甲醇-0.4%甲酸水溶液,体积流量0.1 mL·min^(-1),梯度洗脱,检测波长350 nm;建立13批不同产地饮片提取的旋覆代赭汤样品指纹图谱,进行相似度评价,并结合聚类分析(CA)、主成分分析(PCA)和正交偏最小二乘法-判别分析(OPLS-DA),对旋覆代赭汤指纹图谱数据进行分析。结果:13批旋覆代赭汤指纹图谱确定22个共有特征峰,指认出12个多酚类成分,其中11个多酚类成分来自于旋覆花,其相似度均在0.9以上;通过CA发现,13批样品可按旋覆花产地分为三类;PCA与CA结果基本一致,并提取出4个主成分;OPLS-DA筛选出影响分类的差异性质量标志物,其中已指认出的成分有异绿原酸、异槲皮苷、异鼠李素-3-O-葡萄糖苷、异绿原酸B、1,5-二咖啡酰奎宁酸、绿原酸、槲皮万寿菊苷、槲皮素。结论:通过指纹图谱结合化学模式识别技术的分析策略,可快速有效地筛选不同批次旋覆代赭汤中多酚类成分中的差异质量标志物,为后续旋覆代赭汤的药效物质基础研究和质量评价提供参考。

基于FrFT和RVM的变压器局部放电模式识别

快速准确识别局部放电类型对于保证变压器安全稳定运行具有重要意义。针对局部放电信号模式识别中面临的最优特征参数提取和分类器设计难题,提出一种基于分数阶傅里叶变换(fractional Fourier transform,FrFT)和相关向量机(relevance vector machine,RVM)的局部放电模式识别方法。首先将FrFT引入局部放电信号分析领域,利用FrFT将局部放电信号转换至分数域并对其进行多尺度分析,在扩充信息提取维度的同时,提取可反映不同局部放电信号波形差异的14维特征构成特征向量;然后将特征向量作为输入,建立RVM模型进行最优特征选择和分类判决函数的联合优化,从而实现对不同局部放电信号的分类识别。建立电晕放电、沿面放电和气隙放电试验模型并采集局部放电超声信号开展试验,结果表明所提方法对于每种局部放电信号均能获得较高的识别精度,平均正确识别率相对于常规支持向量机(support vector machine,SVM)分类方法提升超过2.7%。

并行特征提取和渐进特征融合的计算机主板装配缺陷检测

针对计算机主板装配缺陷检测中的元器件位置分布复杂、缺陷目标不显著及多尺度等问题,本文提出了一种并行特征提取和互交叉渐进特征融合的端到端的缺陷检测算法。首先,结合部分卷积和视觉Transformer提出了一种并行残差特征提取网络,利用部分卷积的低计算复杂度的优势提取局部特征,同时利用视觉Transformer的长距离建模能力扩大模型的感受野,增强网络的特征提取能力。其次,引入注意力机制和特征渐进融合机制,提出了一种多尺度注意力互交叉的渐进特征融合网络,增强检测模型的特征融合能力。在公开数据集上的实验结果表明,该算法的平均精度均值(mAP)达到了94.63%,相较于基线模型YOLOv5提升了4.62%,并优于其他几种先进模型,检测速度达到了25 FPS。实现了较好的检测精度与速度的平衡,为实际工业环境下计算机主板表面装配缺陷检测自动化和智能化的实现提供了一种快速、有效的方法。

塔里木盆地鸟山地区铀矿化信息探测技术应用

塔里木盆地分布有我国最大的塔克拉玛干沙漠,石油、天然气资源丰富,铀矿产资源潜力巨大。当前,该盆地沙漠覆盖区铀矿勘查程度低,铀矿找矿确定亟待投入大量的铀矿勘查工作。在沙漠覆盖区开展放射性地球物理和穿透性地球化学方法勘探,可有效探测识别深部铀矿化信息异常,有助于定位预测深部铀矿体。为此,在塔里木盆地巴楚隆起区西南缘的鸟山沙漠覆盖区内,采用活性炭吸附测氡和分量化探方法,应用弱信息提取技术、组合、综合信息提取技术及方向导数,提取沙漠覆盖区深部铀矿化信息异常,异常受鸟山断裂带及鸟山东断裂带控制,鸟山沙漠覆盖区氡浓度平均值为910 Bq·m^(-3),其值约为同地区非沙漠覆盖区的二分之一,氡浓度在平面上呈现环状异常,分量铀平均值为90 ng·g^(-1),其值约为同地区非沙漠覆盖区的三分之一,分量含量在平面上呈团块状,并沿鸟山东断裂带向南延伸趋势,认为活性炭吸附测氡和分量化探组合探测技术用于沙漠覆盖区铀矿找矿工作可行且有效,氡的异常特征与非沙漠覆盖区异常特征一致,环状异常及构造断裂分布一致的氡、分量铀异常以及提取的弱信息异常、元素组合异常、元素及氡气综合异常反映鸟山沙漠覆盖区可能存在砂岩型铀矿化。

综合物化探方法在荷泗地区“交点”型铀矿找矿预测中的应用

“交点”型铀矿是华南地区最重要铀矿类型之一,寻找“交点”型铀矿对华南地区铀矿增储增产具有重要意义。为快速评价“交点”型铀矿浅部、深部铀成矿潜力,筛选最佳铀成矿有利地段,选择地面伽马能谱、土壤氡气及分量化探3种方法进行组合测量。实践表明:地面伽马能谱测量能直接获取浅表岩石中铀含量,探寻浅表异常矿化信息;土壤氡气测量、分量化探测量则能大致查明深部铀矿化信息,判别深部铀成矿潜力;三者组合能有效预测“交点”型铀矿的浅部和深部铀矿化信息,并被320矿点的方法有效性试验所证实。综合运用上述3种方法,在荷泗地区成功预测3片铀成矿有利地段,并优选Um-1号异常进行探槽揭露,发现较好铀矿化信息,取得明显的找矿应用效果。

基于辐照度分析的干式空心电抗器局部放电定位方法

干式空心电抗器局部放电的定位对于保证电能质量和节省维修成本具有重要意义。针对现有局部放电定位方法在电抗器实际应用中精度较低的问题,提出一种基于辐照度分析的干式空心电抗器局部放电定位方法。首先,根据干式空心电抗器典型结构建立三维光学仿真模型,分析不同位置放电光源在包封底部的辐照度分布特性,基于最大方差理论对辐照分布矩阵进行特征提取,构建光学仿真数据集;然后,采用交叉验证的方法对随机森林和反向传播神经网络模型进行参数调优,使用测试集评估模型的定位效果;最后,利用干式空心电抗器局部放电实验平台,对定位方法进行实测验证。结果表明,2种模型在实验数据集下的定位准确率均高于90%,相比于反向传播神经网络,随机森林的定位准确率更高且推理时间更短,验证了基于辐照度分析的放电定位方法的有效性。

基于自适应权重的多源部分域适应

作为机器学习的一个新兴领域,多源部分域适应(MSPDA)问题由于其源域自身的复杂性、领域之间的差异性以及目标域自身的无监督性,给相关研究带来了挑战,以致目前鲜有相关工作被提出.在该场景下,多个源域中的无关类样本在域适应过程中会造成较大的累积误差和负迁移.此外,现有多源域适应方法大多未考虑不同源域对目标域任务的贡献度不同.因此,提出基于自适应权重的多源部分域适应方法(AW-MSPDA).首先,构建了多样性特征提取器以有效利用源域的先验知识;同时,设计了多层次分布对齐策略从不同层面消除了分布差异,促进了正迁移;此外,为量化不同源域贡献度以及过滤源域无关类样本,利用相似性度量以及伪标签加权方式构建自适应权重;最后,通过大量实验验证了所提出AW-MSPDA算法的泛化性以及优越性.

基于双目视觉的部分遮挡行人检测算法

针对行人被障碍物部分遮挡导致的检测准确率降低问题,提出了基于多特征融合的树形路径半全局立体匹配的部分遮挡行人检测算法。使用简单线性迭代聚类(simple linear iterative clustering,SLIC)算法进行超像素分割,提升行人的轮廓信息,并使用多特征融合的树形路径半全局立体匹配算法生成深度图;对行人信息和背景信息及障碍物信息使用自适应分割算法进行分离,获取感兴趣区域;将感兴趣区域放置在行人特征明显且稳定的头肩部,进行感兴趣区域的约束;使用降维梯度直方图特征(histogram of gradient,HOG)进行特征提取并生成样本集,训练支持向量机(support vector machines,SVM)分类器,最终实现部分遮挡的行人检测。实验表明,所提算法与其他行人检测算法相比,在行人部分遮挡场景下,有着更高的行人检测准确率,证明所提算法的有效性。