基于Hotelling T2-BiLSTM的同步调相机故障诊断方法

同步调相机故障会严重影响特高压直流输电工程的稳定性。针对特高压电网中不对称电压条件下同步调相机励磁绕组匝间短路故障难以识别问题,提出基于霍特林T2-双向长短期记忆网络(Hotelling T2-BiLSTM)的同步调相机故障诊断方法。首先,从三相电压、三相电流、励磁电流、转子振动和无功功率中提取故障特征,计算Hotelling T2统计量以构建故障检测指标;然后,采用BiLSTM模型对特征进行分类,并利用逻辑回归层进行故障诊断;最后,在动模实验室中进行实验验证其有效性。实验结果表明,所提方法具有较高的故障诊断精度。

稳健Hotelling T2控制图的构建及其比较

传统Hotelling T2控制图易受离群值的影响,从而导致传统Hotelling T2控制图监控效果与实际存在偏差。针对这一现象,文章运用稳健统计的思想对传统Hotelling T2控制图进行改进,构建出Hotelling T2控制图的稳健算法,以克服离群值的影响,并对此方法进行了模拟和实证分析。模拟实验和实证分析均表明:当数据不存在离群值时,传统和稳健Hotelling T2控制图的监测效果基本保持一致;当数据中存在离群值时,运用传统Hotelling T2控制图方法得到的结果与实际有较大偏差,而运用稳健Hotelling T2控制图方法得到的结果基本不变。这说明相对于传统Hotelling T2控制图方法,稳健Hotelling T2控制图方法更能有效抵抗离群值的影响,具有良好的抗干扰性和高抗差性。

Monitoring Freeway Incident Detection Using a Hotelling T2 Control Chart

In real-life freeway transportation system, a few number of incident observation （very rare event） is available while there are large numbers of normal condition dataset. Most of researches on freeway incident detection have considered the incident detection problem as classification one. However, because of insufficiency of incident events, most of previous researches have utilized simulated incident events to develop freeway incident detection models. In order to overcome this drawback, this paper proposes a wavelet-based Hotelling 7a control chart for freeway incident detection, which integrates a wavelet transform into an abnormal detection method. Firstly, wavelet transform extracts useful features from noisy original traffic observations, leading to reduce the dimensionality of input vectors. Then, a Hotelling T2 control chart describes a decision boundary with only normal traffic observations with the selected features in the wavelet domain. Unlike the existing incident detection algorithms, which require lots of incident observations to construct incident detection models, the proposed approach can decide a decision boundary given only normal training observations. The proposed method is evaluated in comparison with California algorithm, Minnesota algorithm and conventional neural networks. The experimental results present that the proposed algorithm in this paper is a promising alternative for freeway automatic incident detections.

Self-normalization:Taming a wild population in a heavy-tailed world

The past two decades have witnessed the active development of a rich probability theory of Studentized statistics or self-normalized processes, typified by Student’s t-statistic as introduced by W. S. Gosset more than a century ago, and their applications to statistical problems in high dimensions, including feature selection and ranking, large-scale multiple testing and sparse, high dimensional signal detection. Many of these applications rely on the robustness property of Studentization/self-normalization against heavy-tailed sampling distributions. This paper gives an overview of the salient progress of self-normalized limit theory, from Student’s t-statistic to more general Studentized nonlinear statistics. Prototypical examples include Studentized one- and two-sample U-statistics. Furthermore, we go beyond independence and glimpse some very recent advances in self-normalized moderate deviations under dependence.

SOUND SOURCE LOCALIZATION OF DIGITAL HEARING AIDS USING WAVELET BASED MULTIVARIATE STATISTICAL METHOD

The letter proposed a sound source localization method of digital hearing aids using wavelet based multivariate statistics with the Generalized Cross Correlation (GCC) algorithm. Haar wavelet is used to decompose GCC sequences and extract four wavelet characteristics. And then, Hotelling T2 statistical method is used to fuse the four wavelet characteristics. The statistical value is used to judge the number of sound sources and obtain corresponding time delay estimation which is used to localize the position of sound source. The experimental results show that the proposed method has better robustness in an environment with severe noise and reverberation. Meanwhile, the complexity of al-gorithm is moderate, which is available for sound source localization of hearing aids.

Classification, Gender and Age Differences, and Seasonal Changes in Relation to Patterns of Distribution Curves for Physique and Motor Performance in Preschool-Aged Japanese Children

It is predicted that the distribution on score of children＇s motor performance show abnormal curve, because their physical fitness shows decline and gap between good and poor. The aim of present study is to type pattern of distribution curve on physique and motor performance （PMP） in preschoolers, and to investigate gender difference, and change in season and age on distribution of PMP. Seven hundred and seven preschoolers participate in measurement on 21 items of PMP. And mix-longitudinal data which made from 6 terms （2 seasons of spring and autumn for 3 years） is completed. The histograms of 10 sections are trained for each gender, term, and test item based on range of data including mean ± 3 standard deviation. By cluster analysis and multi-correspondence analysis, ＂physique ＆ jumping＂ show closely normal curve, but ＂manipulation （MP）＂ is skewed to poor and ＂running ＆ prompt （RP）＂ is skewed to good. By Hotelling＇s T2-test and Mahalanobis＇ Distance, gender difference is shown that boys＇ RP is skewed to good, and their ＂throwing＂ and ＂weight ＆ flexibility （WF）＂ are skewed to poor. About season, WF is skewed to poor and RP is skewed to good, in spring. In annual change about pattern of distribution, skewing poor in MP and good in ＂running straight ＆ prompt＂ are changed to normal curve with aging.

泥土物证XRF元素定量检验数据比对评价方法比较

目的在微量物证领域,元素分析是进行泥土物证检验的有效手段之一,X射线荧光光谱法(XRF)具有分析速度快、检验无损、可进行无标样分析等特点在泥土物证的比对检验中应用广泛。但是针对两个样品间多维泥土元素检验数据的相似性评价一直是困扰微量物证工作者的一个难题。本文旨在对比欧氏距离和Hotelling T2这两种统计学方法在评价样品检验数据差异性方面的准确性。方法采用XRF法对泥土样品进行10种元素定量分析,采用欧氏距离和Hotelling T2两种方法对检验数据进行统计处理,比较两种方法在评价两个样品间相似性时的差异。结果欧氏距离和Hotelling T2法在处理20个(190组)不同来源的泥土样品之间的两两比对时,不可区分率分别为4.21%和7.37%。结论欧氏距离和Hotelling T2两种多参数数据比对检验法均可用于泥土物证XRF定量检验数据间相似性评价,两种方法各有特点,在实际应用中要注意其使用条件和前提,才能保证统计结果的准确性。

高维数据下两个正态总体均值的逐段检验

高维数据下的均值检验是统计学检验的重要组成部分.在高维情形下,样本协方差矩阵往往是奇异矩阵,传统的均值检验方法因此失效.为解决该问题,对高维数据进行分段,使分得的每一段的维数均小于样本容量,继而运用Hotelling T 2检验依次对每一段进行检验,同时给出一种控制犯第一类错误的概率的新方法,使原假设下的检验水平稳定在事先给定的显著性水平左右.经模拟显示,该逐段检验方法比已有方法能更好地控制犯第一类错误的概率.

On the k-sample Behrens-Fisher problem for high-dimensional data

For several decades, much attention has been paid to the two-sample Behrens-Fisher (BF) problem which tests the equality of the means or mean vectors of two normal populations with unequal variance/covariance structures. Little work, however, has been done for the k-sample BF problem for high dimensional data which tests the equality of the mean vectors of several high-dimensional normal populations with unequal covariance structures. In this paper we study this challenging problem via extending the famous Scheffe’s transformation method, which reduces the k-sample BF problem to a one-sample problem. The induced one-sample problem can be easily tested by the classical Hotelling’s T 2 test when the size of the resulting sample is very large relative to its dimensionality. For high dimensional data, however, the dimensionality of the resulting sample is often very large, and even much larger than its sample size, which makes the classical Hotelling’s T 2 test not powerful or not even well defined. To overcome this difficulty, we propose and study an L 2-norm based test. The asymptotic powers of the proposed L 2-norm based test and Hotelling’s T 2 test are derived and theoretically compared. Methods for implementing the L 2-norm based test are described. Simulation studies are conducted to compare the L 2-norm based test and Hotelling’s T 2 test when the latter can be well defined, and to compare the proposed implementation methods for the L 2-norm based test otherwise. The methodologies are motivated and illustrated by a real data example.

Optimal Estimation for Power of Variance with Application to Gene-Set Testing

Detecting differential expression of genes in genom research(e.g.,2019-nCoV)is not uncommon,due to the cost only small sample is employed to estimate a large number of variances(or their inverse)of variables simultaneously.However,the commonly used approaches perform unreliable.Borrowing information across different variables or priori information of variables,shrinkage estimation approaches are proposed and some optimal shrinkage estimators are obtained in the sense of asymptotic.In this paper,we focus on the setting of small sample and a likelihood-unbiased estimator for power of variances is given under the assumption that the variances are chi-squared distribution.Simulation reports show that the likelihood-unbiased estimators for variances and their inverse perform very well.In addition,application comparison and real data analysis indicate that the proposed estimator also works well.