A blast furnace fault monitoring algorithm with low false alarm rate:Ensemble of greedy dynamic principal component analysis-Gaussian mixture model

The large blast furnace is essential equipment in the process of iron and steel manufacturing. Due to the complex operation process and frequent fluctuations of variables, conventional monitoring methods often bring false alarms. To address the above problem, an ensemble of greedy dynamic principal component analysis-Gaussian mixture model(EGDPCA-GMM) is proposed in this paper. First, PCA-GMM is introduced to deal with the collinearity and the non-Gaussian distribution of blast furnace data.Second, in order to explain the dynamics of data, the greedy algorithm is used to determine the extended variables and their corresponding time lags, so as to avoid introducing unnecessary noise. Then the bagging ensemble is adopted to cooperate with greedy extension to eliminate the randomness brought by the greedy algorithm and further reduce the false alarm rate(FAR) of monitoring results. Finally, the algorithm is applied to the blast furnace of a large iron and steel group in South China to verify performance.Compared with the basic algorithms, the proposed method achieves lowest FAR, while keeping missed alarm rate(MAR) remain stable.

Robust Frequency Estimation Under Additive Symmetric α-Stable Gaussian Mixture Noise

Here the estimating problem of a single sinusoidal signal in the additive symmetricα-stable Gaussian(ASαSG)noise is investigated.The ASαSG noise here is expressed as the additive of a Gaussian noise and a symmetricα-stable distributed variable.As the probability density function(PDF)of the ASαSG is complicated,traditional estimators cannot provide optimum estimates.Based on the Metropolis-Hastings(M-H)sampling scheme,a robust frequency estimator is proposed for ASαSG noise.Moreover,to accelerate the convergence rate of the developed algorithm,a new criterion of reconstructing the proposal covar-iance is derived,whose main idea is updating the proposal variance using several previous samples drawn in each iteration.The approximation PDF of the ASαSG noise,which is referred to the weighted sum of a Voigt function and a Gaussian PDF,is also employed to reduce the computational complexity.The computer simulations show that the performance of our method is better than the maximum likelihood and the lp-norm estimators.

Color-texture segmentation using JSEG based on Gaussian mixture modeling

An improved approach for J-value segmentation （JSEG） is presented for unsupervised color image segmentation. Instead of color quantization algorithm, an automatic classification method based on adaptive mean shift （AMS） based clustering is used for nonparametric clustering of image data set. The clustering results are used to construct Gaussian mixture modelling （GMM） of image data for the calculation of soft J value. The region growing algorithm used in JSEG is then applied in segmenting the image based on the multiscale soft J-images. Experiments show that the synergism of JSEG and the soft classification based on AMS based clustering and GMM overcomes the limitations of JSEG successfully and is more robust.

Threshold-Based Adaptive Gaussian Mixture Model Integration(TA-GMMI)Algorithm for Mapping Snow Cover in Mountainous Terrain

Snow cover is an important parameter in the fields of computer modeling,engineering technology and energy development.With the extensive growth of novel hardware and software compositions creating smart,cyber physical systems’(CPS)efficient end-to-end workflows.In order to provide accurate snow detection results for the CPS’s terminal,this paper proposed a snow cover detection algorithm based on the unsupervised Gaussian mixture model(GMM)for the FY-4A satellite data.At present,most snow cover detection algorithms mainly utilize the characteristics of the optical spectrum,which is based on the normalized difference snow index(NDSI)with thresholds in different wavebands.These algorithms require a large amount of manually labeled data for statistical analysis to obtain the appropriate thresholds for the study area.Consideration must be given to both the high and low elevations in the study area.It is difficult to extract all snow by a fixed threshold in mountainous and rugged terrains.In this research,we avoid relying on a manual analysis for different elevations.Therefore,an algorithm based on the GMM is proposed,integrating the threshold-based algorithm and the GMM.First,the threshold-based algorithm with transferred thresholds from other satellites’analysis results are used to coarsely classify the surface objects.These results are then used to initialize the parameters of the GMM.Finally,the parameters of that model are updated by an expectation-maximum(EM)iteration algorithm,and the final results are outputted when the iterative conditions end.The results show that this algorithm can adjust itself to mountainous terrain with different elevations,and exhibits a better performance than the threshold-based algorithm.Compared with orbit satellites’snow products,the accuracy of the algorithm used for FY-4A is improved by nearly 2%,and the snow detection rate is increased by nearly 6%.Moreover,compared with microwave sensors’snow products,the accuracy is increased by nearly 3%.The validation results show that the proposed algorithm can be adapted to a complex terrain environment in mountainous areas and exhibits good performance under a transferred threshold without manually assigned labels.

Soft Image Segmentation Based on the Mixture of Gaussians and the Phase-Transition Theory

In this paper, we propose a new soft multi-phase segmentation model where it is assumed that the pixel intensities are distributed as a Gaussian mixture. The model is formulated as a minimization problem through the use of the maximum likelihood estimator and phase-transition theory. The mixture coefficients, which are estimated using a spatially varying mean and variance procedure, are used for image segmentation. The experimental results indicate the effectiveness of the method.

Automated segmentation of intraretinal cystoid macular edema based on Gaussian mixture model

We introduce a method based on Gaussian mixture model(GMM)clustering and level-set to automatically detect intraretina fluid on diabetic retinopathy(DR)from spectral domain optical coherence tomography(SD-OCT)images in this paper.First,each B-scan is segmented using GMM clustering.The original chustering results are refined using location and thickness infor-mation.Then,the spatial information among every consecutive five B-scans is used to search potential fluid.Finally,the improved level-set method is used to obtain the accurate boundaries.The high sensitivity and accuracy demonstrated here show its potential for detection of fluid.

基于模态理论和改进GMM的声发射源识别研究

基于模态声发射信号理论,提出了一种利用声学对数倒谱统计参数作为声发射信号特征参数的分析与提取方法。从声发射信号多模态特性出发,提出了一个基于改进高斯混合模型的声发射源信号识别系统。理论分析和实验结果表明,该方法能准确地判断声发射信号源,不仅能够应用于突发型声发射信号的识别,而且可以应用于连续型声发射信号的识别。

基于MFCC和GMM的瓷砖空鼓率识别系统及方法

针对瓷砖因内部空鼓而引起的松动、脱落等质量问题或其他安全隐患问题,研制了一套用于瓷砖空鼓率识别的试验系统。该系统采用梅尔倒谱系数(MFCC)法提取瓷砖敲击声的特征参数,再用高斯混合模型(GMM)法对MFCC特征参数进行分类和识别。试验结果表明,采用MFCC和GMM相结合的方法,可以对瓷砖空鼓情况进行有效识别,该方法具有良好的应用前景。

基于GMM的流体旋转设备运行可靠性在线评价方法

针对流体旋转设备运行工况多变且难以区分,导致运行可靠性评价准确率低的问题,提出了一种基于高斯混合模型(GMM)的流体旋转设备运行可靠性在线性评价方法。首先,根据设备历史运行数据,基于快速搜索和发现密度峰值的聚类算法(DPC),进行工况划分,构建不同工况条件下的基于GMM的运行可靠性基准模型;其次,使用XGBoost算法对设备实时运行状态进行工况识别,约减冗余指标,构建设备运行可靠性的评价指标体系;然后,计算度量评价指标与对应工况下基准模型指标的偏离程度,以马氏距离作为度量标准,进一步计算得到设备运行可靠性评价指数;最后,以矿用离心机设备为例,进行了多工况下的运行可靠性实例分析和模型验证。研究结果表明,该方法能够在线实时反应设备当前的运行可靠性水平,当离心机设备运行可靠性低于0.857时,认为设备进入劣化状态,且评价准确率达到98%以上。

基于GMM的湿筛混凝土轴拉损伤演化机制研究

为研究不同加载阶段下的二级配湿筛混凝土开裂模式与损伤演化过程,将声发射技术(AE)与高斯混合模型(GMM)进行结合作为损伤识别手段,以3种加载速率(1×10^(-6)、5×10^(-6)、25×10^(-6)s^(-1))作为试验变量,对二级配湿筛混凝土开展单轴拉伸损伤时空演化机制试验研究。结果表明:随着加载速率增大,湿筛混凝土试件内部裂缝开展更加密集,并且裂缝种类随机性更高;利用GMM法对声发射数据进行处理分类结果显示,拉伸裂缝为试验加载过程的主要开裂模式,加载速率升高会导致剪切裂缝占比增大;随着加载速率增大,拉伸裂缝频率分布明显扩大,而剪切裂缝与混合裂缝频率分布基本不变;随着加载进行,拉伸裂缝与剪切裂缝概率密度区域均向AF轴趋近;GMM法所得裂缝开裂模式有拉伸裂缝、剪切裂缝与混合裂缝3种类别,并且随着加载进行,混合断裂区所处位置也会发生变化;相较于常规裂缝模式分类方法,GMM法提供了更好的裂缝分类近似值分析,对裂缝开裂模式表述更加可靠。

Improved pruning algorithm for Gaussian mixture probability hypothesis density filter

With the increment of the number of Gaussian components, the computation cost increases in the Gaussian mixture probability hypothesis density(GM-PHD) filter. Based on the theory of Chen et al, we propose an improved pruning algorithm for the GM-PHD filter, which utilizes not only the Gaussian components’ means and covariance, but their weights as a new criterion to improve the estimate accuracy of the conventional pruning algorithm for tracking very closely proximity targets. Moreover, it solves the end-less while-loop problem without the need of a second merging step. Simulation results show that this improved algorithm is easier to implement and more robust than the formal ones.

An efficiency algorithm on Gaussian mixture UKF for BDS/INS navigation system

To further improve the performance of UKF(Unscented Kalman Filter) algorithm used in BDS/SINS(BeiDou Navigation Satellite System/Strap down Inertial Navigation System), an improved GM-UKF(Gaussian Mixture Unscented Kalman Filter) considering non-Gaussian distribution is discussed in this paper. This new algorithm using SVD(Singular Value Decomposition) is proposed to alternative covariance square root calculation in UKF sigma point production. And to end the rapidly increasing number of Gaussian distributions, PDF(Probability Density Function) re-approximation is conducted. In principle this efficiency algorithm proposed here can achieve higher computational speed compared with traditional GM-UKF. And simulation experiment result show that, compared with UKF and GM-UKF algorithm, new algorithm implemented in BDS/SINS tightly integrated navigation system is suitable for handling nonlinear/non-Gaussian integrated navigation position calculation, for its lower computational complexity with high accuracy.

Adaptive learning algorithm based on mixture Gaussian background

The key problem of the adaptive mixture background model is that the parameters can adaptively change according to the input data. To address the problem, a new method is proposed. Firstly, the recursive equations are inferred based on the maximum likelihood rule. Secondly, the forgetting factor and learning rate factor are redefined, and their still more general formulations are obtained by analyzing their practical functions. Lastly, the convergence of the proposed algorithm is proved to enable the estimation converge to a local maximum of the data likelihood function according to the stochastic approximation theory. The experiments show that the proposed learning algorithm excels the formers both in converging rate and accuracy.

Application of a Novel Method for Machine Performance Degradation Assessment Based on Gaussian Mixture Model and Logistic Regression

The currently prevalent machine performance degradation assessment techniques involve estimating a machine＇s current condition based upon the recognition of indications of failure features,which entail complete data collected in different conditions.However,failure data are always hard to acquire,thus making those techniques hard to be applied.In this paper,a novel method which does not need failure history data is introduced.Wavelet packet decomposition（WPD） is used to extract features from raw signals,principal component analysis（PCA） is utilized to reduce feature dimensions,and Gaussian mixture model（GMM） is then applied to approximate the feature space distributions.Single-channel confidence value（SCV） is calculated by the overlap between GMM of the monitoring condition and that of the normal condition,which can indicate the performance of single-channel.Furthermore,multi-channel confidence value（MCV）,which can be deemed as the overall performance index of multi-channel,is calculated via logistic regression（LR） and that the task of decision-level sensor fusion is also completed.Both SCV and MCV can serve as the basis on which proactive maintenance measures can be taken,thus preventing machine breakdown.The method has been adopted to assess the performance of the turbine of a centrifugal compressor in a factory of Petro-China,and the result shows that it can effectively complete this task.The proposed method has engineering significance for machine performance degradation assessment.

基于DTW-GMM的光纤传感系统声纹识别方法

为了满足易燃易爆环境的声纹识别需求,设计了直线型萨格奈克干涉光纤声音传感系统,利用维纳滤波算法对语音数据进行了降噪,通过三电平削波法获取了基音周期特征,采用动态时间规整算法筛选了说话人样本,并提取了梅尔频率倒谱系数特征,运用高斯混合模型-期望最大化算法开展了声纹识别实验研究,同时探究了光纤声音传感系统的频率响应特性与声纹特征,研究了采集语音幅值对声纹识别结果的影响。实验结果表明,系统可实现300~3500 Hz频率段的声音信号感知,声音幅值从0.9 V降至0.15 V时最大与次大对数似然值之差由35.5降至10.9,识别结果从成功变为失败。重复性实验表明,在10 km的传感光纤上,距声源2 m位置处,传感系统可对400段时长为3~5 s之间的文本无关语音段实现准确检测,且综合识别准确率为94.75%。本系统有望为易燃易爆环境中的设备故障、应急救援、渗漏监测等领域提供声纹识别的解决方案。

RBMDO Using Gaussian Mixture Model-Based Second-Order Mean-Value Saddlepoint Approximation

Actual engineering systems will be inevitably affected by uncertain factors.Thus,the Reliability-Based Multidisciplinary Design Optimization(RBMDO)has become a hotspot for recent research and application in complex engineering system design.The Second-Order/First-Order Mean-Value Saddlepoint Approximate(SOMVSA/-FOMVSA)are two popular reliability analysis strategies that are widely used in RBMDO.However,the SOMVSA method can only be used efficiently when the distribution of input variables is Gaussian distribution,which significantly limits its application.In this study,the Gaussian Mixture Model-based Second-Order Mean-Value Saddlepoint Approximation(GMM-SOMVSA)is introduced to tackle above problem.It is integrated with the Collaborative Optimization(CO)method to solve RBMDO problems.Furthermore,the formula and procedure of RBMDO using GMM-SOMVSA-Based CO(GMM-SOMVSA-CO)are proposed.Finally,an engineering example is given to show the application of the GMM-SOMVSA-CO method.

Research on Initialization on EM Algorithm Based on Gaussian Mixture Model

The EM algorithm is a very popular maximum likelihood estimation method, the iterative algorithm for solving the maximum likelihood estimator when the observation data is the incomplete data, but also is very effective algorithm to estimate the finite mixture model parameters. However, EM algorithm can not guarantee to find the global optimal solution, and often easy to fall into local optimal solution, so it is sensitive to the determination of initial value to iteration. Traditional EM algorithm select the initial value at random, we propose an improved method of selection of initial value. First, we use the k-nearest-neighbor method to delete outliers. Second, use the k-means to initialize the EM algorithm. Compare this method with the original random initial value method, numerical experiments show that the parameter estimation effect of the initialization of the EM algorithm is significantly better than the effect of the original EM algorithm.

Multimodal process monitoring based on transition-constrained Gaussian mixture model

Reliable process monitoring is important for ensuring process safety and product quality.A production process is generally characterized bymultiple operation modes,and monitoring thesemultimodal processes is challenging.Most multimodal monitoring methods rely on the assumption that the modes are independent of each other,which may not be appropriate for practical application.This study proposes a transition-constrained Gaussian mixture model method for efficient multimodal process monitoring.This technique can reduce falsely and frequently occurring mode transitions by considering the time series information in the mode identification of historical and online data.This process enables the identified modes to reflect the stability of actual working conditions,improve mode identification accuracy,and enhance monitoring reliability in cases of mode overlap.Case studies on a numerical simulation example and simulation of the penicillin fermentation process are provided to verify the effectiveness of the proposed approach inmultimodal process monitoring with mode overlap.

Automatic Delineation of Lung Parenchyma Based on Multilevel Thresholding and Gaussian Mixture Modelling

Delineation of the lung parenchyma in the thoracic Computed Tomography(CT)is an important processing step for most of the pulmonary image analysis such as lung volume extraction,lung nodule detection and pulmonary vessel segmentation.An automatic method for accurate delineation of lung parenchyma in thoracic Computed Tomography images is presented in this paper.The proposed method involves a segmentation phase followed by a lung boundary correction technique.The tissues in the thoracic Computed Tomography can be represented by a number of Gaussians.We propose a histogram utilized Adaptive Multilevel Thresholding(AMT)for estimating the total number of Gaussians and their initial parameters.The parameters of Gaussian components are updated by Expectation Maximization(EM)algorithm.The segmented lung parenchyma from the Gaussian Mixture model(GMM)undergoes an Adaptive Morphological Filtering(AMF)to reduce the boundary errors.The proposed method has been tested on 70 diseased and 119 normal lung images from 28 cases obtained from Lung Image Database Consortium(LIDC).The performance of the proposed system has been validated.

IGMM结合区间统计的机械故障预警方法研究

针对机械工况恶劣、结构复杂,单特征门限报警的故障预警方法对其预警常出现误、漏报警事件的现状,提出一种无限高斯混合模型(IGMM,Infinite Gaussian Mixture Model)结合区间统计的机械故障预警方法。首先,将机械振动信号映射为高维特征空间,对其所在空间进行区间划分。然后,利用IGMM估计出机械健康状态下高维特征空间在各区间频数的分布;利用累计计数方法统计出机械在实时状态下高维特征空间在各区间频数的分布。最后,对以上两个频数分布计算距离并将其与自学习得出的预警阈值作比较,实现故障预警。验证结果表明,提出方法的预警准确率较高且时效性较好。