A Power Data Anomaly Detection Model Based on Deep Learning with Adaptive Feature Fusion

With the popularisation of intelligent power,power devices have different shapes,numbers and specifications.This means that the power data has distributional variability,the model learning process cannot achieve sufficient extraction of data features,which seriously affects the accuracy and performance of anomaly detection.Therefore,this paper proposes a deep learning-based anomaly detection model for power data,which integrates a data alignment enhancement technique based on random sampling and an adaptive feature fusion method leveraging dimension reduction.Aiming at the distribution variability of power data,this paper developed a sliding window-based data adjustment method for this model,which solves the problem of high-dimensional feature noise and low-dimensional missing data.To address the problem of insufficient feature fusion,an adaptive feature fusion method based on feature dimension reduction and dictionary learning is proposed to improve the anomaly data detection accuracy of the model.In order to verify the effectiveness of the proposed method,we conducted effectiveness comparisons through elimination experiments.The experimental results show that compared with the traditional anomaly detection methods,the method proposed in this paper not only has an advantage in model accuracy,but also reduces the amount of parameter calculation of the model in the process of feature matching and improves the detection speed.

Experimental validation of a signal-based approach for structural earthquake damage detection using fractal dimension of time frequency feature

This article extends a signal-based approach formerly proposed by the authors, which utilizes the fractal dimension of time frequency feature （FDTFF） of displacements, for earthquake damage detection of moment resist frame （MRF）, and validates the approach with shaking table tests. The time frequency feature （TFF） of the relative displacement at measured story is defined as the real part of the coefficients of the analytical wavelet transform. The fractal dimension （FD） is to quantify the TFF within the fundamental frequency band using box counting method. It is verified that the FDTFFs at all stories of the linear MRF are identical with the help of static condensation method and modal superposition principle, while the FDTFFs at the stories with localized nonlinearities due to damage will be different from those at the stories without nonlinearities using the reverse-path methodology. By comparing the FDTFFs of displacements at measured stories in a structure, the damage-induced nonlinearity of the structure under strong ground motion can be detected and localized. Finally shaking table experiments on a 1：8 scale sixteen-story three-bay steel MRF with added frictional dampers, which generate local nonlinearities, are conducted to validate the approach.

Cloud detection from visual band of satellite image based on variance of fractal dimension

Cover ratio of cloud is a very important factor which affects the quality of a satellite image, therefore cloud detection from satellite images is a necessary step in assessing the image quality. The study on cloud detection from the visual band of a satellite image is developed. Firstly, we consider the differences between the cloud and ground including high grey level, good continuity of grey level, area of cloud region, and the variance of local fractal dimension (VLFD) of the cloud region. A single cloud region detection method is proposed. Secondly, by introducing a reference satellite image and by comparing the variance in the dimensions corresponding to the reference and the tested images, a method that detects multiple cloud regions and determines whether or not the cloud exists in an image is described. By using several Ikonos images, the performance of the proposed method is demonstrated.

On-line Chatter Detection Using an Improved Support Vector Machine

On-line chatter detection can avoid unstable cutting through monitoring the machining process.In order to identify chatter in a timely manner,an improved Support Vector Machine(SVM)is developed in this paper,based on extracted features.In the SVM model,the penalty factor(e)and the core parameter(g)have important influence on the classification,more than from Kernel Functions(KFs).Hence,first the classification results are conducted using different KFs.Then two methods are presented for exploring the best parameters.The chatter identification results show that the Genetic Algorithm(GA)approach is more suitable for deciding the parameters than the Grid Explore(GE)approach.

On-Line Detection of EDM Spark Locations Using Potential Difference Method

The valuation relation of potential difference with discharging time in Electrical Discharge Machining (EDM) is analyzed theoretically and tested and verified by experiments designed in this paper and the relation between potential difference and spark location is induced and analyzed, and proceed by experiments under the condition of onedimension.

Adaptive subspace detection based on two-step dimension reduction in the underwater waveguide

In the underwater waveguide,the conventional adaptive subspace detector(ASD),derived by using the generalized likelihood ratio test(GLRT)theory,suffers from a significant degradation in detection performance when the samplings of training data are deficient.This paper proposes a dimension-reduced approach to alleviate this problem.The dimension reduction includes two steps:firstly,the full array is divided into several subarrays;secondly,the test data and the training data at each subarray are transformed into the modal domain from the hydrophone domain.Then the modal-domain test data and training data at each subarray are processed to formulate the subarray statistic by using the GLRT theory.The final test statistic of the dimension-reduced ASD(DR-ASD)is obtained by summing all the subarray statistics.After the dimension reduction,the unknown parameters can be estimated more accurately so the DR-ASD achieves a better detection performance than the ASD.In order to achieve the optimal detection performance,the processing gain of the DR-ASD is deduced to choose a proper number of subarrays.Simulation experiments verify the improved detection performance of the DR-ASD compared with the ASD.

IMPROVING VOICE ACTIVITY DETECTION VIA WEIGHTING LIKELIHOOD AND DIMENSION REDUCTION

The performance of the traditional Voice Activity Detection （VAD） algorithms declines sharply in lower Signal-to-Noise Ratio （SNR） environments. In this paper, a feature weighting likelihood method is proposed for noise-robust VAD. The contribution of dynamic features to likelihood score can be increased via the method, which improves consequently the noise robustness of VAD. Divergence based dimension reduction method is proposed for saving computation, which reduces these feature dimensions with smaller divergence value at the cost of degrading the performance a little. Experimental results on Aurora Ⅱ database show that the detection performance in noise environments can remarkably be improved by the proposed method when the model trained in clean data is used to detect speech endpoints. Using weighting likelihood on the dimension-reduced features obtains comparable, even better, performance compared to original full-dimensional feature.

PRECONCENTRATION OF TRACE ZINC IN SEAWATER ON CPPI RESIN BY FIA AND ON-LINE DETECTION BY ICP-AFS

A quite new type of chelating resin Carboxymethylated Polyethylenimine-Polymethylenepolyphenylene Isocyanate(CPPI)is used for the preconcentration of Zn from high salt water such as seawater. The preconcentration is controlled through the technique of Flow Injection Analysis(FIA).The concentrated sample solution is then directly transferred to an Inductively Coupled Plasma-Atomic Fluorescence Spectrometer(ICP-AFS)for determination.The detection limit of Zn by the technique is about 0.06 ppb.

Microfluidic Analytical System with On-Line Luminol Chemiluminescence Detection Based on Annular Flow of Phase Separation Multiphase Flow

Microfluidic analytical system was developed based on annular flow of phase separation multiphase flow with a ternary water-hydrophilic/hydrophobic organic solvent solution. The analytical system was combined with on-line luminol chemiluminescence detection for catechin analysis. The water (10 mM phosphate buffer, pH 7.3)-acetonitrile-ethyl acetate mixed solution (3:8:4, volume ratio) containing 60 μM luminol and 2 mM hydrogen peroxide as a carrier was fed into the capillary tube (open-tubular fused-silica, 75 μm inner diameter, 110 cm effective length) at a flow rate of 1.0 μL·min-1. The carrier solution showed stable chemiluminescence as a baseline on the flow chart. Eight catechins were detected as negative peaks for their antioxidant potential with different detection times. The system was applied to analyze the amounts of catechin in commercially available green tea beverages.

Theory analysis and experimental research on on-line contamination detecting technology in hydraulic oil

A system of on-line contamination detecting in hydraulic oil based on silting principle is accomplished, where, metal filter membrane as detector, solenoid as active force to propel piston to blotter and gain differential pressure, step motor drives the membrane to filtrate and counter-flush, LabVIEW as detecting software platform, oil's contamination detecting indirectly by gauging differential pressure. Based on theory analysis, accomplished is relation between contamination level and differential pressure, realizing polynomial curve fitting, and calibration experiment. Field experiment is simulated in the condition of experimental laboratory, has credible precision and real-time performance, which can popularize to the field of production.

DETECTION METHOD OF SPOT WELDING BASED ON MULTI-INFORMATION FUSION AND FRACTAL

A novel detection method of support vector machine （SVM） based on fractal dimension of signals is presented. And models of SVM are made based on nugget size defects of spot welding. Classification using these trained SVM models is done to signals of spot welding. It is shown from effect of different SVM models that these models with different inputs. In detection of defects, these models with inputs including sound signal have a high percentage of accuracy, the detection accuracy of these models with inputs including voltage signal will reduce. So the SVM models based on fractal dimensions of sound are some optimal nondestructive detection ones. At last a comparison between SVM detection model and ANNS detection model is researched which indicates that SVM is a more effective measure than Artificial neural networks in detection of nugget size defects during spot welding.

Category-Based Intrusion Detection Using PCA

Existing Intrusion Detection Systems (IDS) examine all the network features to detect intrusion or misuse patterns. In feature-based intrusion detection, some selected features may found to be redundant, useless or less important than the rest. This paper proposes a category-based selection of effective parameters for intrusion detection using Principal Components Analysis (PCA). In this paper, 32 basic features from TCP/IP header, and 116 derived features from TCP dump are selected in a network traffic dataset. Attacks are categorized in four groups, Denial of Service (DoS), Remote to User attack (R2L), Remote to User attack (U2R) and Probing attack. TCP dump from DARPA 1998 dataset is used in the experiments as the selected dataset. PCA method is used to determine an optimal feature set to make the detection process faster. Experimental results show that feature reduction can improve detection rate for the category-based detection approach while maintaining the detection accuracy within an acceptable range. In this paper KNN classification method is used for the classification of the attacks. Experimental results show that feature reduction will significantly speed up the train and the testing periods for identification of the intrusion attempts.

Flow-Injection Microcolumn on-Line Preconcentration for theDetermination of Chromium(Ⅵ)

A rapid and sensitive on-line preconcentration method for spectrophotometric determination of chromium (VI) in nature water is described. Preconcentration and determination are based on (i) the quantitative and fast adsorption of chromium (VI) on the high surface area nanometer-size TiO2 (anatase) powders, which prepared by a sol-gel method from hydrolysis of TiCI4 and (ii) the quantitative and reproducible elution of Cr (VI) by 2. 0 mol. L-1 HCI. A mini-column system for preconcentration is developed, Cr(VI)on the mini-column is eluted and merged with a stream water and DPCB (1, 5-diphenylcarbazide ) as the chromogenic reagent. The Proposed system permits throughputs of 6 sample h--l (0. 001 μg mL-1 Cr(VI) ) or 20 sample h-1 (0. 1 μg mL-1Cr (VI) . The preconcentration factor is 55. The detection limit is 0. 8 ng·mL-1 Cr(VI). The reproducibility is satisfactory with a relative standard deviation of less than 3. 35% (0. Of μg'mL-1Cr (VI), n = 5).

PMOXA/PAA Brushes toward on-Line Preconcentration for BSA in Capillary Electrophoresis

In this work, a binary-mixed-brushes-coated (BBC) capillary with switchable protein adsorption/desorption properties was developed and applied for on-line preconcentration of proteins. Firstly, amine-terminated poly(2-methyl-2-oxazoline)(PMOXA-NH2) and thiolterminated poly(acrylic acid)(PAA-SH) were synthesized by using cationic ring-opening polymerization (CROP) and reversible addition fragmentation chain transfer (RAFT) polymerization, respectively. Then, the BBC capillary based on poly(2-methyl-2-oxazoline)(PMOXA) and poly(acrylic acid)(PAA) was prepared by sequentially grafting of PMOXA-NH2 and PAA-SH onto fused-silica capillary inner surface through poly(dopamine)(PDA) as an anchor. The obtained PMOXA/PAA coating formed on the capillary or capillary's raw material was characterized in terms of the thickness, surface chemical composition by using scanning electron microscope (SEM) and X-ray photoelectron spectrum (XPS). The switchable protein adsorption/desorption performance of the BBC capillary was investigated by using fluorescence microscope under di erent solutions with certain pH and ionic strength(I). The results showed that bovine serum albumin (BSA) could be adsorbed on BBC capillary at pH=5.0 (I=10^-5 mol/L), and then the adsorbed BSA could be released at pH=9.0 (I=0.1 mol/L). This switchable protein adsorption/desorption property of coated capillary was then used to preconcentrate proteins on-line for increasing the detection sensitivity of BSA in capillary electrophoresis (CE). With this method, a sensitivity enhancement factor (SEF) more than 5000 for BSA detection was obtained.

On-line coupling of derivatization with pre-concentration to determine trace levels of methotrexate

A new simple, sensitive and precise green analytical procedure using an automated packed-reactor derivatization technique coupled with on-line solid-phase enrichment （SPEn） has been developed and evaluated to determine trace levels of methotrexate （MTX）. The method was based on injection of MTX into a flowing stream of phosphate buffer （0.04 M, ptt 3.4）, carried through the packed oxidant reactor of Cerium （IV） trihydroxyhydroperoxide for oxidative cleavage of the drug into highly fluorescent product, 2,4-diaminopteridine- 6-carboxylic acid, followed by SPEn on a head of short ODS column （10mm x 4.6 mm i.d., 5 I＋tm particle size）. The flow rate was 0.25 mL/min and packed reactor temperaturc was 40 ~C. The trapped product was back-flush eluted from the ODS column to the detector by column-switching with an environmentally friendly mobile phase consisting of ethanol and phosphate buffer （0.04 M, pH 3.4） in the ratio of 5：95 （v/v）. The eluent was monitored at emission and excitation wavelengths of 460 and was linear over the concentration range of 1.25-50 360 nm, respectively. The calibration curve ng/mL with a detection limit of 0.08 ug/ml..The method was successfully applied to determine MTX in pharmaceutical formulations with mean percentage recovery ranging from 99.48 to 99.60.

Automatic System for Failure Detection in Hydro-Power Generators

This paper presents an automatic system for failure detection in hydro-power generators. The main idea of this system is to detect failure using current and voltage signals acquired without any type of internal interference in the generator operation. The detected failures could be mechanical or electrical origins, such as: problems in bearings, unwanted vibrations, partial discharges, misalignment, unbalancing, among others. It is possible because the generator acts as a transducer for mechanical problems, and they appear in current and voltage signals. This automatic system based on electric signature analysis has been installed in Itapebi Power Plant generators since 2012. Some results are presented in this paper.

Overview of Industrial Materials Detection Based on Prompt Gamma Neutron Activation Analysis Technology

Prompt gamma neutron activation analysis (PGNAA) is a non-destructive online measurement nuclear analysis method. With its unique advantages, it has been widely used in online analysis of industrial materials such as coal, cement, and minerals in recent years. However, there are many kinds of literature on PGNAA in the field of industrial materials detection, and there are still a few concluding articles. To this end, based on the principle of PGNAA online analysis, the status quo and development of the real-time online detection of industrial material components in the field are reviewed and discussed by consulting a large number of domestic and foreign PGNAA related literature and data, to facilitate the reference of relevant scientific researchers.

An Intrusion Detection Approach Based on Particle Method

In recent years,the network continues to enter people’s lives,followed by network security issues that continue to appear,causing substantial economic losses to the world.As an effective method to tackle the network security issues,intrusion detection system has been widely used and studied.In this paper,the NSL-KDD data set is used to reduce the dimension of data features,remove the features of low correlation and high interference,and improve the computational efficiency.To improve the detection rate and accuracy of intrusion detection,this paper introduces the particle method for the first time that we call it intrusion detection with particle(IDP).To illustrate the effectiveness of this method,experiments are carried out on three kinds of data-before dimension reduction,after dimension reduction and importing particle method based on dimension reduction.By comparing the results of DT,NN,SVM,K-NN,and NB,it is proved that the particle method can effectively improve the intrusion detection rate.

基于双目视觉的车前行人检测方法研究

当前的汽车安全辅助驾驶和无人驾驶汽车是图像领域的研究热点,针对汽车在启动或行驶时车前存在行人可能导致的安全问题,着重研究了基于双目视觉的车前行人检测方法。进行了双目相机的相机标定和立体标定;通过改进后半全局立体匹配算法获取深度图,确定车前行人所处位置的感兴趣区域(Region of Interest,ROI),剔除冗余的背景信息;分割并提取了图像的降维梯度直方图(Histogram of Gradients,HOG)特征信息;将特征输入到支持向量机(Support Vector Machine,SVM)分类器训练,检测并标记出车前的行人目标。实验证明,所提算法对车前场景下的动态行人可以更为有效地检测,具备更优的检率精度、时效性和鲁棒性。

基于墙角族语义尺寸链的绑架定位研究

针对目前原始自适应蒙特卡洛定位(Adaptive Monte Carlo Localization,AMCL)在相似环境下绑架检测容易出错且重定位极易失败等问题,提出基于墙角族语义尺寸链的改进AMCL算法.融合机器人多传感器信息和Gmapping算法构建二维栅格地图,基于Yolov5获取室内环境的目标检测框和类别信息,结合GrabCut算法和贝叶斯方法构建增量式语义映射地图;通过墙角的凸、凹和墙角相对于机器人的方位角对墙角进行分类,充分发掘语义映射地图中各墙角之间、墙角与室内物体之间的类别和位置关系,构建墙角族语义尺寸链和相应检索表;在定位过程中,基于墙角族语义尺寸链进行全局预定位,提出绑架检测机制进行绑架检测,在检测到绑架事件发生后,基于改进AMCL算法实现定位自恢复.最后,通过真实环境下的绑架实验验证了本文方法的有效性,实验表明,所提方法的全局定位准确率、全局定位速率、绑架检测准确率和绑架后定位准确率在相似环境下分别提升了42%、214%、88%和72%;在非相似环境下分别提升了44%、152%、12%和92%;在长走廊环境下分别提升了36%、426%、26%和68%.