Cross-Dimension Attentive Feature Fusion Network for Unsupervised Time-Series Anomaly Detection

Time series anomaly detection is crucial in various industrial applications to identify unusual behaviors within the time series data.Due to the challenges associated with annotating anomaly events,time series reconstruction has become a prevalent approach for unsupervised anomaly detection.However,effectively learning representations and achieving accurate detection results remain challenging due to the intricate temporal patterns and dependencies in real-world time series.In this paper,we propose a cross-dimension attentive feature fusion network for time series anomaly detection,referred to as CAFFN.Specifically,a series and feature mixing block is introduced to learn representations in 1D space.Additionally,a fast Fourier transform is employed to convert the time series into 2D space,providing the capability for 2D feature extraction.Finally,a cross-dimension attentive feature fusion mechanism is designed that adaptively integrates features across different dimensions for anomaly detection.Experimental results on real-world time series datasets demonstrate that CAFFN performs better than other competing methods in time series anomaly detection.

Unsupervised Time Series Segmentation: A Survey on Recent Advances

Time series segmentation has attracted more interests in recent years,which aims to segment time series into different segments,each reflects a state of the monitored objects.Although there have been many surveys on time series segmentation,most of them focus more on change point detection(CPD)methods and overlook the advances in boundary detection(BD)and state detection(SD)methods.In this paper,we categorize time series segmentation methods into CPD,BD,and SD methods,with a specific focus on recent advances in BD and SD methods.Within the scope of BD and SD,we subdivide the methods based on their underlying models/techniques and focus on the milestones that have shaped the development trajectory of each category.As a conclusion,we found that:(1)Existing methods failed to provide sufficient support for online working,with only a few methods supporting online deployment;(2)Most existing methods require the specification of parameters,which hinders their ability to work adaptively;(3)Existing SD methods do not attach importance to accurate detection of boundary points in evaluation,which may lead to limitations in boundary point detection.We highlight the ability to working online and adaptively as important attributes of segmentation methods,the boundary detection accuracy as a neglected metrics for SD methods.

H_-/H_∞ fault detection filter design for interval time-varying delays switched systems

The problem of the robust fault detection filter design for time-varying delays switched systems is considered in the framework of mixed H-/H∞. Firstly, the weighted H∞ performance index is utilized as the robustness performance, and the H- index is used as the sensitivity performance for obtaining the robust fault detection filter. Then a novel multiple Lyapunov-Krasovskii function is proposed for deriving sufficient existence conditions of the robust fault detection filter based on the average dwell time technique. By introducing slack matrix variable, the coupling between the Lyapunov matrix and system matrix is removed, and the conservatism of results is reduced. Based on the robust fault detection filter, residual is generated and evaluated for detecting faults. In addition, the results of this paper are dependent on time delays,and represented in the form of linear matrix inequalities. Finally,the simulation example verifies the effectiveness of the proposed method.

Fuzzy Neural Network Model of 4-CBA Concentration for Industrial Purified Terephthalic Acid Oxidation Process

A fuzzy neural network (FNN) model is developed to predict the 4-CBA concentration of the oxidation unit in purified terephthalic acid process. Several technologies are used to deal with the process data before modeling.First,a set of preliminary input variables is selected according to prior knowledge and experience. Secondly,a method based on the maximum correlation coefficient is proposed to detect the dead time between the process variables and response variables. Finally, the fuzzy curve method is used to reduce the unimportant input variables.The simulation results based on industrial data show that the relative error range of the FNN model is narrower than that of the American Oil Company (AMOCO) model. Furthermore, the FNN model can predict the trend of the 4-CBA concentration more accurately.

Cooperative interception with fast multiple model adaptive estimation

For the case that two pursuers intercept an evasive target,the cooperative strategies and state estimation methods taken by pursuers can seriously affect the guidance accuracy for the target,which performs a bang For the case that two pursuers intercept an evasive target,the cooperative strategies and state estimation methods taken by pursuers can seriously affect the guidance accuracy for the target,which performs a bang-bang evasive maneuver with a random switching time.Combined Fast multiple model adaptive estimation(Fast MMAE)algorithm,the cooperative guidance law takes detection configuration affecting the accuracy of interception into consideration.Introduced the detection error model related to the line-of-sight(LOS)separation angle of two interceptors,an optimal cooperative guidance law solving the optimization problem is designed to modulate the LOS separation angle to reduce the estimation error and improve the interception performance.Due to the uncertainty of the target bang-bang maneuver switching time and the effective fitting of its multi-modal motion,Fast MMAE is introduced to identify its maneuver switching time and estimate the acceleration of the target to track and intercept the target accurately.The designed cooperative optimal guidance law with Fast MMAE has better estimation ability and interception performance than the traditional guidance law and estimation method via Monte Carlo simulation.

Research and application of an intelligent recloser controller installed on outdoor rod

A new type of intelligent recolser controller installed on the outdoor rod is developed, which is mainly composed of microcontroller of Intel 87C196KC 20 and CPLD devices. This controller integrates all the functions of measuring, controlling, protection, fault diagnosis, communication, remote controlled operation and self power devices with infra red remote control devices as a unit. The controller applies the distributed structure, field concentration line and intelligent technology to seal up the synthetic servomechanisms such as the microcomputer based protection and measuring devices in the second stage of the mini out door transformer substation, which are distributed on the outdoor circuit switches on the spot and formed as a whole. Therefore, this technology can transform a large number of ordinary homemade SF 6 circuit beaker and vacuum circuit breaker into intelligent circuit recloser, thus replacing the expensive imported automatic circuit recolser.

Performance of Stochastically Intermittent Sensors in Detecting a Target Traveling between Two Areas

We study the problem of detecting a target that moves between a hiding area and an operating area over multiple fixed routes. The research is carried out with one or more cookie-cutter sensors with stochastic intermission, which turn on and off stochastically governed by an on-rate and an off-rate. A cookie-cutter sensor, when it is on, can detect the target instantly once the target comes within the detection radius of the sensor. In the hiding area, the target is shielded from being detected. The residence times of the target, respectively, in the hiding area and in the operating area, are exponentially distributed and are governed by rates of transitions between the two areas. On each travel between the two areas and in each travel direction, the target selects a route randomly according to a probability distribution. Previously, we analyzed the simple case where the sensors have no intermission (i.e., they stay on all the time). In the current study, the sensors are stochastically intermittent and are synchronized (i.e., they turn on or off simultaneously). This happens when all sensors are affected by the same environmental factors. We derive asymptotic expansions for the mean time to detection when the on-rate and off-rate of the sensors are large in comparison with the rates of the target traveling between the two areas. Based on the mean time to detection, we evaluate the performance of placing the sensor(s) to monitor various travel route(s) or to scan the operating area.

Parameter-Free Search of Time-Series Discord

Time-series discord is widely used in data mining applications to characterize anomalous subsequences in time series. Compared to some other discord search algorithms, the direct search algorithm based on the recurrence plot shows the advantage of being fast and parameter free. The direct search algorithm, however, relies on quasi-periodicity in input time series, an assumption that limits the algorithm＇s applicability. In this paper, we eliminate the periodicity assumption from the direct search algorithm by proposing a reference function for subsequences and a new sampling strategy based on the reference function. These measures result in a new algorithm with improved efficiency and robustness, as evidenced by our empirical evaluation.

Space moving target detection using time domain feature

The traditional space target detection methods mainly use the spatial characteristics of the star map to detect the targets, which can not make full use of the time domain information. This paper presents a new space moving target detection method based on time domain features. We firstly construct the time spectral data of star map, then analyze the time domain features of the main objects(target, stars and the background) in star maps, finally detect the moving targets using single pulse feature of the time domain signal. The real star map target detection experimental results show that the proposed method can effectively detect the trajectory of moving targets in the star map sequence, and the detection probability achieves 99% when the false alarm rate is about 8×10^(-5), which outperforms those of compared algorithms.

Real-time Face Detection using Skin Color Model

This paper presents a new face detection approach to real-time applications, which is based on the skin color model and the morphological filtering. First the non-skin color pixels of the input image are removed based on the skin color model in the YC rC b chrominance space, from which we extract candidate human face regions. Then a mathematical morphological filter is used to remove noisy regions and fill the holes in the candidate skin color regions. We adopt the similarity between the human face features and the candidate face regions to locate the face regions in the original image. We have implemented the algorithm in our smart media system. The experiment results show that this system is effective in real-time applications.

Dispersion-Induced Waveform Distortion Detection in 42.7 Gbps CS-RZ Signals by Optical Time Domain Level Monitoring

We propose a technique for chromatic dispersion monitoring based on optical time domain level monitoring. Experimental and simulation results show that the technique is effective for the monitoring of dispersion in 42.7-Gbps CS-RZ signals for dynamic dispersion compensation.

An optimization of the relationship between crystal volume and MDAC

The volume and exposure time of nuclear radiation detectors are different in the Marine environment.This paper selects γ-rays emitted by ^(131)I,^(137)Cs and ^(208)Tl radionuclides,and uses NaI detectors of different volumes to simulate the minimum detectable activity concentration(MDAC)at different exposure time.And this paper studies the relationship between the increase multiple of crystal volume and the decrease multiple of MDAC.In this paper,based on MDAC,the existence of nuclides at different crystal volumes and different exposure times was qualitatively calculated and analyzed,which will be of guiding significance to the in situ γ spectrum measurement and long-term monitoring of seawater.

Analysis of fault detection method based on predictive filter approach

A new detection method for component faults based on predictive filters together with the fault detectability, false alarm rate, missed alarm rate and upper bound of detection time are proposed. The efficiency of the method is illustrated by a simulation example of a second-order system. It is shown that the fault detection method using predictive filters has a small delay, a low false alarm rate and a low missed alarm rate. Furthermore the filter can give accurate estimates of states even after a fault occurs. The real-time estimation provided by this method can also be used for fault tolerant control.

Detection of food source by PCR analysis of the gut contents of Aldrichina grahami （Aldrich） （Diptera： Calliphoridae） during post-feeding period

The last meal of sarcophagous maggots may be useful in identifying the species on whose flesh they have fed （the＂host＂species）. The DNA profile of the host species may indeed be detectable in the＂last meal＂. In this paper, mitochondrial DNA analysis of gut contents was used to identify the prior host of post-feeding larvae of Aldrichina grahami （Aldrich） （Diptera： Calliphoridae）. A modified logistic equation was fitted to estimate the probability of identifying the host under five different constant temperatures （16, 20, 24, 28 and 32 ℃）. Our results shows that the detected time ranged from a maximum of 24 h at 32℃ to 42 h at 16℃ and a minimum of 12 h at 32~C to 30 h at 16℃. Furthermore, the host detection time was also calculated to give the maximal time after larval hatching from the egg. These results indicate that, in criminal cases where the maggots stray from the corpse, the last meal of the larvae should not be overlooked as potentially critical evidence.

Balancing Frequencies and Fault Detection in the In-Parameter-Order Algorithm

The In-Parameter-Order （IPO） algorithm is a widely used strategy for the construction of software test suites for combinatorial testing （CT） whose goal is to reveal faults triggered by interactions among parameters. Variants of IPO have been shown to produce test suites within reasonable amounts of time that are often not much larger than the smallest test suites known. When an entire test suite is executed, all faults that arise from t-way interactions for some fixed t are surely found. However, when tests are executed one at a time, it is desirable to detect a fault as early as possible so that it can be repaired. The basic IPO strategies of horizontal and vertical growth address test suite size, but not the early detection of faults. In this paper, the growth strategies in IPO are modified to attempt to evenly distribute the values of each parameter across the tests. Together with a reordering strategy that we add, this modification to IPO improves the rate of fault detection dramatically （improved by 31% on average）. Moreover, our modifications always reduce generation time （2 times faster on average） and in some eases also reduce test suite size.

Strongyloides stercoralis prevalence and diagnostics in Vientiane, Lao People’s Democratic Republic

Background:Despite the high prevalence of strongyloidiasis in the Laotian population,Laotian hospitals still lack diagnostic capacity to appropriately diagnose Strongyloides stercoralis infections.This cross-sectional hospital-based study was conducted to assess the prevalence of Strongyloides stercoralis infection among hospitalized patients treated at Mahosot Hospital,the primary reference hospital of Lao People’s Democratic Republic(Lao PDR),and to validate feasible methods for diagnosing S.stercoralis infection at hospital’s laboratory.Methods:Between September and December 2018,stool samples of 104 inpatients were investigated for S.stercoralis infection by wet smear,Baermann technique,Koga Agar plate culture(KAPC),and real-time detection polymerase chain reaction(RTD-PCR)at the Infectious Diseases Ward of the Mahosot Hospital in Vientiane.The sensitivity,the specificity,the negative predictive value(NPV)of each diagnostic test,as well as their combination(s)was calculated using a composite reference standard(CRS).The correlation of the different test methods was assessed by chi-square or Fisher’s exact test.Cohen’s kappa coefficient was used to assess the diagnostic agreement of the different test methods.Results:The overall prevalence of S.stercoralis infections among the study population was 33.4%.The cumulative infection prevalence statistically significantly increased from the lowest age group of 40 years and below(22.4%),to the medium(40.0%)and to the oldest age group of 61 year and above(72.7%)(P=0.003).The cumulative infection prevalence of CRS was considerably higher in male(40.4%)compared to female patients(28.1%),but not statistically different(P=0.184).The diagnostic sensitivity of Baermann technique,KAPC,RTD-PCR,and the combination of Baermann technique and KAPC were 60.0,60.0,74.3,and 77.1%,respectively.Only 13 patients(37.1%)of the total 35 S.stercoralis patients diagnosed with any technique had a simultaneously positive diagnostic test with Baermann,KAPC and RTD-PCR.Conclusions:We identified Baermann technique and KAPC to be currently the most feasible and implementable standard methods for diagnosing S.stercoralis at a hospital setting such as Mahosot Hospital and provincial and district hospitals in Lao PDR and other low-and middle income countries in Southeast Asia.Trial registration:This study was approved by the National Ethics Committee for Health Research in Lao PDR(reference no.083/NECHR)and by the Ethics Committee Northwest and Central Switzerland(reference no.2018–00594).

A position-sensitive detector with lithium glass and MaPMT

A position-sensitive detector is designed for neutron detection. It uses a single continuous screen of a self-made lithium glass scintillator, rather than discrete crystal implementations, coupling with a multi-anode PMT （MaPMT）. The scintillator is fast and efficient; with a decay time of 34 ns and thermal neutron detection efficiency of around 95.8% for the 3 mm thick screen, and its light yield is around 5670 photons per neutron and 3768 photons per MeV γ rays deposition. The spatial resolution is around 1.6 mm （FWHM） with the energy resolution around 34.7% by using α （5.2 MeV） rays test.