Normal State Data Research Articles

Research on Fault Monitoring Technology of Distribution Network Based on Fuzzy Association Rules Mining

In order to monitor the operation status of distribution network in real time and ensure the security and stability of distribution internet of things, a fault monitoring method based on fuzzy association rule mining was proposed. Fuzzy association rules mining was used to build the normal state parameter model of distribution network. The degree of interest index replaced the traditional index, aiming to improve the association degree between the rule parameters and form an effective expert knowledge base. According to the rules in the expert knowledge base, the normal state data of distribution network were obtained by fuzzy reasoning. Further, the state similarity function between the monitoring states and the normal states was used as the state judgement index. When the similarity is lower than the threshold, the fault alert should be triggered. Taking the single-phase ground fault data of distribution network simulated by MATLAB Simulink as an example, it was verified that the proposed method can effectively monitor the single-phase ground fault.

Online detection of bearing incipient fault with semi-supervised architecture and deep feature representation

Although researchers have made substantial progress in bearing fault detection and diagnosis recently, incipient fault detection, especially online detection, is still at an initial stage. Generally speaking, online detection of incipient faults is still subject to the following challenges: (1) improving discriminative ability of incipient fault features; (2) adaptive recognition of the distribution inconsistency that exists in online sequential data; (3) achieving automatic detections with avoiding manual adjustment of detection criterion; and (4) reducing false alarm rate. To address these challenges, this paper presents a new approach for bearing incipient fault online detection using semi-supervised architecture and deep feature representation. This approach is simple and effective. First, we extract deep features using stacked denoising auto-encoder from the target bearing's normal state data and an auxiliary bearing's fault state data. Second, we introduce safe semi-supervised support vector machine (S4VM), a kind of semi-supervised classifier, to identify the sequentially arrived data of the target bearing as normal or anomalous. To update the classifier effectively, we use the principal curve to generate synthetic fault data for keeping data classes balanced during online condition monitoring. Finally, we propose a new fault alarm criterion based on S4VM generalization error upper bound to adaptively recognize the occurrence of an incipient fault. The experimental results on three datasets (IEEE PHM Challenge 2012, IMS and XJTU-SY) demonstrate the effectiveness and high reliability of the proposed approach.

Robust Detection of Bearing Early Fault Based on Deep Transfer Learning

In recent years, machine learning techniques have been proven to be a promising tool for early fault detection of rolling bearings. In many actual applications, however, bearing whole-life data are not easy to be historically accumulated, while insufficient data may result in training a detection model that is not good enough. If utilizing the available data under different working conditions to facilitate model training, the data distribution of different bearings are usually quite different, which does not meet the precondition of i n d e p e n d e n t a n d i d e n t i c a l d i s t r i b u t i o n ( i . i . d ) and tends to cause performance reduction. In addition, disturbed by the unstable noise under complex conditions, most of the current detection methods are inclined to raise false alarms, so that the reliability of detection results needs to be improved. To solve these problems, a robust detection method for bearings early fault is proposed based on deep transfer learning. The method includes offline stage and online stage. In the offline stage, by introducing a deep auto-encoder network with domain adaptation, the distribution inconsistency of normal state data among different bearings can be weakened, then the common feature representation of the normal state is obtained. With the extracted common features, a new state assessment method based on the robust deep auto-encoder network is proposed to evaluate the boundary between normal state and early fault state in the low-rank feature space. By training a support vector machine classifier, the detection model is established. In the online stage, along with the data batch arriving sequentially, the features of target bearing are extracted using the common representation learnt in the offline stage, and online detection is conducted by feeding them into the SVM model. Experimental results on IEEE PHM Challenge 2012 bearing dataset and XJTU-SY dataset show that the proposed approach outperforms several state-of-the-art detection methods in terms of detection accuracy and false alarm rate.

Wind Turbine Fault Diagnosis and Predictive Maintenance Through Statistical Process Control and Machine Learning

This study applies statistical process control and machine learning techniques to diagnose wind turbine faults and predict maintenance needs by analyzing 2.8 million sensor data collected from 31 wind turbines from 2015 to 2017 in Taiwan. Unlike previous studies that only relied on historical wind turbine data, this study analyzed the sensor data with practitioners’ insight by incorporating maintenance check list items into the data mining processes. We used Pareto analyses, scatter plots, and the cause and effect diagram to cluster and classify the failure types of wind turbines. In addition, control charts were used to establish a monitoring mechanism to track whether operation data are deviated from the controls (i.e., standard deviations) as a mean to detect wind turbine abnormalities. While statistical process control was applied to fault diagnosis, machine learning algorithms were used to predict maintenance needs of wind turbines. First, the density-based spatial clustering of applications with noise algorithm was used to classify abnormal-state wind turbine data from normal-state data. Then, random forest and decision tree algorithms were employed to construct the predictive models for wind turbine anomalies and tested with K-fold cross-validation. The results indicate a high level of accuracy: 92.68% for the decision tree model, and 91.98% for the random forest model. The study demonstrates that, by data mining and modeling, the failures of wind turbines can be detected, and the maintenance needs of parts can be predicted. Model results may provide technicians early warnings, improve equipment efficient, and decrease system downtime of wind turbine operation.

Data-Driven Incipient Fault Prediction for Non-Stationary and Non-Linear Rotating Systems: Methodology, Model Construction and Application

Many researches have been carried out on incipient fault prediction technology for key machine components (such as bearings) based on historical and real-time condition monitoring data. However, there is still lack of well-understood systematic methodologies for detecting incipient fault for rotating machines. Based on machine learning technology, this paper studies an incipient fault prediction model applying with wavelet packet decomposition and dynamic kernel principal component analysis (WPD-DKPCA) to meet the needs of engineering applications. The incipient fault prediction WPD-DKPCA model, which does not require knowledge on equipment structure and failure mechanisms, only requires normal state data of the machine, and incipient fault prediction can be achieved through self-learning. Run-to-failure experimental data and engineering case data have been used to verify the constructed model, and the verification results show that the constructed model can reliably and accurately detect an incipient bearing fault. Comparisons of fault prediction effects prove that using T2 statistic monitoring can detect upcoming faults of machines much earlier than Kurtosis and Root Mean Square (RMS).

Study of the Superconducting to Normal Transition

A model, based on classical mechanics and thermodynamics and valid for \textit{all} superconductors, is devised to investigate the properties of the \textit{current-driven}, superconducting to normal transition. This process is shown to be reversible. An original derivation of the BCS variational procedure is given. Two different critical temperatures are introduced. The temperature dependence of the critical current is worked out and found to agree with observation. The peculiar transport properties of high-$T_c$ compounds in the normal state and old magnetoelastic data are also interpreted within this framework. A novel experiment is proposed to check the relevance of this analysis to high-$T_c$ superconductivity.

Study of transport and magnetic properties of electron-doped superconducting cuprates Eu1.85Ce0.15Cu1-yZnyO4+α-δ

Transport and magnetic properties of electron-doped superconducting cuprates Eu1.85Ce0.15Cu1-yZnyO4+α-δ (ECCZO) with y = 0, 0.01, 0.02 and 0.05 have been studied from electrical resistivity and magnetic susceptibility measurements, in order to elucidate the effect of partially Zn-substitution for Cu to its superconductivity and localization of electron in normal state. It has been found that the onset of the critical temperature (Tc) from electrical resistivity and magnetic susceptibility measurements was observed at 13 K for y = 0 and δ values of 0.095. With increasing Zn concentration, the onset of Tc disappeared, indicating the destruction of superconductivity properties by Zn. Moreover, 0.01 of Zn substitution for Cu decreased radius of localization of electron (r) in normal state at low temperatures that may related to the evident of stripe-pinning effects by Zn in electron-doped system of ECCZO. Analysis of magnetic susceptibility data in normal state found a Zn-induced local moments in electron-doped high-Tc superconducting cuprates of ECCZO as observed in hole-doped cuprates.

Self-Adaptive Fault Detection and Performance Assessment of a Rotary Actuator Based on Residual Analysis

A rotary actuator that employs hydraulic oil as the power source has a direct rotary structure. It is an important structure and has been widely utilized in aircrafts and ships because of its advantages, including large torque/quality ratio, simple compact structure, and fast dynamic response. Huge damage may be caused when a rotary actuator breaks down during operation. However, only a few studies have focused on fault detection and performance assessment for rotary actuators. In this study, a method that detects the fault in and assesses the performance of the rotary actuator based on residual analysis is proposed. The data in normal state are utilized to build an observer with two radial basis function (RBF) neural networks. One RBF neural network is employed to estimate the expected output required to generate the residuals. The self-adaptive thresholds are obtained through the other RBF neural network. The residual is then inputted into the self-organizing mapping neural network trained by the residual values in normal state to normalize the performance of the rotary actuator into confidences values between 0 and 1. Finally, the detection and assessment of two typical faults of the rotary actuator are simulated. Results verify the efficiency of the proposed method.

Study of Steering Wheel Movement under Fatigue Driving and Drunk Driving Based on Sample Entropy

In order to distinguish the fatigue, drunk and normal driving effectively, sample entropy is introduced in this paper. An experiment was designed based on a driving simulator. 13 drivers were recruited to participate in the experiment. The steering wheel movement data in fatigue, drunk and normal state were collected. Sample entropy was used to analyze the characteristic of steering wheel movement under different radius curves in three driving states. The results show that the sample entropy in normal state is higher than that in fatigue state, while it is lower than that in drunk state. This indicates that sample entropy is feasible for distinguishing fatigue, drunk and normal driving.

Two-band conductivity of a FeSeTe film by reflectance measurements in the terahertz and infrared range

We report an infrared spectroscopy study of a 200 nm thick FeSeTe film grown on LaAlO3 with Tc = 13.7 K. We analyze the 20 K normal state absolute reflectance RN measured over a broad infrared range and the reflectance ratio RS/RN, RS being the superconducting state reflectance, measured at 6 K in the terahertz range down to 12 cm−1. We show that the normal state model conductivity is given by two Drude components, one of which much broader and intense than the other. In the superconducting state, we find that a gap Δ = 37 ± 3 cm−1 opens up in the narrow Drude band only, while the broad Drude band results to be ungapped, at least in the explored spectral range. Our results show that only a two-band model can coherently describe both normal and superconducting state data.

Use Subsampling to Solve Imbalanced Dataset Problem for Automatic Incident Detection Algorithm

Considering the fact that the amount of traffic incident data is rare compared to the large amount of normal traffic state data in the real word, we proposed an Automatic Incident Detection (AID) algorithm based on subsampling method. First, an improved subsampling method based on Edited Nearest Neighbor Rule (ENN) algorithm was used to reconstruct the training set to get a balanced dataset. Then, the Support Vector Machine (SVM) was adopted as a classifier to detect traffic incidents. The real traffic data collected from the I-880 freeway in American was used to build the model and test the performance of the proposed AID algorithm. In addition, we made a comparison of the detection performances between the AID algorithm obtained by the original training set and the one by the relatively balanced training set. The experimental results show that the proposed AID algorithm based on subsampling is suitable for imbalanced dataset and can obtain a better detection performance.

Superconducting and normal-state properties ofAPd2As2(A = Ca, Sr, Ba) single crystals

The synthesis and crystallography, magnetic susceptibility chi, magnetization M, specific heat Cp, in-plane electrical resistivity rho and in-plane magnetic penetration depth measurements are reported for single crystals of APd2As2 (A = Ca, Sr, Ba) versus temperature T and magnetic field H. The crystals were grown using PdAs self-flux. CaPd2As2 and SrPd2As2 crystallize in a collapsed body-centered tetragonal ThCr2Si2-type structure (I4/mmm), whereas BaPd2As2 crystallizes in the primitive tetragonal CeMg2Si2-type structure (P4/mmm), in agreement with literature data. The rho(T) data exhibit metallic behavior for all three compounds. Bulk superconductivity is reported for CaPd2As2 and SrPd2As2 below Tc = 1.27 and 0.92 K, respectively, whereas only a trace of superconductivity is found in BaPd2As2. No other phase transitions were observed. The chi(T) and M(H) data reveal anisotropic diamagnetism in the normal state, with chi_c > chi_ab for CaPd2As2 and BaPd2As2, and chi_c < chi_ab for SrPd2As2. The normal and superconducting state data indicate that CaPd2As2 and SrPd2As2 are conventional type-II nodeless s-wave electron-phonon superconductors. The electronic superconducting-state heat capacity data for CaPd2As2, which has an extremely sharp heat capacity jump at Tc, are analyzed using our recent elaboration of the alpha-model of the BCS theory of superconductivity, which indicates that the s-wave gap in this compound is anisotropic in momentum space.

은닉 마르코프 모델을 이용한 속도 변화가 있는 회전 기계의 상태 진단 기법

In industry, various rotating machinery such as pumps, gas turbines, compressors, electric motors, generators are being used as an important facility. Due to the industrial development, they make high performance(high-speed, high-pressure). As a result, we need more intelligent and reliable machine condition diagnosis techniques. Diagnosis technique using hidden Markov-model is proposed for an accurate and predictable condition diagnosis of various rotating machines and also has overcame the speed limitation of time/frequency method by using compensation of the rotational speed of rotor. In addition, existing artificial intelligence method needs defect state data for fault detection. hidden Markov model can overcome this limitation by using normal state data alone to detect fault of rotational machinery. Vibration analysis of step-up gearbox for wind turbine was applied to the study to ensure the robustness of diagnostic performance about compensation of the rotational speed. To assure the performance of normal state alone method, hidden Markov model was applied to experimental torque measuring gearbox in this study.

THE LIGHT CURVE OF HERCULES X-1 AS OBSERVED BY THEROSSI X-RAY TIMING EXPLORER

Analysis of the light curve of Hercules X-1 using the full set of archival observations of Hercules X-1 by the Rossi X-Ray Timing Explorer/Proportional Counter Array (RXTE/PCA) is reported. The observations cover time periods that Her X-1 is in main high, short high, and low states, and an anomalous low state (ALS). They include over 1.4 Ms of net exposure time. We present 35 day and orbital phase folded light curves of the count rates and softness ratios, showing the range of behaviors of Her X-1 with the high sensitivity of the RXTE/PCA. New phenomena are uncovered and previous phenomena are seen in greater detail. For both main high and short high states, the fraction of time in dips is found to be a function of orbital phase and of 35 day phase. It increases steadily with orbital phase past orbital phase 0.3 and is higher at the start and end of both main high and short high states. It is higher for short high state (62%) than for main high state (28%). The normal low state data and ALS data are compared: the low state count rate is ~twice as high as for ALS data. The 2-4 keV to 9-20 keV softness ratio changes smoothly with orbital phase for low states and ALSs, and is indistinguishable between the two, yet very different than for the high states. This supports models for which the cause of the ALS is changed disk geometry that prevents a direct line of sight from neutron star to observer at all 35 day phases.

Preformed Excitonic Liquid Route to a Charge Density Wave in2H−TaSe2

Recent experiments on 2H-TaSe(2) contradict the long-held view of the charge density wave arising from a nested band structure. An intrinsically strong coupling view, involving a charge density wave state arising as a Bose condensation of preformed excitons emerges as an attractive, albeit scantily investigated alternative. Using the local density approximation plus multiorbital dynamic mean field theory, we show that this scenario agrees with a variety of normal state data for 2H-TaSe(2). Based thereupon, the ordered states in a subset of dichalcogenides should be viewed as instabilities of a correlated, preformed excitonic liquid.

Study on Dynamic Machining Performance of Machine Tool Based on BP Network

It is analyzed on the relationship of the vibration and the machining performance of machine tool in this paper, and estimate the dynamics machining performance of machine tool on line based on BP (Back Propagation) network. Clustering is used to process data in order to reduce the number of data. Calculate speed of estimating system is increased by this method of data processing and satisfied the need of estimating on line. The 718 CNC is simulated with 25 groups of faulty state data and 25 groups of normal state data, and the simulation results show that the system can accurately estimate the dynamic machining performance of machine tool. The estimated method is very important to maintenance for machine tool.

Normal-state correlated electronic structure of iron pnictides from first principles

Discovery of high-Tc superconductivity (HTSC) in the Fe-based pnictides [1] is the latest among a host of other, ill-understood phenomena in d-band oxides. HTSC in Fe-pnictides emerges upon doping a bad metal with spin density wave (SDW) order at q = (π,0). Preliminary experiments indicate [2, 3] unconventional SC. Existent normal state data indicate a “bad metal” without Landau Fermi Liquid (FL) quasiparticles at low energy [1]. These observations in Fe-pnictides are reminiscent of cuprate SC. The small carrier density (giving rise to carrier pockets), along with Uemura scaling from � -SR [4] similar to hole-doped cuprates strongly suggests a SC closer to the Bose condensed, rather than a BCS (ξ ≃ 1000a) limit.

Optical response of high-Tccuprates: Possible role of scattering rate saturation and in-plane anisotropy

We present a generalized Drude analysis of the in-plane optical conductivity ${\ensuremath{\sigma}}_{ab}(T,\ensuremath{\omega})$ in cuprates taking into account the effects of in-plane anisotropy. A simple ansatz for the scattering rate $\ensuremath{\Gamma}(T,\ensuremath{\omega})$, that includes anisotropy, a quadratic frequency dependence, and saturation at the Mott-Ioffe-Regel limit, is able to reproduce recent normal state data on an optimally doped cuprate over a wide frequency range. We highlight the potential importance of including anisotropy in the full expression for ${\ensuremath{\sigma}}_{ab}(T,\ensuremath{\omega})$ and challenge previous determinations of $\ensuremath{\Gamma}(\ensuremath{\omega})$ in which anisotropy was neglected and $\ensuremath{\Gamma}(\ensuremath{\omega})$ was indicated to be strictly linear in frequency over a wide frequency range. Possible implications of our findings for understanding thermodynamic properties and self-energy effects in high-${T}_{c}$ cuprates will also be discussed.

Integration of Principal Component Analysis and Fuzzy Logic Systems for Comprehensive Process Fault Detection and Diagnosis

A fault diagnosis system is developed by integrating principal component analysis (PCA) with fuzzy logic knowledge-based (FLKB) systems. A PCA model is created using normal state data. Then it is used to project normal and faulty data during the training stage. It evaluates the process variables values and their correlation, allowing fast and reliable fault detection. Once detection is performed, a FLKB system is used to evaluate the contributions of each variable to changes in the process, finding the root causes of the abnormal event detected. A simple methodology to automatically extract compact process information is presented. Then, an optimization algorithm is implemented to improve the isolation performance. The methodology is demonstrated in an academic case study and in the Tennessee Eastman process benchmark.

Heat transport inBi2+xSr2−xCuO6+δ: Departure from the Wiedemann-Franz law in the vicinity of the metal-insulator transition

We present a study of heat transport in the cuprate superconductor ${\mathrm{Bi}}_{2+x}{\mathrm{Sr}}_{2\ensuremath{-}x}{\mathrm{CuO}}_{6+\ensuremath{\delta}}$ at sub-Kelvin temperatures and in magnetic fields as high as 25 T. In several samples with different doping levels close to optimal, the linear-temperature term of thermal conductivity was measured both at zero field and in presence of a magnetic field strong enough to quench superconductivity. The zero-field data yields a superconducting gap of reasonable magnitude displaying a doping dependence similar to the one reported in other families of cuprate. The normal-state data together with the results of the resistivity measurements allows us to test the Wiedemann-Franz (WF) law, the validity of which was confirmed in an overdoped sample in agreement with previous studies. In contrast, a systematic deviation from the WF law was resolved for samples displaying either a lower doping content or a higher disorder. Thus, in the vicinity of the metal-insulator crossover, heat conduction in the zero-temperature limit appears to become significantly larger than predicted by the WF law. Possible origins of this observation are discussed.