Search

Supporting the grid voltage and suppressing the active power oscillation have recently become two major requirements for high penetration of distributed generations during asymmetrical faults. But they are rarely considered in the low-voltage microgrid including zero sequence voltage and current components. Therefore, this paper proposes an optimal voltage support control without active power oscillation in the four-wire inverter system, which also satisfies the constraints of maximum allowable peak current. Both the increase of positive sequence voltage, and the decrease of negative and zero sequence voltage can be well realized and optimized through the equivalent grid impedance. Then, the sum of zero sequence active power oscillation and the related oscillations caused by positive and negative sequence coupling is guaranteed to be zero. Finally, the validity of the proposed method is verified by the simulation and experimental results.

In this paper, a novel image splicing detection and localization scheme is proposed based on the local feature descriptor which is learned by deep convolutional neural network (CNN). A two-branch CNN, which serves as an expressive local descriptor is presented and applied to automatically learn hierarchical representations from the input RGB color or grayscale test images. The first layer of the proposed CNN model is used to suppress the effects of image contents and extract the diverse and expressive residual features, which is deliberately designed for image splicing detection applications. In specific, the kernels of the first convolutional layer are initialized with an optimized combination of the 30 linear high-pass filters used in calculation of residual maps in spatial rich model (SRM), and is fine-tuned through a constrained learning strategy to retain the high-pass filtering properties for the learned kernels. Both the contrastive loss and cross entropy loss are utilized to jointly improve the generalization ability of the proposed CNN model. With the block-wise dense features for a test image extracted by the pre-trained CNN-based local descriptor, an effective feature fusion strategy, known as block pooling, is adopted to obtain the final discriminative features for image splicing detection with SVM. Based on the pre-trained CNN model, an image splicing localization scheme is further developed by incorporating the fully connected conditional random field (CRF). Extensive experimental results on several public datasets show that the proposed CNN based scheme outperforms some state-of-the-art methods not only in image splicing detection and localization performance, but also in robustness against JPEG compression.

As an important topic in hyperspectral image (HSI) analysis, band selection has attracted increasing attention in the last two decades for dimensionality reduction in HSI. With the great success of deep learning (DL)-based models recently, a robust unsupervised band selection (UBS) neural network is highly desired, particularly due to the lack of sufficient ground truth information to train the DL networks. Existing DL models for band selection either depend on the class label information or have unstable results via ranking the learned weights. To tackle these challenging issues, in this article, we propose a Gumbel-Softmax (GS) trick enabled concrete autoencoder-based UBS framework (CAE-UBS) for HSI, in which the learning process is featured by the introduced concrete random variables and the reconstruction loss. By searching from the generated potential band selection candidates from the concrete encoder, the optimal band subset can be selected based on an information entropy (IE) criterion. The idea of the CAE-UBS is quite straightforward, which does not rely on any complicated strategies or metrics. The robust performance on four publicly available datasets has validated the superiority of our CAE-UBS framework in the classification of the HSIs.

With the development of renewable energy generation technology, the DC power distribution system (DPSs) commonly contains various buses with different voltage levels. Under this circumstance, it is much more challenging to carry out stability analysis for such multi-voltage level DPSs. According to the mainstream methods, the multi-voltage level DPSs are directly equivalent to the source subsystem and load subsystem. However, this kind of impedance equivalence and stability analysis hide each voltage-level DC bus's impedance characteristic. When the impedance instability occurs, it will be challenging to identify the instability DC bus in DPSs. In this paper, a stepwise impedance analysis method for multi-level DPSs is proposed. The impedance stability is firstly implemented in the lowest voltage level of DPSs with the stability criterion. If the lowest voltage level of DPSs is stable, then the voltage level's impedance is converted to the upper voltage level through an interface converter. With the step-by-step stability analysis and equivalent impedance calculation, the multi-voltage level DPSs can be guaranteed if all of the voltage levels are derived stable. Lastly, the effectiveness of the proposed method is validated by simulations in Matlab/Simulink.

With the exhaustion of energy and more and more stringent requirements on the ecological environment, more and more people attach importance to new energy. Wi5nd energy is favored all over the world because of its pollution-free and huge capacity. However, the wind speed and direction of the natural wind change at all times, which brings great difficulties to stabilize the rotation speed and power of the wind wheel. In this paper, a variable pitch system is designed based on active disturbance rejection control (ADRC), and then the mathematical modeling of wind turbine is carried out by MATLAB/SIMULINK software. The changes of wind turbine speed and pitch Angle are discussed in high wind speed. Through the simulation of step wind and random wind at high wind speed, the simulation situation is closer to the actual natural situation. In view of the difficulty in parameter setting of ADRC, on the one hand, the empirical method is used to set parameters, and on the other hand, particle swarm optimization (PSO) is used to self-tune parameters of ADRC.

Grid-feeding converters are required to maintain the operation capability and support the grid voltage during asymmetrical faults. This paper considers these two demands and proposes an active power oscillation control strategy of grid-feeding converters. By setting a flexible reference current expression with four control parameters, power allocations are mathematically derived to achieve different objectives. Thus, the asymmetrical current is injected to meet the next generation grid code and reduce the voltage unbalance factor (VUF) at the point of common coupling (PCC). The proposed strategy can also attenuate the active power oscillation and take full advantage of the converter capacity, which is implemented in two phase stationary frame without phase-locked loop. Experimental results verify the correctness of the proposed control strategy.

This paper investigates the current coupling characteristics of N-paralleled grid-connected inverters, which affects the design and operation of inverter, and makes fault current more complex and difficult to analyze. In this paper, by establishing the fault models of paralleled droop-controlled inverter, an asymmetrical fault current calculation method for N-paralleled droop-controlled inverter is proposed. On this basis, influence of main factors, including controller parameters, reference power and inverter line impedance, on fault current coupling degree are analyzed quantitively. Simulation results and calculation results verify the practical of the proposed calculation method and the correctness of theoretical analysis of coupling effect.

Motor bearing is subjected to the joint effects of much more loads, transmissions, and shocks that cause bearing fault and machinery breakdown. A vibration signal analysis method is the most popular technique that is used to monitor and diagnose the fault of motor bearing. However, the application of the vibration signal analysis method for motor bearing is very limited in engineering practice. In this paper, on the basis of comparing fault feature extraction by using empirical wavelet transform (EWT) and Hilbert transform with the theoretical calculation, a new motor bearing fault diagnosis method based on integrating EWT, fuzzy entropy, and support vector machine (SVM) called EWTFSFD is proposed. In the proposed method, a novel signal processing method called EWT is used to decompose vibration signal into multiple components in order to extract a series of amplitude modulated–frequency modulated (AM-FM) components with supporting Fourier spectrum under an orthogonal basis. Then, fuzzy entropy is utilized to measure the complexity of vibration signal, reflect the complexity changes of intrinsic oscillation, and compute the fuzzy entropy values of AM-FM components, which are regarded as the inputs of the SVM model to train and construct an SVM classifier for fulfilling fault pattern recognition. Finally, the effectiveness of the proposed method is validated by using the simulated signal and real motor bearing vibration signals. The experiment results show that the EWT outperforms empirical mode decomposition for decomposing the signal into multiple components, and the proposed EWTFSFD method can accurately and effectively achieve the fault diagnosis of motor bearing.

Three-phase grid-connected converter usually adopts different power and current control strategies under unbalanced conditions according to the practical application, but the power relationship is mainly discussed in three-phase three-wire system, and zero sequence power is not considered. Though theoretical analysis of the vector relationship between instantaneous power and current, a general expression of reference current for three-phase four-wire converter under unbalanced faults is proposed in this paper, which can realize switching between five fixed objective controls. On this basis, three coordinated objective control strategies are presented, which can be applied to three-phase four-wire converter with different neutral phase overcurrent capability. The proposed control can be easily implemented in αβ0 stationary frame by using only two parameters without phase-locked loop, and can satisfy the case of single-phase and two-phase grounding faults. Simulation results are provided to verify the proposed method.

As a voltage-controlled inverter, droop-controlled inverter will provide large inrush fault current when subjected to short-circuit faults, causing unplanned disconnection from the utility grid or even power devices damage. Therefore, to ensure the reliability of protection system, it is very important to figure out the fault current characteristics. This paper proposes an asymmetrical fault current iterative algorithm of droop-controlled inverter by using systems’ symmetrical components. It is found that the fault current of each phase is composed of an attenuated periodic component and a dc component. The periodic component is mainly determined by the control parameters, grid equivalent voltage and equivalent line impedance. While the dc component is only related to grid parameters and fault moment. Comparisons between calculation results and simulation results or experimental results verified the correctness and feasibility of proposed method.

This paper proposes a new method for the testing of accelerometers by using a laser feedback interferometer (LFI). In this method, a portion of light from a laser diode is reflected by an external target and then returns back to the laser cavity, laser feedback interference occurs. Depending on various vibration frequencies, fringe counting method or minimum point method is introduced into the interference signal analysis of LFI to determine the sensitivity of accelerometers. Theoretical analysis and experimental results are given. The proposed system is applicable to test accelerometers in industrial applications with a low-cost and simple mechanical structure.

Based on the feature analysis of image content, this paper proposes a novel steganalytic method for grayscale images in spatial domain. In this work, we firstly investigates directional lifting wavelet transform (DLWT) as a sparse representation in compressive sensing (CS) domain. Then a block CS (BCS) measurement matrix is designed by using the generalized Gaussian distribution (GGD) model, in which the measurement matrix can be used to sense the DLWT coefficients of images to reflect the feature residual introduced by steganography. Extensive experiments are showed that proposed scheme CS-based is feasible and universal for detecting stegography in spatial domain.

Data secure processing is the important issue of video authentication in cloud environment. This research presents a novel scheme to protect integrity of video content for common video data operations by using a semi-fragile CS-watermark technology. In proposed scheme, the CS-watermark data are generated from the block compressed sensing (CS) measurements which rely on the knowledge of the measurement matrix used for sensing I frame's DCT coefficients. Our analysis and results indicate that the CS-watermark data can accurately verify the integrity of the original video content, and have higher security than other watermarking methods.

Singular value decomposition (SVD) can realize denoising without relying on spectral characteristics. It is more useful for small scale denoising. Bispectrum can effectively inhibit the interference of non Gaussian noise, which makes the signal feature extraction convenient. The two methods are combined in this research. In the beginning, the vibration signals of engine crankshaft bearings go through SVD-based denoising, and then the high-order spectral theory is adopted to get the bispectrum of the signals after denoising. In the end, the frequency band of the fault crankshaft bearings signal is extracted by searching the whole 2-D frequency field, and favorable diagnosing result is obtained.

This article describes and analyses the main technologies of MVC (Multi-view video coding) based on H.264, which include MVC temporal scalable, inter-view prediction, parallel processing and structure of MVC bit streams. For further analysis of MVC coding efficiency, some experiments are designed. Compared with the single view coding of simulcast, the MVC compression coding based on the JMVC model can obtain high coding efficiency.

In this paper, by using the synchrotron radiation real-time imaging, the growth of gas bubbles on solid-liquid interface during the soldering process was investigated. It was proved that the formation and growth of the bubbles was a process of heterogeneous nucleation. During the soldering heating insulation stage, the bubbles volume increased first and remained unchanged afterwards. When there were more than one bubbles, the annexation process was happened among the adjacent bubbles, the larger bubbles were pulled to the smaller ones. The bubbles volume remained unchanged during the cooling stage. At the same time, the dissolution of the Cu substrate and the growth of the IMCs were influenced by the bubbles.

For the shortcomings of current video watermarking algorithms on tamper detection, combined semi-fragile watermarking with Compressive Sensing theory, the paper proposes a novel video content authentication algorithm for MPEG-2. The image features of I-frame are extracted by Compressed Sensing algorithm, and semi-fragile content authentication watermarking are generated, in which watermarking bits are embedded into the low-frequency DCT coefficients of I-frame. Simulation results show that the algorithm has better detection ability and detection accuracy than invertible semi-fragile watermarking algorithm distinguishing MPEG-2 compression from malicious manipulation the algorithm. For video inner frames tampering, accuracy of the algorithm can reach to the sub-block of image.

Many security-related applications can benefit from constructing models to predict the behavior of an entity. However, such models do not provide the user with explicit knowledge that can be directly used to influence the behavior for his/her interest. This type of knowledge is called actionable knowledge. Actionability is a very important aspect of the interestingness of mined patterns. In this paper, we formally define a new problem of mining actionable behavioral rules from group data. We also propose an algorithm for solving the problem. Using terrorism group data, our experiment shows the validity of our approach as well as the practical value of our defined problem in security informatics.

The vibration signal of transmission bearing is decomposed by empirical mode decomposition, and fractal dimensions of decomposed intrinsic mode functions are calculated to extract fault characters of transmission bearings of different conditions. The results show that empirical mode decomposition method is able to separate signals of different frequency bands effectively, and the fractal dimension of specific IMF component is able to reflect the technology state of transmission bearing sensitively, which can be selected as the characteristic parameters to diagnose transmission bearing fault. The combination of empirical mode decomposition and fractal dimension is an effective method of transmission bearing's fault character extraction.

For Multi-channel data acquisition and analysis, Blind signal separation technique and Fast ICA is an effective method; In this paper, data acquisition objects is the vibration signal of electric traction experiment system, Fast ICA algorithm is adapted to separate the complex and aliasing measured vibration signal, separation results show that major source signals have been successfully and clearly separated from the complex and aliasing vibration signals through the separation algorithm, Then the characteristics of each source vibration signals can be effectively extracted in different frequency by analyzing source signals. The results are helpful for the later data analysis.

Timely and effective surveillance is critical for the prevention and control of epidemics. However, due to technical challenges and shortage of human resources, comprehensive and timely data collection required for effective surveillance, especially collection of data about sudden epidemic outbreaks, is still very difficult. In this paper, we propose the use of multi-source web data for epidemic surveillance. We use the 2009 Influenza A (H1N1) pandemic in Beijing as a case study to demonstrate the utility of our proposed approach. Experiments using data from the Beijing Center for Disease Control and Prevention (CDC) and several search engines show encouraging results. This case study also has direct practical values in the real setting.

A major concern in open source software development is regarding bugs emerging during the life circle of a software project. Understanding the topological characteristics of interrelated bugs can provide useful insights into software project management and potentially facilitate the development of new complex network models. In this paper, we analyze the bug network in Gentoo Linux, one of the most popular Linux distributions. We model software bugs as nodes and the blocking relationships among them as edges. As the resulting Gentoo bug network cannot be adequately explained by some commonly-used complex network models, we propose a new model, which can better explain this network.

Topic Detection and Tracking provides a flat and unorganized view of a document collection and cannot adequately reflect the content of the complete collection as some of the information is lost in the process. Topic models account for more information and lead to a more organized view of the document collection. In this paper, we propose a more efficient model named Related Topic Network with a new term weighting method. Empirical evaluation using two real-world datasets consisting of 953 and 5,550 news documents demonstrates the utility of the proposed model and shows that the new term weighting method leads to performance improvement.

In this paper, a probe into the stabilization effects of ferrous sulfate (FeSO4) came from two aspects-stabilization efficiency and leachate content. Fe/As molar ratio, arsenic pollution intensity, soil pH, reacted time interval, competitive ions, such as sulfate(SO42-), phosphate(PO43-) and zinc ion(Zn2+), were studied in order to discuss their effects on the stabilization of arsenic. Results show that ,to all different levels of arsenic contaminated soils ,when Fe/As molar ratio reaches 6, arsenic stabilization efficiency exceeds 90%,and arsenic concentration in leachate could also reduce to less than 5 mg/L; in the soil pH range from 8.10 to 8.57,ferrous sulfate has obtained the preferable results, and pH=8.23 is the optimum value; because of existing of competitive actions ,the exogenous adding of SO42-、PO43- and Zn2+ could lead to much poorer performance for FeSO4.

The spatio-temoral characteristics of statistical data was research challenges in Statistical GIS. In this paper, give an effective method to implement spatio-temporal organization and query of statistical data. First, introducing the research situation of spatio-temporal data model and analyzing the spatio-temporal feature of statistical data. Then proposing a data model using object-oriented method and introduce how to realize it in the database. At last, introducing t some of the most common query algorithm in the Statistical GIS. Proved by programming realization, long-term statistical data can be effectively organized and stored in the database, spatio-temporal snapshot query and sequences query can be well executed in this way.

This paper presents a new approach to the problem of absolute stability analysis of Lur's systems subject to time-varying delay. A more general Lyapunov-Krasovskii functional is constructed to obtain some delay-dependent stability criteria by combining the free-weighting matrix technique. Since these criteria utilize more information on the state variables and no useful item is ignored throughout the estimate of upper bound of the derivative of Lyapunov functional, the developed conditions show less conservatism. An numerical example is given to illustrate the effectiveness of our proposed method.