Search

In this article, a fixed and smooth-switch-sequence modulation strategy is proposed to equalize the dc-link voltages among the cascaded rectifier. The structure of a single-phase three-level neutral-point-clamped cascaded rectifier is studied. Three core steps of this strategy are analyzed respectively, including carrier wave disposition calculation, fixed pulse generator, and voltage rank function. In addition, a criterion of the degree of the load unbalance is established in order to measure the voltage balancing boundary of the presented approach. The superiorities of the proposed control strategy embody the smooth switch sequence, the strong voltage balance capability, as well as the fixed switch sequence. Owe to these advantages, the cascaded rectifier could effectively reduce the switching frequency, achieve the goals of voltage balancing, and reduce the calculation burden of the controller when the number of cascaded modules is large. The effectiveness of the proposed strategy is verified by simulation and experiment results. [ABSTRACT FROM AUTHOR]

Local geometric descriptors act as an essential component for 3D rigid data matching. A rotational invariant local geometric descriptor usually consists of two components: local reference frame (LRF) and feature representation. However, existing evaluation efforts have mainly been paid on the LRF or the overall descriptor and the quantitative comparison of feature representations remains unexplored. This paper fills the gap by comprehensively evaluating nine state-of-the-art local geometric feature representations. In particular, our evaluation assesses feature representations based on ground-truth LRFs such that the ranking of tested methods is more convincing as compared with existing studies. The experiments are deployed on six standard datasets with various application scenarios (shape retrieval, point cloud registration, and object recognition) and data modalities (LiDAR, Kinect, and Space Time) as well as perturbations including Gaussian noise, shot noise, data decimation, clutter, occlusion, and limited overlap. The evaluated terms cover the major concerns for a feature representation, e.g., distinctiveness, robustness, compactness, and efficiency. The outcomes present interesting findings that may shed new light on this community and provide complementary perspectives to existing evaluations on the topic of local geometric feature description. A summary of evaluated methods regarding their peculiarities is finally presented to guide real-world applications and new descriptor crafting. [ABSTRACT FROM AUTHOR]

In this article, a graphic processing unit (GPU)-based acceleration implementation of a hybrid discontinuous Galerkin time domain method based on Maxwell’s equations and Helmholtz vector wave equation (HDGTD) has been developed. The computational domain is discretized by tetrahedrons and the resultant meshes are categorized into two regions solved by Maxwell’s equations and Helmholtz vector wave equation, respectively. The hierarchical vector basis functions are used to expand the unknowns in the HDGTD method, and a universal matrix technique is proposed to decompose the geometry-dependent matrices in each tetrahedron into the summation of universal matrices defined in barycentric coordinates, thus giving rise to a great decrease of the memory usage. A local time stepping (LTS) method based on a simple interpolation is introduced in the proposed HDGTD method to achieve highly efficient solution of multiscale problems. Two kinds of compute unified device architecture (CUDA)-based mapping techniques, i.e., 1-D and 2-D blocks, are implemented to achieve a tradeoff between the parallel speedup and the memory usage. With the 1-D block mapping, over 590 times speedup can be achieved, and in the case of the 2-D block mapping, over 150 times acceleration and 13 times memory reduction are obtained. Some practical complex examples are given to demonstrate a good performance of the proposed parallel method. [ABSTRACT FROM AUTHOR]

Deep learning (DL) architecture, which exploits multiple hidden layers to learn hierarchical representations automatically from massive input data, presents a promising tool for characterizing fault conditions. This paper proposes a DL-based multi-signal fault diagnosis method that leverages the powerful feature learning ability of a convolutional neural network (CNN) in images. The proposed deep model is able to learn from multiple types of sensor signals simultaneously so that it can achieve robust performance and finally realize accurate induction motor fault recognition. First, the acquired sensor signals are converted to time–frequency distribution (TFD) by wavelet transform. Then, a deep CNN is applied to learning discriminative representations from the TFD images. Since then, a fully connected layer in deep architecture gives the prediction of induction motor condition based on learned features. In order to verify the effectiveness of the designed deep model, experiments are carried out on a machine fault simulator where both vibration and current signals are analyzed. Experimental results indicate that the proposed method outperforms traditional fault diagnosis methods, hence, demonstrating effectiveness in induction motor application. Compared with conventional methods that rely on delicate features extracted by experienced experts, the proposed deep model is able to automatically learn and select suitable features that contribute to accurate fault diagnosis. Compared with single-signal input, the multi-signal model has more accurate and stable performance and overcomes the overfitting problem to some degree. [ABSTRACT FROM AUTHOR]

Different from single-bus dc microgrids (MGs), a multibus MG normally bridge multiple dc buses via a complex line impedance network. This intricate configuration together with tickling constant power loads may cause severe stability issues. For system stability improvement, conventional ways include designing small-signal stabilizers in a decentralized way and constructing large-signal stabilizers by solving Lyapunov equation in a central controller. Note that centralized methods may be vulnerable to single point of failures, while decentralized patterns are commended to reinforce MG reliability and scalability. To implement large-signal stabilization in decentralized mechanisms, which has rarely been investigated in the existing literature, a novel uniform nonsmooth control scheme (UNCS) is proposed, in this article. UNCS has the following fourfold advantages. First, UNCS helps to express commonly used dc/dc converters, boost, buck, and buck–boost, into a uniform model. The model substantially facilitates a uniform stabilizer design that accommodates the preceding three converters despite their differences in topology. Second, UNCS assigns each converter based distributed generator (DG) with a composite large-signal stabilizer comprising a nonsmooth observer and a nonsmooth controller. The controller neutralizes the disturbances estimated by the observer, and it also rigorously stabilizes the internal states in DG uniform model. Third, the stabilizer also provides an additional control degree of freedom, a finetuning factor, to refine DG dynamic responses to meet high requirements for voltage and power qualities. Fourth, when numerous DGs with the proposed stabilizer congregate to build a multibus MG, the overall system large-signal stabilization is autonomously attained without communication links, which is mathematically justified by Lyapunov stability analyses. As such, DGs are allowed to randomly plug in/out without worrying about impairing system level stability. Simulations help to properly select relevant key parameters. Experiments show that UNCS outperforms the standardly designed PI controls, and it will stabilize the multibus MG in a large operation range. [ABSTRACT FROM AUTHOR]

In this paper, a novel high efficient dual-wavelength mid-infrared optical parametric oscillator (OPO) was demonstrated, which was pumped by a dual-wavelength Raman fiber oscillator for the first time. The adoption of oscillator structure reduced the waste of power, increased Raman pump power and promoted its frequency conversion. The dual-wavelength pump source was fixed at 1070 nm and 1120 nm and the Raman conversion efficiency was greatly enhanced from 26.7% to 72.1%. The high Raman efficiency and low Raman threshold enabled the Raman laser to build optical parametric oscillation easily and the pump-to-idler conversion efficiency of Raman laser improved greatly from 4.2% to 13.3%. The generated dual-wavelength mid-infrared laser was fixed at 3272 nm and 3663 nm, with a maximum power of 6.87 W, indicating a 13.9% pump-to-idler conversion efficiency. The highest mid-infrared output power generated by optical parametric oscillation of Raman laser was also achieved, which is almost ten times than before. [ABSTRACT FROM AUTHOR]

In this article, an electrical doping method to realize the high n-type doping for a germanium (Ge) layer in a low-doped n-type Ge metal-oxide-semiconductor (MOS) structure in the accumulation condition is proposed, which maintains the initial crystal quality of Ge. The direct-bandgap electroluminescence (EL) from Ge light-emitting diode (LED) has been demonstrated with 1.5– 1.6μm infrared emission using the metal/graphene/high-k/ interfacial layer (IL)/n-Ge MOS structure fabricated on bulk-Ge substrate. An onset current density of 5 A cm−2 is observed, which is, up to now, the lowest onset current density compared with the Ge LEDs fabricated on unstrained Ge substrates. These results indicate the superiority of the electrical doping method and the interface passivation technique on yielding the high-performance Ge-based LEDs and lasers. [ABSTRACT FROM AUTHOR]

The performance of the long-wavelength infrared (LWIR) HgCdTe photodetectors with nBn structure is susceptible to the thickness of barrier layer and the huge mismatch of the energy band between the barrier layer and the absorption layer. Herein, a numerical simulation analysis for LWIR HgCdTe nBn device is carried out to present an enhanced performance with optimized structural and physical parameters (including thickness, doping concentration, and composition). The results provide that the valence band-offset (VBO) could be reduced by structure optimization, and could greatly improve the transport properties of photogenerated carriers. In addition, the effect of composition diffusion caused by the annealing process has also been studied to demonstrate that larger composition diffusion leads to a much higher dark current. The model and calculation results established in this article can provide important theoretical support and guidance for further researches of LWIR HgCdTe nBn devices. [ABSTRACT FROM AUTHOR]

The uncertainty of power system is intensified by the integration of large-scale renewable energy resources such as wind farm (WF). Considering the impact of system uncertainties, the probabilistic stability analysis methods have been used in the stability assessments of power system with the WF integration. However, in traditional probabilistic analysis methods, WF is normally considered as PQ bus or aggregated as one single-machine equivalent model regardless of wake effect, which might reduce the accuracy of probabilistic stability studies. In this article, a dynamic equivalent modeling method of WF for probabilistic stability assessments is proposed. The wake effect is considered in the modeling process and a practical four-machine clustering method is used in wind turbines clustering. Besides, the fisher discriminant analysis (FDA) is adopt to combine the similar WTs clustering results reasonably. Then, the WF is aggregated to a multi-machine equivalent model by capacity weighted method. Also, the established WF probabilistic model can be directly used in time-domain simulation based on Monte Carlo simulation and FDA. Finally, the efficiency of the proposed method is verified in an actual WF in China. [ABSTRACT FROM AUTHOR]

Managing the distributed energy resources in cybersecure smart grids will involve placing trust in organizations that act between operators and proactive consumers. We present a Stackelberg game in a three-party neighborhood area network in conjunction with open access and preservation of private data. [ABSTRACT FROM AUTHOR]

Automatic vehicle detection from a video synthetic aperture radar (VideoSAR) system presents significant potential to enhance the surveillance performance in dynamic region of interest (DROI). In this paper, a novel VideoSAR low-rank plus sparse decomposition (LRSD) perspective for single-channel single-pass configuration is proposed to track the ground defocusing vehicles. Vehicle imaging features with 2-D motion parameters are derived theoretically by exploiting a priori knowledge of polar format algorithm (PFA). In accordance with the revealed characteristics, a vision-based VideoSAR-LRSD algorithm, called three-term decomposition (TTD) with proximal exchange-based alternating directions method of multipliers (PEADMM), is then proposed to improve the performance of vehicle detection. It can be used to break the limitation for the application of emergency response not permitting the acquisition of multi-channel or multi-pass data. We comprehensively demonstrate using extensive VideoSAR DROI experiments that in comparison with the state-of-the-art algorithms, TTD-PEADMM algorithm presents the improved accuracy and is able to offer competitive results. [ABSTRACT FROM AUTHOR]

The utilization of voltage-source converter-based multi-terminal HVDC (VSC-MTDC) technology has been envisaged as a cost-effective solution to transmit bulk renewables from multiple sending ends to receiving-end AC systems. However, with complex configurations and various types of components, the reliability evaluation of a VSC-MTDC system is more complicated than that of a conventional HVDC system. In addition, due to stochastic failures and complex operating modes, the application of a VSC-MTDC system can have a significant influence on the reliability of hybrid AC/DC power systems. This paper presents a new technique for evaluating the nodal reliability of a hybrid AC/DC power system based on the VSC-MTDC system. The proposed technique is incorporated within a multi-state model for the VSC-MTDC system. Then, an optimization model for contingency management of a hybrid AC/DC power system is formulated. The model is established based on the optimal power flow technique for a VSC-MTDC based hybrid AC/DC power system. The objective of the model is to minimize the total system cost when determining the load curtailment and generation re-dispatch results for each contingency state. Nodal reliability indices for hybrid AC/DC power systems are proposed to evaluate the reliability performance of customers at each node. The effectiveness of the proposed methods is validated using a modified IEEE reliability test system. [ABSTRACT FROM AUTHOR]

Objective: The interaction between superparamagnetic iron oxide (SPIO) nanoparticles and mesenchymal stem cells (MSCs) in the presence of pulsed magnetic field (PMF) has become an important area of research in recent years. Methods: A parameter-adjustable pulsed magnetic field was developed based on the principle of insulated gate bipolar translator transistor-controlled discharge of large capacitances. The internalizations of SPIO nanoparticles by MSCs were investigated under the treatment of PMF in both intermittent stimulation mode and continuous stimulation mode. Results: The intensities and frequencies of pulsed magnetic field can be adjustable in the range of 1.9–4.6 mT and 3–5 kHz, respectively. This PMF was safe to the MSCs. However, the uptake of SPIO nanoparticles by MSCs was significantly prohibited under the treatment of kHz-ranged PMF while the 10 Hz PMF enhanced the cellular uptake of nanoparticles. This phenomenon was relative with the magnetic effect of the PMF with different frequency. Conclusion: The PMF can be used to effectively regulate the cellular uptake of SPIO nanoparticles and the mechanism lies in the magnetic effect. Significance: The interaction between SPIO nanoparticles and the MSCs is a fundamental and important issue for nanomedicine and stem cell research. Our results demonstrate that the external magnetic field can be used to regulate their interaction. We believe that this safe, facile, and flexible method will greatly promote the development and clinical translation of regenerative medicine and nanomedicine. [ABSTRACT FROM AUTHOR]

Power loss and voltage regulation are two fundamental and challenging issues in DC distributed generation system. In this paper, a multi-objective optimization problem is established to achieve trade-offs between reducing power loss and decreasing voltage deterioration by adjusting the weighting coefficients. This problem is proven to be a convex optimization under constant impedance load and constant current load. A distributed algorithm based on random coordinate descent method is developed to solve this problem without acquiring the information of line impedance. Thus, the optimal operating point of the system can be found automatically in real-time, and the deviation of optimal solution caused by the fluctuation of network parameters is avoided. In addition, the complexity of this algorithm is low and its computational requirement is reduced in practice. Finally, case studies are implemented to verify the effectiveness of the proposed algorithm under the scenarios of line parameter fluctuation, load variation, communication line failure, and plug-play capabilities. [ABSTRACT FROM AUTHOR]

This paper reports an intracavity terahertz parametric source based on the stimulated polariton scattering in RbTiOPO4 crystal with a high repetition rate and a high average power. The side pumping for the gain medium by laser diodes is adopted to get a higher fundamental power and a larger fundamental laser beam size than those in the diode-end-pumping in order to improve the THz output power. A non-collinear convex-plane fundamental cavity is adopted so that the thermal effect can be offset in some degree and the fundamental beam size is further increased. The obtained maximum average THz output power is 367 μW at 3.88 THz when the diode pump power is 105 W and the pulse repetition frequency (PRF) is 7 kHz. The terahertz power of 367 μW is the highest ever obtained in SPS sources. [ABSTRACT FROM AUTHOR]

Due to the maturity of wide area measurement technology, phasor measurement unit (PMU)-based dynamic equivalent modeling method (PDM) has been wildly used in large-scale power system modeling. However, traditional “event-based” PDM equivalent model may face serious robustness issues, such as although derived equivalent models show the good ability on reproduction of the system behaviors under recorded events where they derived from, they may not work well for other possible disturbances. In this article, to improve the robustness of equivalent model, a novel method on dynamic equivalent modeling of power system is proposed. Based on the result of coherency identification, generators inside the equivalent system can be aggregated into different coherent subgroups and represented by multimachine equivalent model. Key parameter selection, which can shrink the scale of identified parameters, is designed to avoid multiple solutions of the equivalent model. Moreover, a multiobjective parameter identification algorithm is developed to fully consider the PMU measurements in different system events. In addition, hybrid dynamic simulation is performed so that each equivalent submodel can be identified separately. Finally, the effectiveness of the proposed method is verified in an actual power grids in China. [ABSTRACT FROM AUTHOR]

SIN is constructed to support real-time data acquisition, massive data transmission and processing, and systematized information services. Different from static networks, both the network topologies and the available network resources are dynamic in SIN. However, the existing networking technologies treat SIN as segmented static networks, and ignore the relationship among different segmented subnetworks, which results in low resource utilization and poor transmission QoS. This article exploits the time-varying graph theory to characterize and allocate the dynamic resources of SIN. We first introduce the overall SIN architecture, and discuss its key technologies in terms of network slicing and dynamic routing. Then, we construct a novel STAG to precisely depict the multi-dimensional time-varying resources and reveal their connection relationships in SIN. With STAG, we propose the dynamic network slicing strategy to build the dedicated network slices for different services, and design the intra-slice routing scheme to guarantee the service transmission QoS. Simulation results demonstrate our methods can complete more space missions with higher resource utilization. Finally, we discuss the promising future of the time-varying graph theory. [ABSTRACT FROM AUTHOR]

This paper proposes a semi-consensus strategy for multi-functional hybrid energy storage systems (HESSs) in DC microgrids. Batteries in a HESS are regulated by conventional V-P droops and supercapacitors (SCs) are with integral droops (ID). Only batteries are assigned with local distributed compensators which exchange information through sparse communication links. Those SCs are exempted from data exchange process, which would save system investment costs. Within the semi-consensus scheme, the most essential function is the cooperation of V-P droop and ID that helps to naturally allocate low frequency components of load power to batteries and high frequency components to SCs, thus prolonging the overall life time of HESS. In addition to the transient power allocation function, there are other three functions endowed by the proposed strategy, which are autonomous DC bus voltage recovery to its nominal level, spontaneous SC state of charge (SOC) restoration, autonomous power sharing and SOC balancing among batteries. It is the simultaneous realization of above four functions with limit communications that makes up the main contributions in this paper. A generic mathematical modeling of HESS with the semi-consensus strategy is established. The model allows for dynamic analyses to theoretically validate the effectiveness of proposed method in both frequency and time domains. In-house experimental results are shown fully consistent with the dynamic analyses and also effectively corroborate the intended HESS multi-functional operations. [ABSTRACT FROM AUTHOR]

With more distributed energy resources penetrated into the residential community, nanogrid based peer-to-peer (P2P) energy market has rapidly emerged over recent years. Due to the complexities on the decision-making process of each market participant, an efficient, fair and beneficial oriented bidding strategy is thus necessary. In this article, a two-stage bidding strategy for P2P trading of nanogrid is proposed. To overcome the limitations of traditional methods, in the first stage, a supply-demand relationship considered two-step price predictor, which aims to promote the usage of local renewable energy, is formulated to provide the guidance on transaction adjustment. In the second stage, trading preference based simultaneous game-theoretic approach is fully introduced, which can optimize the market equilibrium and then increase the social welfare of the P2P market. Additionally, to mitigate the possible failure of price matching, value-at-risk is implemented through the trading process as a risk hedging tool. To verify the effectiveness of the proposed strategy, usages of local renewable energy, economic benefits and success rates of transaction is compared against the traditional method for various cases. [ABSTRACT FROM AUTHOR]

Considering the high cost and complexity of multiple-channel blind source separation, single-channel solutions have elicited a significant amount of interest in source separation. Joint maximum likelihood sequence estimation implemented by the Viterbi algorithm (VA) is a seemingly ultimate solution to the problem of data transmission over dispersive channels. However, both joint reduced-state sequence estimation (JRSSE) and joint delayed-decision feedback sequence estimation (JDDFSE) have numerous shortcomings when applied to single-channel blind signal separation. Hence, joint double-direction DDFSE (JDD-DDFSE) is proposed in this article. JDD-DDFSE employs joint VA (JVA) of the low-order channel twice. The estimated symbols from the initial JVA are utilized to equalize the causal and non-causal taps of the channel beyond the trellis state in the second JVA. A method that can derive the log likelihood ratio of the estimated symbols is provided; the complexity of this method is considerably lower than that of soft-output VA or the BCJR algorithm. Hence, the entire computation complexity of the iterative separation structure basically remains unchanged compared with that of the no-iteration structure. For the non-causal channel, the proposed algorithm has the lowest computation complexity and demonstrates a separation performance similar to that of conventional JVA, JRSSE, and JDDFSE algorithms. [ABSTRACT FROM AUTHOR]

Effective detection of multifaults in bearings and gears is a challenging issue in rotary machinery health monitoring. As such, a generalized Vold–Kalman filtering (GVKF)-based compound faults diagnosis method is presented in this paper. The technique includes four main steps: 1) a time–frequency ridge is separated from the time–frequency representation (TFR) of the vibration signal using a peak search method; 2) according to the time–frequency ridge, GVKF parameters corresponding to all the fault characteristic frequencies (FCFs) are estimated; 3) the fault feature components are obtained using the generalized demodulation transform (GDT) and the VKF with the GVKF parameters; and 4) the spectra obtained by the fast Fourier transform (FFT) are used to fault detection. The main contributions of the proposed method are as follows: 1) the influence of speed fluctuations and the unrelated harmonic components are removed through the integration of the GDT and the VKF and 2) the tachometerless GVKF parameters are defined and calculated to quantitatively detect different fault types, which avoids missed diagnosis and misdiagnosis. The proposed multifault diagnosis algorithm is verified by both simulation and experiment data. Comparison with other commonly used techniques has shown the advantage of the new method. [ABSTRACT FROM AUTHOR]

Electromagnetic pulse (EMP)-induced nonlinear degradation of an L-band gallium arsenide-based low-noise amplifier (LNA) is studied by an injection experiment, failure analysis, and mixed-mode simulation. The experimental results indicate that the real-time response, radio frequency, and direct current characteristic of LNA samples under an EMP show nonlinear and permanent degradation features. In detail, the characteristics first degrade and then recover slightly as the injection power increases. In addition, the nonlinear degradation shows a close dependence on the pulse properties. Failure analysis reveals that the degradation of the LNA is attributed to latent failure defects under and around the gate metal strip of the first-stage high electron mobility transistor, with recovery well interpreted by an EMP-induced “annealing effect.” Simulation results show that if the pulse repetition frequency is smaller than 2.5 kHz, a pulse repetition-induced thermal accumulation effect will not occur and that the pulsewidth τ will mainly determine the actual action duration and the temperature of the “annealing effect.” Based on the above mechanism, the dependence of the pulse properties on nonlinear degradation is well explained. [ABSTRACT FROM AUTHOR]

A virtual synchronous generator (VSG) control based on adaptive virtual inertia is proposed to improve dynamic frequency regulation of microgrid. When the system frequency deviates from the nominal steady-state value, the adaptive inertia control can exhibit a large inertia to slow the dynamic process and, thus, improve frequency nadir. And when the system frequency starts to return, a small inertia is shaped to accelerate system dynamics with a quick transient process. As a result, this flexible inertia property combines the merits of large inertia and small inertia, which contributes to the improvement of dynamic frequency response. The stability of the proposed algorithm is proved by Lyapunov stability theory, and the guidelines on the key control parameters are provided. Finally, both hardware-in-the-loop and experimental results demonstrate the effectiveness of the proposed control algorithm. [ABSTRACT FROM AUTHOR]

In arc flash hazards calculations, protection boundary can be used to gain insight in personnel protection against thermal injuries. Although IEEE 1584-2002 has provided the calculation guide based on tests with the arcing electrodes in a vertical plane, applications where the traditional vertical arrangement is not directly applicable have been drawing more attention over the past few years. Due to the varied thermal behaviors of arc under different equipment configuration and the varied bus connection mode, for the purpose to achieve the adequate protection on arc flash thermal injury, it is necessary to analyze the effect of different electrode geometries on arc flash protection boundary (AFPB). According to the prior work on the development of IEEE 1584-2018, this article fully discusses the impact of electrode geometry on incident energy exposure level and AFPB. With the multiple electrode geometry designed tests, thermal behaviors of arc under different electrode orientation, gap distance, and equipment configurations have been analyzed. According to test results, correction equations were derived under each typical electrode geometry to adjust the AFPB calculation model of IEEE 1584-2002. Through the comparisons, the corrected model can better reflect the actual effects of electrode geometry, meanwhile showing a good ability to provide reasonable AFPB for practical uses. [ABSTRACT FROM AUTHOR]

With ground-based synthetic aperture radar (GB-SAR) discontinuous observation, the effect of rail error, caused by multiple installations, on imaging is difficult to eliminate with difference correction. In this article, the impact of rail error on each point, which is related to the incident angle of each point, is theoretically analyzed. This is the first attempt to introduce a baseline in the application of GB-SAR microdeformation to build a model that estimates the incident angle of each point in the image. Then, the rail error correction model is established using the incident angle, and the influence of the rail error on the image point is weakened using this model. The experimental results show that the precision of the estimated incidence angle cosine is 0.0132, which is comparable to the same angle cosine value measured by the total station. By applying the correction method in this article, different impacts of each point in the image, caused by rail error from multiple instrument installations, can be weakened to below 0.1 mm, meeting the requirement of GB-SAR microdeformation measurement submillimeter accuracy. [ABSTRACT FROM AUTHOR]

Surface flaw inspection is of great importance for quality control in the field of manufacture. In this paper, a novel surface flaw inspection algorithm is proposed based on adaptive multiscale image collection (AMIC) using convolutional neural networks. First, the inspection networks are pretrained with ImageNet data set. Second, the AMIC is established, which consists of adaptive multiscale image extraction and with-contour local extraction from training images. Through the AMIC, the training data set is greatly augmented, and labels of images can be accomplished automatically without artificial consumption. Then, transfer learning is performed with the AMIC established from training data set. Finally, an automatic surface flaw inspection instrument for large-volume metal components embedded with the proposed inspection algorithm is designed. Experiments with small metal components are performed to analyze the influence of parameters, and comparative experiments are carried out. The inspecting precisions for indentation, scratch, and pitted surface of the proposed method are 97.3%, 99.5%, and 100%, respectively. The experimental results demonstrate the effectiveness of the proposed method in the detection of various surface flaws. [ABSTRACT FROM AUTHOR]

A novel thermal infrared pedestrian segmentation algorithm based on conditional generative adversarial network (IPS-cGAN) is proposed for intelligent vehicular applications. The convolution backbone architecture of the generator is based on the improved U-Net with residual blocks for well utilizing regional semantic information. Moreover, cross entropy loss for segmentation is introduced as the condition for the generator. SandwichNet, a novel convolutional network with symmetrical input, is proposed as the discriminator for real–fake segmented images. Based on the c-GAN framework, good segmentation performance could be achieved for thermal infrared pedestrians. Compared to some supervised and unsupervised segmentation algorithms, the proposed algorithm achieves higher accuracy with better robustness, especially for complex scenes. [ABSTRACT FROM AUTHOR]

Memristors with small size, fast speed, low power, CMOS compatibility and nonvolatile modulation of device resistance are promising candidates for the next-generation data storage and in-memory logic computing paradigm. In comparison to the binary logics enabled by memristor devices, multi-valued logics can provide higher computation efficiency with simple operation scheme, reduced circuit complexity, and smaller chip area. In this contribution, we demonstrate that all the 27 univariate ternary logic operations can be realized with a single ZnO three-state resistive switching memristor in at most three steps. The nonvolatile modulation characteristics of the memristor allow the read step to be independent of the logic operation and capacitate logic-in-memory applications. The present methodology could be beneficial for constructing future high-performance computation architectures. [ABSTRACT FROM AUTHOR]

Novel structures of type II, III, and IV linear-phase finite impulse response (FIR) systems are proposed. It is constituted of a linear combination of basic sub-filters, called cardinal filters, with weighting coefficients equal to the derivatives of the system amplitude response at a prescribed frequency. Since the cardinal filters can be synthesized via recursive closed-form expressions, regardless of the desired system amplitude response, the proposed structure provides a universal design for arbitrary derivative-constrained linear-phase FIR filters. Design examples of maximally flat filters show that further improvement by combining derivative constraints with the least squared error optimization technique can be obtained. [ABSTRACT FROM AUTHOR]

High penetration of renewable energies through fast-response power converters results in a considerable displacement of conventional synchronous generators and losing of system inertia for frequency control. Modular multilevel converters (MMCs) can be employed serving as an interface between the large-scale renewable generation and power grids. In a microgrid with high shares of renewables integrating through MMCs, submodule (SM) capacitors can be used as energy storage to provide a degree of synthetic inertia for system frequency support. This paper quantitatively exploits the MMC synthetic inertia in a microgrid with flexible renewable penetration levels. An MMC inertia coefficient concept is proposed. It is mainly affected by the penetration ratio, the SM capacitance, and the modulation index of an MMC, with the system operation constraints. According to the analysis, a substantial portion of system inertia can be provided by properly designed MMCs. Detailed MMC frequency control loops are presented. The capacitor average voltage is proportionally linked to the frequency deviation, in order to flexibly adjust the capacitor energy during frequency events. The MMC output active power is deliberately and simultaneously regulated according to the capacitor energy change rate. By controlling the MMC capacitor voltage, dc side power, and output active power, an amount of energy can be released or absorbed by SM capacitors to improve the system frequency response during the frequency event transients, while the renewable generation is scarcely influenced. The proposed concept is experimentally verified and the results show that, with proper system parameters and control loops, the line frequency deviation and the rate of change of frequency can be significantly reduced. Good agreements of the system frequency characteristics have been achieved between the theoretical calculation and experimental results. [ABSTRACT FROM AUTHOR]

This paper proposes a novel ground-fault location method in resonant grounded systems based on distributed modulation and compensation adjustment. After the permanent single-phase ground fault arc has been suppressed, the Petersen coil will be adjusted to make the zero-sequence currents at the upstream monitoring points of the fault point exceed a certain threshold and be modulated in situ. At this time, the corresponding modulated signals can be extracted from the neutral-to-earth voltage on the busbar for measuring these zero-sequence currents. Taking the carrier-frequency of every monitoring point as its identification tags and the variation characteristic of the zero-sequence currents along feeders as another detection criterion, the fault point can be determined online accurately without a high-performance real-time monitoring network. In this paper, the carrier-frequencies allocation and signal conditioning scheme are proposed for a good detection accuracy of the zero-sequence currents, then the signal detection principle with the control of Petersen coil is described in detail. Simulations and experiments verified the feasibility of the proposed fault location method. [ABSTRACT FROM AUTHOR]

This paper introduces a novel H-bridge structured ac–ac pulsewidth modulation (PWM) converter topology based on two-level non-differential ac chopper legs, which can work at the non-inverting and inverting modes for the utility voltage compensation. Compared to the traditional bipolar-type H-bridge ac–ac converter, no bidirectional switches are switched in a complementary manner to overcome the commutation problem due to the delayed response of electronic circuits and semiconductor switching devices. The detailed PWM control signals and operational principles are presented to regulate the output voltage in a bipolar manner. Especially the main advantage is that the proposed ac–ac converter has the same buck/boost operation process for non-inverting and inverting modes, which ensures the continuously average current supply to the low-voltage output side without a high-value capacitor to support power. Additionally, due to the common sharing ground of the input and output, the feature that output can reverse or maintain phase angle with input is supported well. Then, the detailed analysis, design conditions, and experimental verification based on a 1 kW experimental prototype are presented. [ABSTRACT FROM AUTHOR]

In this article, a novel superresolution mapping (SRM) method based on pixel-, subpixel-, and superpixel-scale spatial dependence (PSSSD) with pansharpening technique is proposed. First, an original coarse resolution remote sensing image and a high-resolution panchromatic image are fused by the pansharpening technique to produce a pansharpened result. A segmentation image with numerous superpixels, which represent irregular objects, is generated by adaptively segmenting the pansharpened result. Second, the class proportions of pixels, subpixels, and superpixels are, respectively, obtained from the original image, the pansharpened image, and the segmentation image. Pixel-scale spatial dependence is obtained using a pixel spatial attraction model. Subpixel-scale spatial dependence is derived using a novel subpixel spatial attraction model. Superpixel-scale spatial dependence is obtained through an extended random walker algorithm. Third, the three-scale spatial dependence is integrated into the PSSSD. Finally, a class allocation method based on object units is used to obtain the SRM results according to the PSSSD. Experimental results for three remote sensing images show that the proposed PSSSD outperforms the existing state-of-the-art SRM methods. [ABSTRACT FROM AUTHOR]

In this paper, alternative operational mode (AOM) of Si-based blocked impurity band (BIB) detectors is thoroughly investigated. The basic structure of the detector evolves from the prototype of the ion-implanted BIB detector, which is potentially compatible with CMOS processes. Benefiting from lower dark current, the detectivity of ion-implanted devices studied in this work is comparable with the Si BIB detectors ever reported. For ion implantation devices, a peak blackbody detectivity of ${5} \times {10}^{{12}}$ cm $\cdot $ Hz1/2/W (background temperature T = 40 K) occurs at 2 V under AOM. These merits make the ion-implanted device a powerful candidate for a wide range of applications from infrared to terahertz band. By analyzing the carrier distribution and the band structure in different functional regions of the device at the measured temperature, depletion junction formed in the region of active layer near active layer/cathode interface side is deduced when the device is operated under AOM. The theory is successfully used to interpret experimental data, including the detectivity comparable to the epitaxial device, significantly reduced dark current. [ABSTRACT FROM AUTHOR]

A brain-computer interface (BCI) provides a novel non-muscular communication pathway for individuals with severe neuromuscular diseases. BCI systems based on steady-state visual evoked potentials (SSVEPs) have high classification accuracy, information transfer rate, and signal-to-noise ratio, giving them high research and application value. However, SSVEP-based BCI has several limitations in real-world applications. The main challenge is how to reduce or eliminate the need for a dedicated training process while maintaining high classification accuracy. Filter bank canonical correlation analysis (FBCCA) is a powerful and widely used feature extraction method for SSVEP-based BCI systems. However, the reference signals of FBCCA are fixed-frequency sine-cosine waves, which makes it difficult to accurately describe the complex, mutative, and individually different physiological SSVEPs. Therefore, there is huge room for improvement in classification performance based on the FBCCA method. In contrast, although spatiotemporal beamforming (BF) detects SSVEPs with high accuracy, it needs an additional training process, which limits its application. In this study, we propose a bimodal decoding algorithm (FBCCA+BF), which combines the advantages of the training-free classification of FBCCA and the data-driven and adaptive features of BF. Six-channel SSVEP data corresponding to eight targets measured from 15 subjects were used to test the effectiveness of three different CCA-based methods, BF, and our proposed FBCCA+BF methods. It was found that the classification accuracies for BF and FBCCA+BF are 95.6% and 92.2%, respectively, which are significantly higher than the other CCA-based methods. Notably, both BF and FBCCA+BF obtain state-of-the-art performance, but FBCCA+BF does this without the need for a dedicated training process. Therefore, we conclude that our proposed FBCCA+BF method provides a training-free and high-accuracy approach for SSVEP-based BCIs. [ABSTRACT FROM AUTHOR]

This paper proposes a modulation reference compensation method to completely eliminate the intercoupling between the output active and reactive powers of modular multilevel converters (MMCs). Different from the existing methods, it is realized not based on an accurate control model, but by directly amending the initial arm modulation references of MMCs. Therefore, the proposed method is essentially a compensation algorithm, which is extremely easy to implement in a microcontroller with low calculation burden. In addition, it does not need any additional closed-loop controllers or voltage/current sensors, which is very beneficial to practical projects. The proposed method finally realizes accurate and fully decoupled active and reactive power regulation, avoids overload operation and protection malfunction, and significantly improves the dynamic performance of MMCs. The effectiveness and accuracy of the proposed analysis and compensation method are carefully verified by both simulation and experimental results. [ABSTRACT FROM AUTHOR]

Recently, hyperspectral image (HSI) classification approaches based on deep learning (DL) models have been proposed and shown promising performance. However, because of very limited available training samples and massive model parameters, DL methods may suffer from overfitting. In this paper, we propose an end-to-end 3-D lightweight convolutional neural network (CNN) (abbreviated as 3-D-LWNet) for limited samples-based HSI classification. Compared with conventional 3-D-CNN models, the proposed 3-D-LWNet has a deeper network structure, less parameters, and lower computation cost, resulting in better classification performance. To further alleviate the small sample problem, we also propose two transfer learning strategies: 1) cross-sensor strategy, in which we pretrain a 3-D model in the source HSI data sets containing a greater number of labeled samples and then transfer it to the target HSI data sets and 2) cross-modal strategy, in which we pretrain a 3-D model in the 2-D RGB image data sets containing a large number of samples and then transfer it to the target HSI data sets. In contrast to previous approaches, we do not impose restrictions over the source data sets, in which they do not have to be collected by the same sensors as the target data sets. Experiments on three public HSI data sets captured by different sensors demonstrate that our model achieves competitive performance for HSI classification compared to several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Partial discharge (PD) is the main sign of the insulation deterioration; therefore, the online PD localization of the power transformers can provide the equipment degradation characteristics and has great value for the power grid. Because of some unique merits, acoustic PD detection is one of the best options for such applications. However, when discharges happen inside the windings, precise PD localization by acoustic method becomes very challenging. As the acoustic PD signal can be reverberated and distorted along its propagation, the type and number of sensors, with their installation, should be carefully studied and selected. In this paper, the acoustic wave propagation inside the transformer is explored via numerical simulation. Based on the understanding of wave propagation, a novel-structure fiber-optic acoustic sensor array is properly designed and installed into one phase winding, which is from a real 35 kV transformer. Experimental verification shows that, by using this proposed design, PD localization with less than 5-cm error can be achieved. Therefore, the novel sensor array developed in this paper, together with the installation and localization method, unveils a novel access to accurate online PD localization, especially for those PDs which happen inside transformer windings. [ABSTRACT FROM AUTHOR]

Photovoltaic (PV) generators have continuously increased in recent years, whose power is usually controlled through the perturbation and observation (P&O) method. In essence, the P&O method is nonlinear and discontinuous. Hence, the conventional small-signal stability analysis is not suitable anymore when the influence of the P&O-based power control is considered. Focusing on this problem, this paper adopts the nonlinear describing function (DF) method to conduct the accurate stability analysis of PV generators with consideration of P&O-based power control. The detailed procedures about the DF method are introduced, and then the related influence factors like perturbation size, filters, and so on are analyzed quantitatively. Furthermore, the comparison with the conventional stability analysis methods is made, which suggests that the DF method can effectively enhance the accuracy of the stability analysis. All the conclusions are verified by the real-time hardware-in-loop tests. [ABSTRACT FROM AUTHOR]

Direct modulation principle is widely used in modular multilevel converters (MMCs) for its simplicity and natural stability. However, it results in strong active and reactive power coupling. This paper carefully analyzes and identifies four power coupling paths and two power influence factors with the direct modulation principle. Full power decoupling of the MMC can be realized by cutting off the derived coupling paths and compensating the influence factors. However, this will further deteriorate the already poor damping of the MMC and significantly increase dc-bus current fluctuations. To solve this problem, the sources resulting in poor damping are identified and then a new closed-loop dc-bus current controller is incorporated into the control system. Finally, the proposed power decoupling strategy with the newly designed current controller realizes completely decoupled power regulation, avoids overload operation, and obviously improves the dynamic performance of the MMC. Furthermore, it increases the stability of the MMC system under all possible operation conditions. The proposed control strategy is confirmed by experimental results. [ABSTRACT FROM AUTHOR]

To improve the interception performance in missile guidance applications, this paper proposes a novel control strategy for the integrated guidance and control design subject to input constraints. The system constraints are translated into a restriction in an invariant set, which is then enforced by manipulating the derivative of the applied reference. For the integrated system in the presence of disturbances and parametric uncertainties, disturbance observers are applied to estimate the lumped disturbances existing in each loop. The quadratic stability via dynamic surface control is guaranteed, and an ellipsoidal approximation of the invariant set is obtained, which can be enlarged to allow some degree of the input saturation. The effectiveness of the proposed scheme is illustrated using numerical simulations against a maneuvering target. [ABSTRACT FROM AUTHOR]

The digital control system of modular multilevel converters (MMCs) normally synchronizes with the switching actions. The switching frequency of an MMC is generally selected as low as possible from an efficiency point of view, which leads to an unavoidable and relatively long time delay in the control system. Such a time delay might deteriorate the MMC performance and even reduce the stability margin by introducing infinite uncertain eigenvalues into the system state functions. Without a proper time-delay model and quantitative system stability analyses, MMCs might suffer from uncertainties and unstable risks during their operations, especially when the switching frequency is low. This paper derives an MMC model considering the time delay in the digital control system. A Padé-based stability analysis method is then proposed, which consists of three steps: information integration and critical eigenvalue extraction, and trajectory visualization and critical angle (CA) calculation, and quantitative assessment and extreme value determination. By the proposed model and method, the critical eigenvalue can be effectively extracted. With the Hopf bifurcation, the time delay in the digital control system induces a pair critical conjugate eigenvalues into MMC systems, which is detected as an instability cause. The CA and the critical impact are defined to quantitatively assess the stability, while selecting the controller parameters and minimum switching frequency. The proposed method for the MMC stability assessment is experimentally verified. [ABSTRACT FROM AUTHOR]

Modern aircrafts are evolving toward more electric aircraft (MEA), resulting in greater reliance on the electrical system for safe flight. On-board power system of MEA integrates a large number of power electronic converters, and it is reported that semiconductor devices and electrolytic capacitors in power converters are the most vulnerable links impacted by loading conditions; thus, reliability becomes a critical concern in an MEA power system. This paper proposes a hierarchical approach for systematic reliability modeling and evaluation for the on-board power system of MEAs. It consists of three hierarchical levels (HLs): component level (HL1), subsystem level (HL2), and system level (HL3). In HL1, failure rates of power electronic components are modeled considering relevant inner structure and loading conditions; in HL2, the reliability of individual subsystems such as converters are constructed; in HL3, the system reliability is quantified based on the network architecture and reliability of the subsystems. The impacts of different parts (components/subsystems) on the overall system are assessed effectively with the identification of the vulnerable parts. This also provides a guideline for reliability enhancement by using thermal control techniques, adding redundancies or performing maintenance on the vulnerable parts to ensure the satisfactory of system reliability requirements. The proposed method is demonstrated on the future MEA power system architectures (hybrid ac–dc architecture and HVdc architecture). [ABSTRACT FROM AUTHOR]

The subpixel mapping (SPM) technique can handle coarse fractional images derived by unmixing coarse original hyperspectral (HS) image to produce a fine land cover map at the subpixel scale. A popular SPM approach is a two-step model. It first increases the spatial resolution of coarse fractional images by subpixel sharpening to produce fine fractional images and then assigns class labels to each subpixel by the class allocation method. However, there is only a single processing path of the current SPM algorithm, and the information type of the fine fractional images is not rich. To enrich the information type, SPM based on dual processing paths (DPP) is proposed. DPP contains two processing paths, namely spatial–spectral path and multiscale path. First, the coarse original HS image and the high spatial resolution multispectral image are fused by component substitution to produce the fine fractional images with more spatial–spectral information in the spatial–spectral path. At the same time, deep Laplacian pyramid networks are used to obtain the fine fractional images with multiscale information in the multiscale path. The fine fractional images from the two paths are then integrated to generate the improved fraction images with multiscale spatial–spectral information. Finally, the multiscale spatial–spectral information is utilized to allocate class labels by the class allocation method. Experimental results on three real HS remote sensing data show that the proposed DPP outperforms the other SPM methods, demonstrating the effectiveness of the use of DPP in enriching the information type of the fine fractional images. [ABSTRACT FROM AUTHOR]

This paper introduces a compensation strategy to deal with both current and voltage measurement errors in the three-phase pulsewidth modulated rectifier system. The dc offset and scaling errors in the voltage and current measurements cause the injection of undesired dc and unbalanced currents into the three-phase input current and subsequently lead to voltage ripple at the dc output voltage. This paper proposes a compensation scheme for current measurement error where the dc offset and scaling errors in the current measurement are estimated from the characteristic of the dc output voltage ripple combining with simple band-pass and low-pass filters. Meanwhile, an advanced current controller designed with a proportional integral plus two resonant controllers tuned at the fundamental grid frequency $({{\omega _s}})$ and $2{\omega _s}$ in the synchronous (d–q) reference frame is suggested to reject the impact of the dc offset and scaling errors in the voltage measurement. The proposed compensation method is developed without the need of extra hardware circuit, sensor, or precise information of system parameters so that it can be considered as more cost-effective and robust solution. The effectiveness of the proposed solution is verified by experimental results. [ABSTRACT FROM AUTHOR]

The rapidly developing data center industry results in a large amount of energy consumption. Considering its unique demand characteristics, it becomes desirable to improve its energy efficiency by reusing its waste heat. In this paper, an integrated energy management scheme is proposed to optimize the operation cost for a data center microgrid. The waste heat recovered from data center operation is optimally scheduled with other resources in the integrated energy management model to minimize the operation cost of data center microgrid. Taking the uncertain nature of renewable supply, power and heat demand into consideration, a two-stage stochastic programming model is formulated and analyzed under different test cases. The numerical study results demonstrate the effectiveness of the proposed integrated energy management model. [ABSTRACT FROM AUTHOR]