Search

Zero-Shot Learning (ZSL) aims to recognize unseen classes that never appear during training. Recently, generative adversarial networks (GANs) have been introduced to convert ZSL into a supervised learning problem by synthesizing unseen visual features. However, since unseen classes are never experienced for the generator during training, the synthesized unseen visual features often become heavily biased towards seen classes, or sometimes there is even no meaningful class that can be assigned to them. This is known as the bias problem. In this paper, we propose a novel method, namely Adaptive Bias-Aware GAN (ABA-GAN), to alleviate generating biased visual features. For this purpose, we build a semantic adversarial network to regularize the feature generator. Specifically, an adaptive adversarial loss is proposed to constrain the feature distributions, which avoids the generation of meaningless visual features. Meanwhile, a domain divider is presented to explicitly distinguish synthesized visual features between seen and unseen domains, such that the bias towards seen classes can be alleviated. Moreover, we propose a novel metric named bias score (BS) to explicitly quantify the degree of the strong bias. Extensive experiments on four widely used benchmark datasets demonstrate that our proposed method outperforms the state-of-the-art approaches under both ZSL and GZSL protocols. [ABSTRACT FROM AUTHOR]

Unsupervised domain adaptation (UDA) is an approach to minimizing the domain gap. Generative methods are common approaches to minimizing the domain gap of aerial images, which improves the performance of the downstream tasks, for example, cross-domain semantic segmentation. For aerial images, the digital surface model (DSM) is usually available in both the source domain and the target domain. Depth information in DSM brings external information to generative models. However, little research utilizes it. In this letter, depth-assisted ResiDualGAN (DRDG) is proposed where depth supervised loss (DSL) and depth cycle consistency loss (DCCL) are used to bring depth information into the generative model. Experimental results show that DRDG reaches state-of-the-art accuracy between generative methods in cross-domain semantic segmentation tasks. Source code is available at https://github.com/miemieyanga/ResiDualGAN-DRDG. [ABSTRACT FROM AUTHOR]

This paper presents a new approach for synthesizing a novel street-view panorama given a satellite image, as if captured from the geographical location at the center of the satellite image. Existing works approach this as an image generation problem, adopting generative adversarial networks to implicitly learn the cross-view transformations, but ignore the geometric constraints. In this paper, we make the geometric correspondences between the satellite and street-view images explicit so as to facilitate the transfer of information between domains. Specifically, we observe that when a 3D point is visible in both views, and the height of the point relative to the camera is known, there is a deterministic mapping between the projected points in the images. Motivated by this, we develop a novel satellite to street-view projection (S2SP) module which learns the height map and projects the satellite image to the ground-level viewpoint, explicitly connecting corresponding pixels. With these projected satellite images as input, we next employ a generator to synthesize realistic street-view panoramas that are geometrically consistent with the satellite images. Our S2SP module is differentiable and the whole framework is trained in an end-to-end manner. Extensive experimental results on two cross-view benchmark datasets demonstrate that our method generates more accurate and consistent images than existing approaches. [ABSTRACT FROM AUTHOR]

Characterizations of families of fuzzy implication functions are a necessary step to fully understand the behaviour of the members of the family, their potential applicability and their relations with other families. These characterizations are based on additional algebraical properties that completely define the family. Recently, the class of power based implications was characterized through, among others, the property I (x , y) ⋅ I (y , z) = I (x , z) in a concrete sub-domain. This property, called in the literature also as multiplicative Sincov's equation, was uncommon to other families, and it was studied in-depth in a previous article. This equality is generalized in this paper by understanding the internal product as the product t-norm and changing it to a general arbitrary continuous Archimedean t-norm. This additional property is analyzed jointly with the weak ordering property or the ordering property, leading to a characterization of those fuzzy implication functions satisfying both additional properties. [ABSTRACT FROM AUTHOR]

We propose a unified game-theoretical framework to perform classification and conditional image generation given limited supervision. It is formulated as a three-player minimax game consisting of a generator, a classifier and a discriminator, and therefore is referred to as Triple Generative Adversarial Network (Triple-GAN). The generator and the classifier characterize the conditional distributions between images and labels to perform conditional generation and classification, respectively. The discriminator solely focuses on identifying fake image-label pairs. Theoretically, the three-player formulation guarantees consistency. Namely, under a nonparametric assumption, the unique equilibrium of the game is that the distributions characterized by the generator and the classifier converge to the data distribution. As a byproduct of the three-player formulation, Triple-GAN is flexible to incorporate different semi-supervised classifiers and GAN architectures. We evaluate Triple-GAN in two challenging settings, namely, semi-supervised learning and the extremely low data regime. In both settings, Triple-GAN can achieve excellent classification results and generate meaningful samples in a specific class simultaneously. In particular, using a commonly adopted 13-layer CNN classifier, Triple-GAN outperforms extensive semi-supervised learning methods substantially on several benchmarks no matter data augmentation is applied or not. [ABSTRACT FROM AUTHOR]

In recent years, remarkable progress in zero-shot learning (ZSL) has been achieved by generative adversarial networks (GAN). To compensate for the lack of training samples in ZSL, a surge of GAN architectures have been developed by human experts through trial-and-error testing. Despite their efficacy, however, there is still no guarantee that these hand-crafted models can consistently achieve good performance across diversified datasets or scenarios. Accordingly, in this paper, we turn to neural architecture search (NAS) and make the first attempt to bring NAS techniques into the ZSL realm. Specifically, we propose a differentiable GAN architecture search method over a specifically designed search space for zero-shot learning, referred to as ZeroNAS. Considering the relevance and balance of the generator and discriminator, ZeroNAS jointly searches their architectures in a min-max player game via adversarial training. Extensive experiments conducted on four widely used benchmark datasets demonstrate that ZeroNAS is capable of discovering desirable architectures that perform favorably against state-of-the-art ZSL and generalized zero-shot learning (GZSL) approaches. Source code is at https://github.com/caixiay/ZeroNAS. [ABSTRACT FROM AUTHOR]

Conditional Generative Adversarial Networks (cGANs) have enabled controllable image synthesis for many vision and graphics applications. However, recent cGANs are 1-2 orders of magnitude more compute-intensive than modern recognition CNNs. For example, GauGAN consumes 281G MACs per image, compared to 0.44G MACs for MobileNet-v3, making it difficult for interactive deployment. In this work, we propose a general-purpose compression framework for reducing the inference time and model size of the generator in cGANs. Directly applying existing compression methods yields poor performance due to the difficulty of GAN training and the differences in generator architectures. We address these challenges in two ways. First, to stabilize GAN training, we transfer knowledge of multiple intermediate representations of the original model to its compressed model and unify unpaired and paired learning. Second, instead of reusing existing CNN designs, our method finds efficient architectures via neural architecture search. To accelerate the search process, we decouple the model training and search via weight sharing. Experiments demonstrate the effectiveness of our method across different supervision settings, network architectures, and learning methods. Without losing image quality, we reduce the computation of CycleGAN by 21×, Pix2pix by 12×, MUNIT by 29×, and GauGAN by 9×, paving the way for interactive image synthesis. [ABSTRACT FROM AUTHOR]

This article is concerned with the problem of the $H_{\infty }$ output feedback control for a class of event-triggered networked systems subject to multiple cyber attacks. Two dynamic event-triggered generators are equipped at sensor and observer sides, respectively, to lower the frequency of unnecessary data transmission. The sensor-to-observer (STO) channel and observer-to-controller (OTC) channel are subject to deception attacks and Denial-of-Service (DoS) attacks, respectively. The aim of the addressed problem is to design an output feedback controller, with the consideration of the effects of dynamic event-triggered schemes (DETSs) and multiple cyber attacks. Sufficient condition is derived, which can guarantee that the resulted closed-loop system is asymptotically mean-square stable (AMSS) with a prescribed $H_{\infty }$ performance. Moreover, we provide the desired output feedback controller design method. Finally, the effectiveness of the proposed method is demonstrated by an example. [ABSTRACT FROM AUTHOR]

Bipedal robots are gaining significant attention due to their wide range of applications in various fields such as manufacturing, transportation, rehabilitation, and many others. However, mimicking human-like walking in bipedal robots is a challenging task as human walking represents a complex learning process that requires the coordination of the human brain motor cell with its leg muscles. Besides, maintaining walking stability in bipedal robots is another significant task that has not been fully explored. To address the challenges mentioned above and improve the performance of bipedal robot walk, this work proposes a long short-term memory (LSTM)-based human gait trajectory generator. A low-cost and markerless motion capture sensor system (Microsoft Kinect V2) is used to capture human gait data during a treadmill walk. The captured data is then analyzed and utilized to extract the significant features using the proposed model. Furthermore, the selection of significant hyperparameter(s) has also been focused on achieving the optimal performance for the proposed model using the Rao-3 metaheuristic optimization algorithm (optimized-LSTM). The stability performance of the generated gait trajectories is verified through standard evaluation measures like cyclograms and phase plots. The gait trajectories obtained through the proposed model are also validated on the HOAP-2 robot simulator. [ABSTRACT FROM AUTHOR]

The healthy operation conditions of wind turbines (WTs) with insufficient data need attention, but they face the problems of data imbalance and lack of labels. Aiming at the condition monitoring (CM) of these WTs, a CM method based on transfer learning and one-class classification (OCC) is proposed. This method uses the source WT data to help learn information about monitoring data from the target WT. First, the data of the target and source WTs are preprocessed to construct a training set. Second, to improve the TrAdaBoost algorithm for the OCC task, a novel weighting method based on an autoencoder (AE)—an unsupervised one-class classifier—is designed to assign weights for samples in the source and target domain dynamically and then the CM model of the target WT is established. Finally, its effectiveness is verified by using the monitoring data to detect the generator fault of the target WT. The comparison with the nontransfer method shows that the proposed method can significantly reduce the number of false alarms and can issue early warnings. [ABSTRACT FROM AUTHOR]

Inspired by an electromagnetic generator containing multi-pole radial ring magnets, we designed a bipolar electret rotatory power generator (e-RPG) with a multi-phase core-shell structure to boost output performance and achieve high-precision self-sensing capabilities. A selectively localized rotary corona discharging method is invented to uniformly inject both positive and negative charges into a single electret’s curved surface. The experimental results show that the output power of the bipolar-charged e-RPG is increased to 384% compared with those of positive or negative charging alone, which is consistent with the results obtained from the theoretical model. By introducing the core-shell-structured multi-phase electrode design, both the rotation direction and speed can be precisely detected. The frequency sensitivity and linearity $\text{R}^{2}$ (coefficient of determination) are up to 3.0 rpm/Hz and 99.9%, respectively. The proposed e-RPG has potential applications in the Internet of Things, self-powered sensing and self-sensing of rotational motions. [ABSTRACT FROM AUTHOR]

Recent research applies patent autoclassification using machine learning to map the technological landscape of an industry value chain. However, when these methods are applied to emerging industries, the available patent sample data are small-scale and unevenly distributed, which cause overfitting and reduce the accuracy of patent classification. Therefore, this article proposes a framework to map the technological landscape of an emerging industry value chain through patent analysis with deep learning, which integrates a generative adversarial network as a data-augmentation method to overcome the problem of low-quality emerging-industry patent samples, and a deep neural network as a patent classifier. Based on this framework, this article conducts an application case of the 3-D printing industry. The evaluation results show that the integrated framework can effectively classify the patents with small-scale and unevenly distributed sample data, and depict the technological landscape of an emerging industry value chain. This article develops an efficient, reliable framework for patent autoclassification of emerging industries to overcome the lack of high-quality training samples, and it sheds light on the emerging industry value chain analysis with deep learning. [ABSTRACT FROM AUTHOR]

The recent success of learning-based algorithms can be greatly attributed to the immense amount of annotated data used for training. Yet, many datasets lack annotations due to the high costs associated with labeling, resulting in degraded performances of deep learning methods. Self-supervised learning is frequently adopted to mitigate the reliance on massive labeled datasets since it exploits unlabeled data to learn relevant feature representations. In this work, we propose SS-StyleGAN, a self-supervised approach for image annotation and classification suitable for extremely small annotated datasets. This novel framework adds self-supervision to the StyleGAN architecture by integrating an encoder that learns the embedding to the StyleGAN latent space, which is well-known for its disentangled properties. The learned latent space enables the smart selection of representatives from the data to be labeled for improved classification performance. We show that the proposed method attains strong classification results using small labeled datasets of sizes 50 and even 10. We demonstrate the superiority of our approach for the tasks of COVID-19 and liver tumor pathology identification. [ABSTRACT FROM AUTHOR]

Designing single-board computers (SBCs) is becoming more challenging given the growing number of discrete components that are made available and the rate at which this number grows. Keeping track of all available components options, revisions, and functionalities is challenging for SBC designers who are striving for faster design cycles. Moreover, the procedure of deciding peripheral components, their values, and connections of an SBC is not only difficult because of various parameters that need to be considered but also is time consuming as there exist numerous components on a typical SBC nowadays. In this article, an SBC generator tool, FASCINET, is presented that uses a neural network (NN) model to design customized peripheral circuits for SBCs. The tool creates a large commercial off-the-shelf database (COTS DB) of existing components, efficiently searches through them, and selects optimal components for both main and peripheral components based on the user’s requirements. Creating such a broad COTS DB requires processing abundant datasheets. A manual approach is time consuming, even if only a fraction of all available datasheets is considered. In order to automate this process, this article describes a novel NN-based approach for automatically categorizing datasheets and proposes an extraction technique for parsing relevant functional information from tables within. Our evaluation using a test set that contains over 770 000 components shows that the category of datasheets is identified correctly over 95% of the time. Additionally, the table extractor has a precision above 96%. Our proposed fully autonomous SBC design approach reduces the time for generating the schematic of an SBC to as little as 2 min. For validating the accuracy of our model, the netlists of 400 SBCs designed by FASCINET are compared to the human-designed versions. This evaluation shows that FASCINET is able to design SBCs that are identical to the manually designed ones except for minor differences. [ABSTRACT FROM AUTHOR]

Phase sequence exchange (PSE) is a recently developed emergency control technology in which power electronic devices are used to switch the three-phase sequence, thereby reducing the power angle of the generator by 120° and preventing the system from losing stability. This article proposes the installation of the PSE device in the fragile cut-set line of the power grid, and to use the local measurement information for control. This method can restore the stability of the system when the power grid oscillates, thereby avoiding the collapse of the island power grid after splitting due to oscillation. First, an improved real-time power angle calculation method is proposed. When a system loses synchronization and the power angle of a line reaches the threshold, the PSE device on the same cut-set line acts simultaneously to prevent the system from falling out-of-step and “splitting” from the cut set. A PSE prototype is used for experimental verification on the real-time digital simulator platform. The results demonstrate that the proposed power angle calculation method greatly reduces the error, and the proposed PSE control method can achieve online monitoring and real-time control, which achieves better control effects than conventional methods. [ABSTRACT FROM AUTHOR]

In this article, a voltage-fed isolated matrix-type ac/dc converter (VF-MC) is proposed for wind energy conversion system (WECS). Instead of the line-frequency transformer, the high-frequency transformer (HFT) is used for isolation between the generator and the grid. The direct matrix converter (MC) is employed to connect the wind power generator and the HFT, and the intermediate dc capacitors are saved in the ac/ac conversion. Moreover, the proposed VF-MC can work without use of filter capacitors on generator side, and enable more compact size. The dedicated space-vector-modulation scheme and commutation method are proposed for safe operation of the MC. The voltage spikes due to the leakage inductance of HFT are thus reduced and the additional snubber circuit is avoided in the proposed system. A 1-kW laboratory prototype of the proposed VF-MC-based WECS is built to verify the effectiveness of the system experimentally. [ABSTRACT FROM AUTHOR]

The current source converter (CSC) based conversion system is a good candidate for next-generation series-connected wind energy conversion systems. In this article, a new power converter is proposed for the CSC-based series-connected wind system. In addition, three versions of the proposed converter are developed. Compared with the existing CSC-based converters, the proposed one provides a practical solution for the series-connected wind system, achieves superior current harmonic performance, and features inherent power balancing, simple control, and high scalability, in addition to keeping all the advantages of the existing CSCs. The operation principle of the proposed converter is discussed, and its performance is verified by the simulation and lab-scaled experiments. [ABSTRACT FROM AUTHOR]

Just-in-time learning (JITL) is a widely used method for online soft sensing. The limitation of available data and the increase of sample dimensions will make the historical dataset sparse, seriously impair the reliability and usability of JITL applications in the industry. To this end, data augmentation technology (DA) is introduced into the JITL framework for the first time to improve the performance of JITL-based soft sensors. In this article, a novel causality-informed variational autoencoder (CIVAE) is developed to generate virtual samples to augment the historical dataset. Based on this, a complete framework of data augmentation just-in-time learning (DA-JITL) is formulated offline and online and implemented with two different strategies. Finally, a numeral example and an actual industrial example are applied to verify the effectiveness of the proposed method. Besides, the effect of virtual data on JITL and the influence of virtual data volume, virtual data ratio and other factors are carefully discussed on the proposed method. In the industrial case, two indicators (rmse and R2) of the proposed method have been improved by an average of 25% and 17%, respectively, compared to the traditional JITL approach. [ABSTRACT FROM AUTHOR]

DC microgrid as a microscale power system has drawn growing attention for its various applications, which calls for good power quality, proper power sharing, and fast response property. However, the existing researches are not yet able to achieve the purpose in a predefined time. In this article, a new predefined-time secondary controller is proposed for DC microgrid to realize the objectives of both voltage regulation and current sharing among DGs within a predefined time. The controller is designed by employing a sign function and a $K^1$ function and by defining a new composite error, which does not need to sample the voltage of DC bus or the current of neighbors, but only the local current and voltage. The upper bound of the convergence time of both the voltage and current in the DC microgrid is just related to one parameter which can be predefined and is independent of initial error. The effectiveness of the proposed controller is verified by multiple simulations and experiments built on MATLAB/Simulink and a hardware-in-the-loop experimental platform. The conducted simulations and experiments include several typical scenarios together with a comparison, which illustrate the advantages of the proposed controller such as fast convergence rate and small overshoot. [ABSTRACT FROM AUTHOR]

In this article, a novel radial–axial (RA) flux switching permanent magnet generator (FSPMG) is introduced. In contrast to other RA flux machines, a major advantage of this generator is its implementation only by the use of conventional nonoriented electrical steel sheets, even though the flux linkage of the novel RA-FSPMG has a 3-D path. Besides that, it benefits from the flux concentration strategy and a passive rotor. A concrete foundation of the novel RA-FSPMG is provided, in which, the 3D-Flux path is described, working principles are discussed, and power-sizing equation is derived. No-load characteristics and the full-load performance of an optimized RA-FSPMG are investigated in contrast to an optimized double stator radial FSPMG and an optimized double stator axial FSPMG with similar volume, number of permanent magnets (PMs), and electrical loading. The results show the competitiveness of the novel RA-FSPMG, in particular, in the output power and the required copper mass. Furthermore, the PM demagnetization possibility and the rotor axial pull force of the optimized RA-FSPMG are investigated, in detail. All discussions are supported by the 3-D finite element analysis and experimental results. [ABSTRACT FROM AUTHOR]

Electrical equipment of the floating nuclear power plant operates in a long-term vibration environment. Ground faults occur frequently and are often accompanied by arc, a reliable arc suppression algorithm for the generator stator winding is necessary. In this paper, an active arc suppression method is proposed by connecting a controllable voltage source to the generator neutral point. Through regulating the neutral point voltage to be equal in amplitude and opposite in phase to the fault potential, the fault point voltage can be clamped to a low level to achieve arc suppression. Since the generator's phase potential contains certain harmonic components, the controllable voltage source includes fundamental and third-order harmonic components. To accurately calculate the output voltage, online calculation methods for the fault location, fault potential and grounding capacitance are proposed. The accuracy of the fault location method is verified by simulation and dynamic test. And simulation results based on PSCAD/EMTDC software show that the proposed method can achieve effective arc suppression at different ground fault conditions. [ABSTRACT FROM AUTHOR]

The stator ground fault location of large generators can guide rapid maintenance. In this paper, a location criterion without fault transition resistance is constructed by using the fundamental potential and the third harmonic potential of the faulty winding simultaneously. The above potentials of the faulty winding are corresponding to the fault position. To precisely locate the grounding fault, it is necessary to clarify the analytical relationship between the faulty winding potential and the fault position. However, the existing methods often take the coil potential as the unit to analyze the winding potential, which has theoretical errors for the generators with short-pitch winding. Aiming at this problem, a more precise analysis method for winding potential is established by taking the slot conductor potential as the unit. To solve the location criterion, the optimization model for each slot conductor of the faulty phase is constructed. The slot conductor with the smallest optimal objective function value is judged as the faulty slot conductor, and the corresponding decision variable is judged as the fault position. Simulation and experiments verify the effectiveness of the proposed method, which can still meet the needs of engineering applications under multiple adverse conditions. [ABSTRACT FROM AUTHOR]

With the increase of photovoltaic penetration, the power-electronic converter (PEC)-interfaced generators with traditional control strategy may cause severe power oscillations after being disturbed due to lack of inertia and damping. The photovoltaic power plant controlled by virtual synchronous generator (VSG) technology, while introducing virtual inertia and virtual damping to improve the stability of the system, will also inherit its oscillation characteristics and cause a new oscillation mode. Aiming at this problem, this paper analyzes the difference between the system low-frequency oscillation (LFO) modes before and after the VSG is connected to the two-area system by the modal analysis. It is pointed out that the PEC-interfaced renewable energy generators based on VSG control technology could change the oscillation mode of the original system. The influence of VSG's main parameters and capacity on system dynamics is analyzed. Finally, a synchrophasor-based inter-area oscillation suppressed comprehensive control strategy is proposed to suppress the LFO of active power on the tie line, and the superiority of the strategy is verified by DIgSILENT /PowerFactory. [ABSTRACT FROM AUTHOR]

The risk of unstable subsynchronous oscillations (SSO) in DFIG-based wind parks radially connected to transmission lines with series capacitive compensation is typically assessed with numerous electromagnetic transient (EMT) simulations, which is a time-consuming process. However, not all operating scenarios present risk of SSO and require such a detailed investigation. This paper proposes two charts based on impedance equivalents of wind generators obtained through frequency scans to quickly map which combinations of parameters can create unstable SSOs and which combinations are risk-free. With the information, engineers can immediately filter out scenarios with no risk of instability and conduct detailed studies on fewer scenarios that are truly critical. These charts can be obtained with simple equations that use only information readily available in practice to engineers, without the need to run any computer simulation. The charts also reveal the key operating conditions that can increase the risk of an unstable SSO and provide real-time quantitative insights into how close a wind park is to an instability. Application examples are provided to illustrate how this approach can help engineers speed up studies to identify the risk of SSO. [ABSTRACT FROM AUTHOR]

Geomagneticallyinduced current (GIC) studies usually focus on transformer response to dc but seldom analyze generator excitation systems. A GIC is driven by an electric field which is induced by a very low-frequency geomagnetic field. It is widely acceptable to model GICs with dc, mostly suitable for static voltage stability analysis, but this does not adequately characterize a power system's dynamic response. Considering this, we studied a multi-machine power system on an OPAL-RT digital real-time simulator (DRTS)-MATLAB/Simulink system. We used real GIC measurement data from high and mid-latitude regions. Wavelet analysis related the frequency components of the GIC to the power system's response. A dynamic generator model with excitation voltage control modelled the source. Our main finding is that excitation system control contributes to reducing the voltage drop caused by increased var demand in saturated transformers with GICs. Measured GICs revealed non-negligible short-term dynamics affecting bus voltages that are not captured with dc-modelling assumptions, yet they are important for voltage stability analysis. This study implies that excitation control creates a buffer for high GICs to flow without violating the grid codes for minimum operating voltages and this paper demonstrates the optimization of this novel approach. [ABSTRACT FROM AUTHOR]

Metal-oxide arresters (MOA) are widely used to protect equipment against transient electromagnetic disturbances (TED) like very fast transient overvoltage (VFTO), high-altitude electromagnetic pulse (HEMP), lightning pulse, et al. Researches on the performance of MOA under microsecond level rise-time TED are abundant, such as lightning pulse. However, when MOA is subjected to TED of rise-time at nanosecond level, the voltage/current responses of MOA are not fully studied due to the lack of a standardized and reliable test setup. In this paper, a test setup is proposed for acquiring the response behavior of MOA under nanosecond level pulse. The setup is comprised of a pulse generator, a two-branch transmission line system and a high voltage power divider. The generator can provide stable double exponential pulse with tunable rise-time ranging from 5 ns to 100 ns. The divider ensures an even distribution of voltage/current surge wave along the two-branch line. Through the numerical analysis, both impinging and residual surge waveforms could be retrieved from measurements. The proposed method has been applied to investigate response behavior of MOA subjected to nanosecond level rise-time pulses. [ABSTRACT FROM AUTHOR]

An integrated generator–rectifier architecture is a promising candidate for harvesting energy from offshore wind. The most attractive feature of this architecture is that the majority of the power is processed by reliable, efficient, and inexpensive diodes operating at the generator frequency. The use of dc collection grids (medium-voltage or high-voltage dc) is an emerging trend in a wind farm. This article proposes a dc grid interface circuit for the integrated generator–rectifier and a control strategy for maximum power point tracking. Using a 10-MW illustrative design, it is shown that the proposed architecture reduces the total switch volt–ampere rating requirement by 22.8%, which translates into overall loss reduction ranging between 28.3% and 71.7%, depending on the extracted power. The maximum power point tracking algorithm is implemented based on the relationship between the active rectifier $d$ -axis current and the generated power. Experimental results obtained from a laboratory prototype validates the effectiveness of the proposed converter architecture and its control strategy. This approach opens up opportunities for integrating dc collection networks with offshore wind farms through an efficient, reliable, and cost-effective energy conversion system. [ABSTRACT FROM AUTHOR]

The wind–thermal-bundled transmission system is a feasible way to transmit wind power generation; however, the stability of the system should be paid more attention under high wind power penetration. In this article, the impact of wind energy penetration on torsional vibration and electrical characteristics is investigated for wind–thermal-bundled transmitted by high-voltage direct current system. The torsional vibration responses of generator shaft with different wind–thermal-bundled ratios are analyzed by time and frequency methods. The electrical characteristics of wind–thermal-bundled transmission system are disclosed under the influence of the wind farm penetration. The analysis illustrates that the penetration of doubly fed induction generation makes the torsional vibration of turbo generator and wind turbine shaft present different mode characteristics, and the risks of different modes are different. The current and voltage of wind–thermal-bundled transmission system show regular evolution with the change of wind–thermal-bundled proportion. Analytical results provide the value for the design and control of the wind–thermal-bundled transmitted system, which contributes to the improvement of power system stability. [ABSTRACT FROM AUTHOR]

A novel compensation control method for heterogeneous multiagent systems under Denial-of-Service (DoS) attacks and transmission delays is investigated in this article. This control method has all the advantages of the cloud-based computation strategy, the adaptive event-triggered strategy, and the predictive control scheme. The adaptive event-triggering mechanism can adjust the event numbers adaptively, the predictive control can reduce or eliminate the negative effects brought out by both DoS attacks and transmission delays actively, while the cloud-based computation strategy can eliminate the negative effects completely as the same as there are no DoS attacks and transmission delays. Through the interval decomposition skill and the augmented system modeling method, the compensated geschlossenes system model is established. Moreover, the joint design for the feedback gain matrices and the event-triggered parameters is implemented. In the simulation part, five VTOL aircraft are used to demonstrate the theoretical results. [ABSTRACT FROM AUTHOR]

Health monitoring of current complex systems significantly impacts the total cost of the system. Centralized fault diagnosis architectures are sometimes prohibitive for large-scale interconnected systems, such as distribution systems, telecommunication networks, water distribution networks, or fluid power systems. Confidentiality constraints are also an issue. This article presents a decentralized fault diagnosis method that only requires the knowledge of local models and limited knowledge of their neighboring subsystems. The method, implemented in the decentralized diagnoser design ($D^{3}$) algorithm, is based on structural analysis and can advantageously be applied to high-dimensional systems, linear or nonlinear. Using the concept of isolation on request, a hierarchy is built according to diagnostic objectives. The resulting diagnoser is based on analytical redundancy relations (ARRs) generated along the hierarchy. Their number is optimized via binary integer linear programming (BILP) while still guaranteeing maximal diagnosability at each level. $D^{3}$ proves of lower time complexity than its centralized equivalent. It is successfully applied to a nonlinear combined cycle gas-turbine power plant. [ABSTRACT FROM AUTHOR]

The work of Helmer [Divisibility properties of integral functions, Duke Math. J. 6(2) (1940) 345–356] applied algebraic methods to the field of complex analysis when he proved the ring of entire functions on the complex plane is a Bezout domain (i.e. all finitely generated ideals are principal). This inspired the work of Henriksen [On the ideal structure of the ring of entire functions, Pacific J. Math. 2(2) (1952) 179–184. On the prime ideals of the ring of entire functions, Pacific J. Math. 3(4) (1953) 711–720] who proved a correspondence between the maximal ideals within the ring of entire functions and ultrafilters on sets of zeroes as well as a correspondence between the prime ideals and growth rates on the multiplicities of zeroes. We prove analogous results on rings of analytic functions in the non-Archimedean context: all finitely generated ideals in the ring of analytic functions on an annulus of a characteristic zero non-Archimedean field are two-generated but not guaranteed to be principal. We also prove the maximal and prime ideal structure in the non-Archimedean context is similar to that of the ordinary complex numbers; however, the methodology has to be significantly altered to account for the failure of Weierstrass factorization on balls of finite radius in fields which are not spherically complete, which was proven by Lazard [Les zeros d'une function analytique d'une variable sur un corps value complet, Publ. Math. l'IHES 14(1) (1942) 47–75]. [ABSTRACT FROM AUTHOR]

Linear codes achieving or nearly achieving the Singleton bound are interesting in both theory and practice. The objective of this paper is to construct several infinite families of MDS, near MDS and almost MDS codes from some special cyclic subgroups of ${\mathbb {F}}_{q^{2}}^{*}$. To this end, the augmentation and extension techniques are used. The codes in this paper have flexible parameters and their lengths could be large. The minimum linear locality of the codes constructed in this paper is also studied. Some infinite families of optimal linearly locally recoverable codes are obtained. Besides, some codes in this paper are proved to be proper for error detection. [ABSTRACT FROM AUTHOR]

The binary quadratic-residue codes and the punctured Reed-Muller codes ${\mathcal {R}}_{2}((m-1)/2, m))$ are two families of binary cyclic codes with parameters $[n, (n+1)/2, d \geq \sqrt {n}]$. These two families of binary cyclic codes are interesting partly due to the fact that their minimum distances have a square-root bound. The objective of this paper is to construct two families of binary cyclic codes of length $2^{m}-1$ and dimension near $2^{m-1}$ with good minimum distances. When $m \geq 3$ is odd, the codes become a family of duadic codes with parameters $[2^{m}-1, 2^{m-1}, d]$ , where $d \geq 2^{(m-1)/2}+1$ if $m \equiv 3 \pmod {4}$ and $d \geq 2^{(m-1)/2}+3$ if $m \equiv 1 \pmod {4}$. The two families of binary cyclic codes contain some optimal binary cyclic codes. [ABSTRACT FROM AUTHOR]

This article presents a new self-powered temperature sensor based on thermoelectric generators (TEGs). Two different types of TEGs measure temperature and supply the interface circuit. The interface circuit calculates and generates an output current proportional to the cold-surface temperature of the TEGs. Since TEG output voltage depends on the temperature difference between the cold and hot surfaces, a three-terminal TEG has been used. A combination structure of the two PbTe and Bi2Te3 TEGs are thermally in parallel and electrically in series. The new method uses a machine-learning algorithm and Seebeck curve fitting for temperature calculation. For the cold surface temperature range of 0 °C to 70 °C, the error is less than ±1 °C. [ABSTRACT FROM AUTHOR]

Aiming at the problem of low prediction accuracy of traditional prediction models due to the limited labeled sample data and the imbalance of multitimescale sample data in the cement production process, a cement clinker-free calcium oxide (f-CaO) data enhancement and prediction model based on semisupervised prediction Wasserstein generative adversarial networks (SSP-WGANs) is proposed in this article. The model is constructed by WGAN and a prediction model. In the generator of WGAN, the traditional noise input is replaced by time-series matrices composed of related variables affecting the f-CaO content of cement clinker. The generator maps the input high-dimensional time-series data into low-dimensional labeled values through its internal convolutional layer, which can fill in the missing labeled values of the input unlabeled samples. The generated labels are spliced with the related variables affecting the f-CaO content of clinker according to the timescale relationship and are uniformly sent to the discriminator with the real sample pairs for discrimination, thus eliminating the influence of multiple timescales. The generated labeled data are matched with the unlabeled data and used to expand the training set of the prediction model composed of CNN-GRU, which significantly improves the prediction accuracy of the model. The results show that the SSP-WGAN model with data enhancement has higher accuracy and stronger robustness. [ABSTRACT FROM AUTHOR]

In this brief, a novel event-triggered intermittent-based impulsive control (ETIIC) strategy is designed to stabilize a class of nonlinear systems, which possesses the strengths of event-triggered control, intermittent control and impulsive control. Particularly, ETIIC allows that impulsive instants appear only at the control intervals. Moreover, the control intervals can be activated by mean of the well-designed event-triggered mechanism that depends explicitly on threshold-value and check-period. A sufficient criterion is provided to ensure exponential stabilization of nonlinear systems via ETIIC, meanwhile, the non-Zeno behavior is confirmed in our results. At last, an example with numerical simulation is presented to illustrate the feasibility of ETIIC and the effectiveness of our results. [ABSTRACT FROM AUTHOR]

This brief studies the stabilization of load frequency control (LFC) system. The event-triggered intermittent control (ETIC) is designed for the multi-area LFC system. The stabilization of the LFC system is achieved under the designed ETIC. Moreover, the scheme of ETIC is improved via state-based control width (SCW) and an ETIC-SCW is proposed. Two performance indices: the rate of control and the cost of control, are defined for comparisons. Specific cases are studied to verify the effectiveness of the proposed ETIC schemes. From the numerical simulations, it is shown that the designed ETIC, especially ETIC-SCW, achieves clearly lower rate of control and cost of control than the time-based intermittent control and the event-triggered control for LFC system in the literature. [ABSTRACT FROM AUTHOR]

In this brief, a 60 GHz variable gain phase shifter based on body floated RF-DAC structure is presented, which is fabricated in 65 nm bulk CMOS process. The proposed variable gain phase shifter consists of I/Q generator and RF-DAC structure based quadrature vector summing amplifier. A low loss and wide band I/Q generator is proposed, which combines a hybrid coupler and a RC poly-phase filter. The vector summing amplifier adapts a pair of RF-DAC structures for each quadrature signal which turn on and off complementarily. This makes the input and output impedances invariant to control states, which are very important characteristics to reduce RMS gain and phase errors of the phase shifter. The RF-DAC cell is implemented with body floated FETs to reduce the Cds at on and off cells for high gain. The RMS gain and phase errors of the proposed phase shifter show 0.13 dB and 0.88°, respectively, at 60 GHz with 3-bits gain control in 8 dB dynamic range, and 5-bits phase control in 360°. In the frequency range of 57 GHz to 62.7 GHz, RMS gain and phase errors show 0.39 dB and 3.01°, respectively. The chip area and overall power consumption are measured to be 0.71 mm2 and 36 mW, respectively. [ABSTRACT FROM AUTHOR]

Comprehensive benchmarking of clustering algorithms is rendered difficult by two key factors: 1) the elusiveness of a unique mathematical definition of this unsupervised learning approach and 2) dependencies between the generating models or clustering criteria adopted by some clustering algorithms and indices for internal cluster validation. Consequently, there is no consensus regarding the best practice for rigorous benchmarking, and whether this is possible at all outside the context of a given application. Here, we argue that synthetic datasets must continue to play an important role in the evaluation of clustering algorithms, but that this necessitates constructing benchmarks that appropriately cover the diverse set of properties that impact clustering algorithm performance. Through our framework, HAWKS, we demonstrate the important role evolutionary algorithms play to support flexible generation of such benchmarks, allowing simple modification and extension. We illustrate two possible uses of our framework: 1) the evolution of benchmark data consistent with a set of hand-derived properties and 2) the generation of datasets that tease out performance differences between a given pair of algorithms. Our work has implications for the design of clustering benchmarks that sufficiently challenge a broad range of algorithms, and for furthering insight into the strengths and weaknesses of specific approaches. [ABSTRACT FROM AUTHOR]

Due to the lack of systematic empirical analyses and comparisons of ideas and methods, a clearly established state of the art is still missing in the optimization-based design of robot swarms. In this article, we propose an experimental protocol for the comparison of fully automatic design methods. This protocol is characterized by two notable elements: 1) a way to define benchmarks for the evaluation and comparison of design methods and 2) a sampling strategy that minimizes the variance when estimating their expected performance. To define generally applicable benchmarks, we introduce the notion of mission generator: a tool to generate missions that mimic those a design method will eventually have to solve. To minimize the variance of the performance estimation, we show that, under some common assumptions, one should adopt the sampling strategy that maximizes the number of missions considered—a formal proof is provided as the supplementary material. We illustrate the experimental protocol by comparing the performance of two offline fully automatic design methods that were presented in previous publications. [ABSTRACT FROM AUTHOR]

A generator of classes of multidimensional test problems for benchmarking continuous constrained global optimization methods is proposed. It is based on the generator of test classes for global optimization proposed in 2003 by Gaviano, Kvasov, Lera, and Sergeyev and extends the previous generation procedure from the box-constrained case to the case of nonlinear constraints. The user has the possibility to fix the difficulty of tests in an intuitive way by choosing several types of constraints. A detailed information (including the global solution) for each of 100 problems in each generated class is provided to the user. The generator is particularly suited for testing black-box optimization algorithms that normally address low or medium dimensional problems with hard to evaluate objective functions. [ABSTRACT FROM AUTHOR]

To deal with low-voltage ride-through (LVRT) and to enhance the transient performance of a grid-connected wind-driven permanent-magnet synchronous generator during voltage dips on the grid side, this study presents a revised control structure for wind-power converters, comprising a machine-side converter (MSC) and grid-side converter (GSC). In the proposed approach, the control variables’ references are modified with grid voltage, and the revised control designs for the MSC and GSC are established. During voltage dips, the captured wind power is stored as the kinetic energy of the turbine rotor. The active component of the stator current is curtailed according to the dip level by terminating the maximum power tracking operation. The modified GSC current references assist the grid in providing the required reactive current and attempt to minimize the power loss by utilizing the maximum GSC current-carrying capacity. The revised controls are responsible not only for maintaining the DC-link voltage and GSC current within safe limits, but also support the grid in providing reactive power during voltage recovery. Simulations verify the suitability and effectiveness of the proposed design in handling LVRT operations and providing additional flexibility by incorporating stability and security constraints.

Wind power plants do not provide frequency support subject to disturbances because wind turbine generators (WTGs) are decoupled from system frequency deviations. To provide frequency regulation from a permanent-magnet synchronous generator-based wind energy conversion system following the change in system load demand, this paper presents a basic controller that directly perturbs the rotor-speed reference according to system frequency deviations. In the proposed approach, the kinetic energy (KE) of the rotating parts of the WTGs is utilized for frequency support, thus enriching the power system stability. The prominent aspect of opting for this concept is the immediate KE release or absorption through a step-change in the speed reference. Furthermore, the system proceeds to its normal operation without causing instability issues. To prove this concept, extensive simulations in MATLAB/Simulink have been performed, and the results demonstrate its good capability in providing system frequency regulation, resulting in the enhancement of power system stability. For comparison, simultaneous control of the power-frequency droop control and DC-link inertial support was also prepared.

La confiabilidad y disponibilidad de hidrogeneradores depende en gran medida de la condición del aislamiento del devanado de estator, que está expuesto a una combinación de esfuerzos térmicos, eléctricos, ambientales y mecánicos. Una interrupción no programada de una máquina en funcionamiento podría poner en riesgo la estabilidad del sistema eléctrico, así como provocar pérdidas económicas, por lo que es necesario contar con técnicas de monitoreo de condición, como es la medición de descargas parciales en-línea. Una descarga parcial se define como una descarga eléctrica localizada, que solo une parcialmente el aislamiento entre los conductores, acompañada de movimiento de cargas, emisión acústica, radiación y reacción química, entre otros efectos. La medición de descargas parciales en-línea es una de las técnicas más utilizadas para seguir la evolución de grandes hidrogeneradores. En este artículo se propone una metodología novedosa para detectar los mecanismos de falla de descargas parciales: i) en ranura; y, ii) por el establecimiento de chispas por vibración, mostrando ejemplos utilizando una máquina en servicio. La metodología se basa en la medición de descargas parciales en-línea para distintos niveles de potencia activa. Lo anterior permite encontrar dependencias térmicas y mecánicas, que logren localizar y caracterizar el mecanismo de falla. Posteriormente, se establecen una serie de posibles causas-raíz de cada uno, y las recomendaciones de reparación. Los resultados obtenidos en las pruebas realizadas muestran que la medición de descargas parciales en-línea es efectiva cuando se trata de detección temprana de fallas repentinas, y se ha aplicado con éxito en el hidrogenerador evaluado.

Radar-based human activity recognition (HAR) has been applied in many fields such as human–computer interaction, smart surveillance, and health assessment. With the development of deep learning, many deep-learning models have been proposed in radar-based HAR to achieve high classification accuracy. Inevitably, new human activity classes appear continuously, and this requires the deep-learning models to efficiently learn novel categories from only a few data samples while at the same time maintaining high accuracy on the initial human activities on which they were trained. Therefore, this article proposes a category-extensible HAR model based on few-shot learning. The classification weight of the new human activity category is quickly generated through the feature vector extracted by the trained feature extractor and the classification weight of the most similar original category. To solve the mismatching of weight value intervals, we also employ the cosine similarity-based classifier. Furthermore, we adopted the large margin in the softmax cross-entropy (LMSC) loss function to improve the model’s performance and depthwise separable convolution to reduce the computation of the model. The experimental results show that the model can quickly adapt to the new human activity classification task and did not sacrifice the performance on category classification of the original activities. With relatively low computational complexity, our method achieves 81.99% of accuracy when each extended category has only one sample to generate the weights and 91.60% when each of them has only five samples in the task of a three-way. [ABSTRACT FROM AUTHOR]

In this article, we propose a shared-aperture metasurface (MS) combining spin-decoupled strategy to generate four coaxially circularly polarized (CP) beams carrying orbital angular momentum (OAM) at Ku-band and Ka-band. First, the frequency selective surface (FSS) is adopted to suppress the mutual coupling between the Ku-band and the Ka-band. Next, at each band, the spin-decoupled meta-atom is designed to realize the independent control for dual circular polarizations. Subsequently, the designed MS illuminated by CP feed source would generate left-hand CP (LHCP) beam with OAM mode +1 and right-hand CP (RHCP) beam with OAM mode −1 at Ku-band; meanwhile, the LHCP beam with OAM mode +2 and the RHCP beam with OAM mode −2 are generated at Ka-band. Finally, the MS is designed, manufactured, and measured. Both simulated and measured near-field and far-field performance verify that the proposed MS can generate four independent CP OAM beams along the same spatial axis at dual bands. On the other hand, the experimental results and the corresponding theoretical analysis verify that the proposed MS illuminated by a linear-polarized (LP) feed source could also generate eight correlative LP OAM beams at dual bands. [ABSTRACT FROM AUTHOR]

This article investigates practical implementation aspects of a novel hybrid chamber concept that has been recently introduced for over-the-air (OTA) testing of wireless devices. The chamber allows to synthesize a spectrum of plane waves (PWs) incident on a device under test (DUT) from a wide range of angles of arrival (AoAs) through exploiting multiple field scattering inside the overmoded waveguide (WG) chamber. An optimal design approach for the plane wave spectrum generator (PWSG) array inside the chamber is presented considering: 1) array antenna impedance matching for the desired test zone (TZ) quality and AoAs’ range; 2) an operational frequency bandwidth of the test environment and an instantaneous signal bandwidth; and 3) the impact of PWSG excitation errors. A robust linearly constrained minimum variance (LCMV) beamformer is formulated to realize the desired test environment over a wide frequency band, while accounting for the varying number of the WG propagating modes as excited by PWSG array. Numerical simulations and measurements with the first prototype of the hybrid chamber for FR1 frequency band (~3.5 GHz) are presented. This prototype includes a ${7\times 7}$ bowtie antenna element PWSG array, a 1.00 m $\times1.25$ m $\times1.75$ m metal WG chamber, an off-line optimal array beamforming setup, and a planar scanner for TZ characterization. [ABSTRACT FROM AUTHOR]

Human beings tend to learn unknown knowledge in a gradual process, from the basic to the complex. Based on this point, we propose a progressive learning method for producing service strategies according to requests, with a hierarchical priori knowledge and reinforcement learning. Service strategy aims to guide how to perform home services and takes into consideration the relationship between actions and objects in home environment. In this paper, strategy generation is regarded as a text generation problem in question answering (QA). Firstly, a hierarchical priori knowledge with service-object correlation at the bottom and action-object correlation at the top is constructed to assist the understanding on the relationship of objects and actions in service strategies. Service-object correlation guides how to select proper objects with the correct order, while action-object correlation associates actions in strategies according to selected objects. Based on the hierarchical priori knowledge, a progressive learning method is proposed to make the model produce effective strategies with a sequential cognition, from service-object correlation (objects) to action-object correlation (actions). After that, reinforcement learning is employed to enhance the progressive guidance, by designing rewards in terms of the hierarchical priori knowledge. Finally, the proposed method is tested with both comparative experiments and ablation studies, and the experimental results demonstrate the superiority in producing comprehensive and logical strategies, indicating that the progressive learning method in our paper can further improve the QA performance. [ABSTRACT FROM AUTHOR]