Search

The calibration procedure for a wearable P300 brain-computer interface (BCI) greatly impact the user experience of the system. Each user needs to spend additional time establishing a decoder adapted to their own brainwaves. Therefore, achieving subject independent is an urgent issue for wearable P300 BCI needs to be addressed. A dataset of electroencephalogram (EEG) signals was constructed from 100 individuals by conducting a P300 speller task with a wearable EEG amplifier. A framework is proposed that initially improves cross- subject consistency of EEG features through a common feature extractor. Subsequently, a simple and compact convolutional neural network (CNN) architecture is employed to learn an embedding sub-space, where the mapped EEG features are maximally separated, while pursuing the minimum distance within the same class and the maximum distance between different classes. Finally, the model’s generalization capability was further optimized through fine-tuning. Results: The proposed method significantly boosts the average accuracy of wearable P300 BCI to $73.23\pm 7.62$ % without calibration and $78.75\pm 6.37$ % with fine-tuning. The results demonstrate the feasibility and excellent performance of our dataset and framework. A calibration-free wearable P300 BCI system is feasible, suggesting significant potential for practical applications of the wearable P300 BCI system.

The analysis of the reluctance coil launcher (RCL) basically relies on the finite element method (FEM), but the method is time-consuming and does not meet the real-time requirements for control, which seriously hampers the expansion of engineering application areas. To address this problem, this study proposes a numerical calculation method dedicated to RCL based on the mathematical equations of electromagnetic coupling, in which the magnetic saturation problem is corrected by the inductance lookup table method, and the inductance in the coupling circuit is updated in real time according to the instantaneous current and the instantaneous position of the projectile. The accuracy of the calculation method is verified and the reasons for the errors are analyzed by establishing finite element simulation model, MATLAB/Simulink model and experimental platform for a single-stage and a three-stage RCL and comparing the obtained data. When the discharge voltage of the capacitor is 300 V, the accuracy of the projectile muzzle velocity in the single-stage and three-stage is 98.42% and 98.20% respectively, which indicates that the calculation method is correct under specific conditions. The time consumed by the method is only about 1/1000 of that of FEM, which even includes the compilation time of the software MATLAB/Simulink, illustrating the advantages of the algorithm and its application prospects.

Reluctance coil launchers usually use a pulse current formed by capacitive discharge to excite the excitation coil when launching. The pulse current peaks are high, and the projectile usually undergoes deep magnetic saturation, resulting in low launching efficiency. To address this problem, this article proposes using distributed feeder circuit to drive reluctance coil launchers, replacing the traditional single set of large-capacity capacitor discharge scheme with multiple small-capacity capacitors discharged in a time-sequential manner, so as to reduce the peak current and suppress magnetic saturation problems. The study includes two main aspects, one is to study the effect of the driving method on the launch performance and analyze the reasons; the other is to propose the calculation of the time-varying data in the launch process by the time-segmented calculation method. The efficiency of the model described in the paper is only 1.301% when using a single high-capacity capacitor for launching, while the launching efficiency is increased to 3.535% when driven by the distributed feeder circuit, indicating that the method can effectively improve the launching performance.

In the field of aerial photography, complete coverage path planning (CCPP) is the key to improving the task efficiency of unmanned aerial vehicles (UAVs) with a limited battery capacity. Previous CCPP algorithms consisted of Cellular Decomposition (CD) class, Probabilistic Roadmap (PRM) class, and Bio-inspired Neural Network (BINN) class. However, when faced with the complex environment of multiple forbidden areas, CD and PRM have high computational complexity and a long path length, and BINN has a large number of turns. In this paper, an improved back-and-forth (IBF) algorithm is proposed to achieve good performance in these three aspects at the same time based on a newly straight-turn mode and the corresponding auxiliary approaches (i.e. Local Traversal, Global Traversal, and continuity constraint function). An important difference between straight-turn mode and back-and-forth mode is that the next straight direction is not restricted to being parallel to the last straight direction but is determined by Local Traversal, which is introduced to evaluate the benefits of different paths and find the best direction for the next straight-turn mode at the ending cell. Global Traversal is available to choose a “jump” cell when “Dead zone” are faced. The continuity constraint function is used to maintain the continuity of the unexplored area. Under the comprehensive consideration of path length, computation time, and number of turns, the simulation and experimental results indicate that the proposed IBF algorithm has better performance in multiple forbidden areas.

Objective: A brain-computer interface (BCI) can be used to translate neuronal activity into commands to control external devices. However, using noninvasive BCI to control a robotic arm for movements in three-dimensional (3D) environments and accomplish complicated daily tasks, such as grasping and drinking, remains a challenge. Approach: In this study, a shared robotic arm control system based on hybrid asynchronous BCI and computer vision was presented. The BCI model, which combines steady-state visual evoked potentials (SSVEPs) and blink-related electrooculography (EOG) signals, allows users to freely choose from fifteen commands in an asynchronous mode corresponding to robot actions in a 3D workspace and reach targets with a wide movement range, while computer vision can identify objects and assist a robotic arm in completing more precise tasks, such as grasping a target automatically. Results: Ten subjects participated in the experiments and achieved an average accuracy of more than 92% and a high trajectory efficiency for robot movement. All subjects were able to perform the reach-grasp-drink tasks successfully using the proposed shared control method, with fewer error commands and shorter completion time than with direct BCI control. Significance: Our results demonstrated the feasibility and efficiency of generating practical multidimensional control of an intuitive robotic arm by merging hybrid asynchronous BCI and computer vision-based recognition.

The rapid and reliable detection of oil contaminants is of crucial importance for environmental protection. This paper focused on the classification and identification of six types of oil using unconventional steady-state fluorescence spectroscopy coupled with XY-fused networks (XY-Fs). The excitation emission matrix fluorescence (EEMF) spectroscopy, synchronous fluorescence spectroscopy (SFS) and total synchronous fluorescence spectroscopy (TSFS) of the oil samples were measured by a FS920 steady-state fluorescence spectrometer. Further, the obtained results were compared with those obtained by using partial least square discriminant analysis (PLS-DA), LDA based on score matrix of multivariate curve resolution-alternating least squares (MCR-ALS-LDA) and parallel factor (PARAFAC-LDA) analysis. The experimental results showed that the best accuracy (100%) for the test samples was obtained by TSFS combined with XY-Fs. These results provide important references for discrimination of oil contaminants.

A brain-computer interface (BCI) measures and analyzes brain activity and converts it into computer commands to control external devices. Traditional BCIs usually require full calibration, which is time-consuming and makes BCI systems inconvenient to use. In this study, we propose an online P300 BCI spelling system with zero or shortened calibration based on a convolutional neural network (CNN) and big electroencephalography (EEG) data. Specifically, three methods are proposed to train CNNs for the online detection of P300 potentials: (i) training a subject-independent CNN with data collected from 150 subjects; (ii) adapting the CNN online via a semisupervised learning/self-training method based on unlabeled data collected during the user’s online operation; and (iii) fine-tuning the CNN with a transfer learning method based on a small quantity of labeled data collected before the user’s online operation. Note that the calibration process is eliminated in the first two methods and dramatically shortened in the third method. Based on these methods, an online P300 spelling system is developed. Twenty subjects participated in our online experiments. Average accuracies of 89.38%, 94.00% and 93.50% were obtained by the subject-independent CNN, the self-training-based CNN and the transfer learning-based CNN, respectively. These results demonstrate the effectiveness of our methods, and thus, the convenience of the online P300-based BCI system is substantially improved.

There are possible non-convergence problems at load power critical value while using conventional continuous power flow methods. Generally, the step control method is used in the calculation to perform the calculation in small steps at the convergence to the load power critical value, and this method sacrifices the speed of the calculation although the calculation accuracy is high. In this paper, the Levenberg-Marquardt algorithm is used to search for the load power critical value firstly, and after that the step control algorithm is designed to perform fast power flow finding with known load power critical value, which takes into account the computational accuracy and computational efficiency. The case shows that the fast calculation method for continuous power flow of microgrid based on Levenberg-Marquardt algorithm proposed in this paper can guarantee the accuracy while effectively improving the calculation speed, and is suitable for online real-time evaluation of voltage stability and preventive control aided decision online calculation of isolated microgrid with high requirements for real-time.

The soft-switching characteristics of double-sided LCC wireless power transmission systems are susceptible to system parameter disturbances. In order to improve the stability of zero voltage switches in constant-current and constant-voltage charging modes, this paper synthesizes eight resonant conditions which make the system realize constant-current/constant-voltage and Zero Phase Angle. The influence of compensating capacitance parameter disturbance on system impedance angle is studied under constant current/constant voltage operation mode. The parameter configuration methods of compensated capacitance realized by zero voltage switches under the influence of third-order harmonic in constant current/constant voltage operation mode are also given. The system’s experimental prototype verifies the theoretical analysis’s accuracy with capacitive and inductive input impedance.

Building change detection (CD) from remote sensing images (RSI) has great significance in exploring the utilization of land resources and determining the building damage after a disaster. This article proposed an attention-based multiscale input–output network, named AMIO-Net, for building CD in high-resolution RSI. It is able to overcome partial drawbacks of existing CD methods, such as insufficient utilization of information (details of building edges) of original images and poor detection effect of small targets (small-scale buildings or small-area changed buildings that are disturbed by other buildings). In AMIO-Net, the input image is scaled down to different sizes, and performed the convolution to extract features. Then, the feature maps are fed into the encoding stage so that the network can fully utilize the feature information (FI) of the original image. More importantly, we design two attention mechanism modules: the pyramid pooling attention module (PPAM) and the Siamese attention mechanism module (SAMM). PPAM combines a pyramid pooling module and an attention mechanism to fully consider the global information and focus on the FI of changed pixels in the image. The input of SAMM is the parallel multiscale output diagram of the decoding portion and deep feature maps of the network so that AMIO-Net can utilize the global contextual semantic FI and strengthen detection ability for small targets. Experiments on three datasets show that the proposed method achieves higher detection accuracy and F1 score compared with the state-of-the-art methods.

Fourier ptychography (FP) is a computational imaging technique with the advantage that it can obtain large field-of-view (FOV) and high-resolution (HR) imaging. We propose an algorithm for Fourier ptychography based on reweighted amplitude flow (RAF) with regularization by denoising (RED) and deep decoder (DD), which is an untrained deep generative model. The proposed method includes two loops, using reweighted amplitude flow with regularization by denoising as an inner loop for phase retrieval and deep decoder for further denoising as an outer loop in the Fourier ptychography recovery system. The proposed method does not need any training dataset, just adds a little computer time during the image recovery process. The proposed method has no bias due to training images, which is different from other deep learning methods. The experimental results show that the proposed method can improve the reconstruction quality in both PSNR and SSIM.

Zhengzhou and its surrounding areas, located in northern Henan Province, China, receive continuous extreme rainfall from July 17 to July 22, 2021. Northern Henan Province experiences extensive flash floods and urban floods, causing severe casualties and property damage. Understanding the variation of hydrologic features during this flood event could be valuable for future flood emergency response work and flood risk management. This study first demonstrates the rainstorm process based on near-real-time precipitation data from the China Meteorological Administration Land Data Assimilation System (CLDAS-V2.0). To meet the temporal resolution required for monitoring this short-term flood event, reconstructed daily terrestrial water storage anomalies (TWSAs) based on GRACE and GRACE-FO data and CLDAS-V2.0 datasets are first introduced. The spatial and temporal evolution of the reconstructed daily TWSA is analyzed in the study area during this heavy rainfall event. We further employ a wetness index based on the reconstructed daily TWSA for flood warnings. Furthermore, the modeled soil moisture data and daily runoff data are used for flood monitoring. Results show that the reconstructed daily TWSA increases by 437.7 mm in just six days (from July 17 to July 22, 2021), with a terrestrial water storage increment of 9.4 km3. Compared with ITSG-Grace2018, the reconstructed daily TWSA has better potential for near-real-time flood monitoring for short-term events in a small region. The wetness index derived from reconstructed daily TWSA is potential for flood early warning.

Climate change has caused an increase in the frequency of flood events. Rapid and accurate flood mapping is essential for disaster monitoring and risk assessment. The normalized difference flood index (NDFI) is a change detection method with the characteristics of efficient processing and less manual intervention, which can quickly obtain flood information. However, the NDFI method would misclassify some permanent water bodies in lakes and rivers into floods. We presented a framework by combining NDFI calculated from synthetic aperture radar images and a summer permanent water bodies (SPWB) exclusion layer derived from optical remote sensing surface reflectance data, abbreviated as NDFI-SPWB. This framework was further verified by the flood event in the Yangtze river basin in July 2020. Results show that the NDFI-SPWB framework can increase the user accuracy by approximately 10% and the Kappa coefficient by approximately 0.08 compared with the original NDFI method, which verifies the feasibility and effectiveness of the proposed framework.

Current multimodal learning in smart feature extraction and classification has reshaped the landscape of remote sensing. The recent developments in smart feature extraction mainly rely on different machine learning and data mining algorithms as powerful classifiers to improve prediction accuracy. This article presents an innovative ensemble learning algorithm for integrated data and classifier fusion via higher order singular value decomposition (IDCF-HOSVD). Based on the fused data, different analytical, semiempirical, and empirical classifiers can be selected and applied to perform information retrieval that can be further synergized via a tensor flow based feature extraction scheme over the classifier space. When preserving core fused image patterns via HOSVD, the final step of IDCF-HOSVD helps rank the contributions from different classifiers via nonlinear correlation information entropy. Practical implementation of the IDCF-HOSVD algorithm was assessed through its application to map the seasonal water quality conditions in Lake Nicaragua, Central America.

To solve the problem of difficult analysis and modeling of the strongly nonlinear characteristics of a proportional solenoid (PS) valve drive component, namely, the PS in engineering, a simplified mathematical model was established using a power bond graph, and the PS was numerically analyzed by Maxwell Simulation. After verifying the accuracy of the simulation model through experimental data, the static and dynamic output characteristics of the PS under different working conditions were analyzed in an auxiliary manner in the simulation model. Finally, the parameters of the mathematical model built were identified based on the principle of least squares (LSQ). The results show that the goodness of fit between the identified mathematical model and the static output characteristics of the PS reached 0.9866, and the relative error of the dynamic characteristics was 2%. The established model can provide engineers and technicians with a method to quickly analyze the dynamic characteristics of the PS and provide a basis for designing the control strategy of the PS’s current and force.

Plant and scene three-dimensional (3-D) modeling, combined with radiative transfer (RT) modeling, are of great importance for mastering canopy reflectance characteristics and further developing target recognition and parameter retrieval in remote sensing images. However, 3-D RT simulation of large, complex landscapes is generally too demanding in terms of computing time and memory space. Simplifying plant models can significantly reduce the computational load, but with the accuracy reducing in radiation simulations. It is necessary to balance the complexity of plant models and the efficiency of 3-D RT simulation while maintaining high simulation accuracy. We investigated this issue for the vegetation of the Yangtze River estuary in eastern China. First, we used a series of created 3-D models of two species (Phragmites australis and Spartina alterniflora) to simulate canopy reflectance with the discrete anisotropic radiative transfer (DART) model. Then, we investigated how the simulated plant model complexity, plant density, and scene unit scale influence the accuracy and computation time of canopy normalized difference vegetation index (NDVI) simulation. The comparison of different parameterization simulations leads to three major conclusions. It is not necessary to simulate the actual vegetation density exactly, given the simplifications and approximations inherent in simulations. A specific 3-D model per species is needed for simulation since plants’ morphological structures different. Simplifying plant 3-D models and using a coarser DART scale of analysis shortens simulation time, but decreases the accuracy of the simulated canopy NDVI to varying degrees. Based on these results, we propose a universal optimization scheme that balances the accuracy and computation time of canopy NDVI simulation.

This paper aims at developing a data-driven optimal control strategy for virtual synchronous generator (VSG) in the scenario where no expert knowledge or requirement for system model is available. Firstly, the optimal and adaptive control problem for VSG is transformed into a reinforcement learning task. Specifically, the control variables, i.e., virtual inertia and damping factor, are defined as the actions. Meanwhile, the active power output, angular frequency and its derivative are considered as the observations. Moreover, the reward mechanism is designed based on three preset characteristic functions to quantify the control targets: (1) maintaining the deviation of angular frequency within special limits; (2) preserving well-damped oscillations for both the angular frequency and active power output; (3) obtaining slow frequency drop in the transient process. Next, to maximize the cumulative rewards, a decentralized deep policy gradient algorithm, which features model-free and faster convergence, is developed and employed to find the optimal control policy. With this effort, a data-driven adaptive VSG controller can be obtained. By using the proposed controller, the inverter-based distributed generator can adaptively adjust its control variables based on current observations to fulfill the expected targets in model-free fashion. Finally, simulation results validate the feasibility and effectiveness of the proposed approach.

Biochemicals, such as chlorophyll (Chl) and nitrogen (N), are closely related to photosynthesis process of vegetation. Their accurate estimation is an important topic in remote sensing of vegetation. Previous studies mainly focused on Chl-N content inversion in leaf and canopy level, and few cared about their 3-D distributions, which was also an important indicator for the growth status of vegetation (GSV). Hyperspectral LiDAR (HSL) is a novel active remote sensing technology, which has target-sensitive band with hyperspectra resolution. Its 3-D point cloud data simultaneously contains rich spectral and precise geometrical characteristics of the target. This work aims to apply HSL data on 3-D Chl-N content mapping in vegetation through constructing HSL-based spectral indices (SIs). Except for following the SI forms of previous works, the normalized differential vegetation index and ratio index (RI) with four broadbands in an HSL spectral space were successively proposed to invert Chl-N content for the whole vegetation based on the artificial neural network (ANN) method. These four broadbands were transformed based on the relative spectral response curve of detector and the feature weights (FWs) of multiwavelength, respectively. Results show that most HSL-based ANN models can accurately invert Chl-N content with a mean R2 of >0.75, and some that fusing broadband data with convolution transformation, namely the FW-based RI, can even obtain a model R2 of 0.84 for N content inversion. Thus, HSL can be efficiently applied to 3-D Chl-N content mapping of vegetation and has great potential in GSV monitoring.

There are several resonance frequencies points in the high-order resonant network to realize constant current/constant voltage output, but the disturbance of the resonant network parameters will affect the output characteristics of the system. Therefore, selecting the appropriate resonant frequency is necessary to achieve stable output characteristics of wireless power transmission. This paper analyzes the influence of sensitivity of the double-sided LCC compensation network parameters on resonance point and output characteristics. Moreover, this paper proposes a design method to realize load-independent constant current and constant voltage output under two ZPA frequencies, and the resonance parameters are optimized for the realization of ZVS. Under constant current and constant voltage output resonant points, the transmission efficiency of the double-sided LCC IPT system reaches 88.9% and 88.5%, respectively.

In practical, the Faraday effect-induced bias error is one of the main sources of bias error for the interferometric fiber optic gyroscope (IFOG). The Faraday effect-induced bias error impairs in the performance of IFOG. Normally, the Faraday effect-induced bias errors is hardly eliminated. To improve the performance of IFOG, the accurate model of Faraday effect-induced bias errors is necessary for its practical applications. In this paper, from the perspective of the finite element method, we calculate Faraday effect-induced bias errors of the fiber coil by the three-dimensional model in the radial magnetic field, and the axial magnetic field. Comparing with the traditional model, the differences mainly come from the size, and direction of the magnetic field, including the differences of the twist ratio, and the bending rate of each layer. Using the Jones matrix, we derive the three-dimensional magnetic field error model, further more theoretical results are obtained by simulation. The experiment with a 1024 m PM fiber coil are performed, and the experimental results support these theoretical predictions.

The hydraulic wind turbine(HWT) is taken as the research object. To realize the quasi-synchronized safe connection of the HWT, the mathematical model of the fixed displacement pump - variable motor is first established, and the multiplicative nonlinearity existing in the modeling process is solved by the small signal linearization method. For the closed-loop speed regulation system of fixed displacement pump-variable motor, a variable speed input-constant speed output control method based on indirect flow feedback and direct-speed closed-loop system is proposed. Combining MATLAB/Simulink@ and a semi-physical simulation experiment platform of 30kVA HWT, the influence law of some parameters is analyzed for the response characteristics of the speed control system, such as the valve control cylinder response speed, the swash plate reference conversion value, initial swash plate angle and so on. At the same time, combined with the grid connection conditions for generators, a quasi-synchronized grid speed control method for HWT is proposed. Based on simulation and experimental methods, the instantaneous impact of grid connection is effectively suppressed.

Under the background of the large-scale construction of photovoltaic (PV) power stations, it is crucial to discover and solve module failures in time for improving the service life and maintaining the normal operation efficiency of modules. Based on analyzing the difference of I-V curves of PV arrays under different fault states, the I-V curves, temperatures and irradiances are taken as input data, and a fusion model of convolutional neural network (CNN) and residual-gated recurrent unit (Res-GRU) is proposed to identify the PV array fault. This model consists of a 1-dimensional CNN module with a 4-layer structure and a Res-GRU module. It has the advantages of end-to-end fault diagnosis, no manual feature extraction, strong anti-interference ability, and usable in the absence of irradiances and temperatures. Moreover, it can not only identify a single fault (e.g., short circuit, partial shading, abnormal aging, etc.), but also can effectively identify hybrid faults. Experimental results show that the classification accuracy of the proposed method is 98.61%, which is better than the ones of the artificial neural network (ANN), the extreme learning machine with kernel function (KELM), the fuzzy C-mean (FCM) clustering, the residual neural network (ResNet), and the stage-wise additive modeling using multi-class exponential loss function based on the classification and regression tree (SAMME-CART). In addition, in the absence of temperatures and irradiances, the classification accuracy still reaches 95.23%, which has a broad application prospect in the online fault diagnoses of PV arrays.

In this research, a new method based on Active Disturbance Rejection Control (ADRC) for controlling the power output of hydraulic wind turbines (HWT) is proposed. There are three major problems in the power control system of HWT. The first problem is the existence of multiplicative nonlinearity in the system, which is an essential nonlinearity. The second problem is that the controlled system is a non-minimum phase (NMP) system. The third problem is the wind speed with strong randomness and large fluctuation range. These three problems make it difficult for HWT to achieve power control. Therefore, in this paper, the power transmission mechanism of HWT is analyzed. Then the ADRC controller combined with the linear quadratic regulator (LQR) method is designed and the stability is proved. In the process of controller design, the multiplicative nonlinear problem is regarded as the total disturbance in ADRC. Then the extended system model of ADRC and HWT is deduced, the range of the tuning parameter b0 is calculated, and the control parameters are tuned through the LQR method. Finally, the effectiveness of the proposed method is verified through simulation experiments.

Unmanned ground vehicles (UGVs) and unmanned aerial vehicles (UAVs) have been widely used in delivery. In the context of the COVID-2019, in order to control the development of the epidemic, many places have adopted measures to isolate and close the area once a confirmed case is found. While reducing the contact between people, it also blocks the normal driving of vehicles. Only by changing the traditional logistics and distribution methods can customers who have been in a closed and isolated area for a long time be served. Therefore, we use the Cooperative UGV-UAV to achieve it. In this article, when commanding cooperative UGV and UAV for emergency resource delivery, we mainly focus on two questions: how to accept the operation order (OPORD) from the commander, how to generate a nested vehicle routing planning. We first employ one intelligent task understanding module to drive the intelligent unmanned vehicles to accept and process the C-BML (Coalition Battle Management Language) formatted OPORD with 5W (what, who, where, why, when) elements. Then, we slove the nested vehicle routing planning problem as a mixed integer linear program (MILP) with the outputs of what is the UGV route, what is the UGV sortie, and how to control the customers' distribution between the UGV and the UAV. Experimental results of random instances and case study show that using the iterative improvement algorithm increase the speed rate of solving more than 10%.

The precision of conventional PV fault diagnostic methods faces challenges due to the nonlinear output power characteristics of photovoltaic (PV) arrays and the implementation of the maximum power point tracking (MPPT) algorithm. In severe cases, it may lead to power losses and even arouse safety issues. In this study, the variation characteristics of the sequence waveforms at the moment of failure are investigated and used to develop a novel PV fault diagnostic framework. Firstly, the sequence waveforms of string voltages and currents before and after the fault occurred are collected; the normalized sequence data of voltages, currents, and powers are used as analytic data. Then, the fault feature extraction is realized via a stacked autoencoder (SAE) model. After that, an improved multi-grained cascade forest (IgcForest) is proposed to diagnose faults, e.g., line-to-line (L-L) fault, open-circuit (OC) fault, partial-shading of PV arrays, etc. The advantages of the proposed method are that the SAE method to extract features with higher recognition automatically, and the IgcForest to enhance and exploit fault features. Particularly, the proposed improvements can reduce the feature vector dimension and enhance the information connectivity between forests at all levels for further improving the accuracy of diagnoses. In addition, the validity of the proposed method is verified by numerical simulations and measured data, and the corresponding accuracy of fault diagnoses for single failure reach 99.33% and 98.61%, respectively, which are superior to traditional methods, such as softmax, support vector machines, random forest, gcForest, and daForest. Furthermore, it also has a high accuracy of 98.83% for data sets with the occurrence of multiple faults.

Plug-in connector of gas insulated bus (GIB) could be subject to the wear and electrical contact resistance (ECR) deterioration processes under cyclic mechanical and current loads. A specific test platform is designed to simulate the current-carrying wear and ECR degradation process of Ag-plated GIB electrical contact spot inside SF6, N2 and Air insulation mediums under current loads. Worn surface morphology and chemical composition are observed and analyzed. Influence of insulation medium and current load on deterioration mechanism of GIB electrical contact are discussed. Raw ECR curve obtained from current-carrying experiment is smoothed by the exponentially weighted moving average (EWMA) method. A long short-term memory (LSTM) neural network is then proposed to realize both single-step and multi-step ECR and failure predictions of GIB electrical contact. Results show that adhesive wear on contact surface happened due to materials transfer. Deterioration of contact spot inside air medium is much severer than those in SF6 and N2 due to oxidation effect. Deterioration of contact spot inside SF6 medium under high current load is severer than those under low current load due to the change of wear behavior and chemical reactions. Compared with other methods, LSTM could achieve better performance in both single-step and multi-step failure prediction of GIB electrical contact.

In this paper, a coordinated control scheme for wind turbine generator (WTG) and supercapacitor energy storage system (ESS) is proposed for temporary frequency supports. Inertial control is designed by using generator torque limit considering the security of WTG system, while ESS releases its energy to compensate the sudden active power deficit during the recovery process of turbine rotor. WTG is modeled using the fatigue, aerodynamic, structure, turbulence (FAST) code, which identifies the mechanical loadings of the turbine and addresses electro-mechanical interactions in the wind energy system. A damping controller is augmented to the inertial control to suppress severe mechanical oscillations in the shaft and tower of the turbine during frequency supports. Furthermore, the result of small-signal stability analysis shows that the WTG-ESS tends to improve the stability of the whole multi-energy power grid. The major contributions of this paper will be validated by utilizing the proposed control method that combines the grid support capability and maintaining the integrity of structural design of the turbine for normal operations.

In this paper, a new fully distributed state estimation (DSE) based on weighted least square (WLS) method and graph theory is proposed for power system. The proposed method is fully distributed so that the centralized facilities, e.g., supervisory control and data acquisition (SCADA) and centralized estimators, are not required. Also, different from the existing DSE methods, the proposed method is a bus-level DSE method, in which the power system is not required to be partitioned into several areas. In order to realize the proposed fully distributed DSE method, a novel information propagation algorithm is developed in this paper. This algorithm has great potential in future applications since it is useful to broadcast the local information of the nodes to the entire system in a fully distributed network. The proposed DSE method is compared with the conventional centralized state estimation method and existing multi-area DSE method in different models in this paper. The results show that the proposed method has better performance than the traditional methods.

Substitution box (S-Box) has a prominent significance being the fundamental nonlinear component of block cipher which fulfils confusion, one of the properties proposed by Claude Shannon in 1949. In this paper, we proposed an S-Box by using the action of modular group PSL $\left ({2,\mathbb {Z} }\right)$ on projective line PL $\left ({F_{257} }\right)$ over Galois field GF $\left ({2^{8} }\right)$ . In the first step we obtained elements of GF $\left ({2^{8} }\right)$ by using powers of $\alpha $ , where $\alpha $ is the primitive root of irreducible polynomial $p\left ({x }\right)$ of order 8 over field $\mathbb {Z}_{2}$ , then applied the generators of PSL $\left ({2,\mathbb {Z} }\right)$ and followed steps to get rid of infinity from output. In the final step of proposed scheme, one of the permutations of $S_{16}$ is applied which enhanced the possible number of S-Boxes obtained by any single specific irreducible polynomial $p(x)$ over field $\mathbb {Z}_{2}$ of order 8. We analyzed performance of the proposed $8\times 8$ S-Box under cryptographic properties such as strict avalanche criterion, bit independence criterion, nonlinearity, differential approximation probability, linear approximation probability; and compared obtained results with a number of renowned S-Boxes. Lastly, we performed statistical analysis (which comprises of contrast analysis, homogeneity analysis, energy analysis, correlation analysis, entropy analysis and mean of absolute deviation analysis) on our proposed S-Box and obtained results have been compared with adequate number of S-Boxes.

Topological descriptors are the numerical quantities of a graph that characterize various structural properties of it. In environmental sciences, pharmacy and in mathematical chemistry such descriptors are used for the quantitative structure-activity and property relationships (QSAR/QSPR) studies in which physicochemical properties of compounds are correlated with their molecular structures. A large spectrum of topological descriptors is available, among which the distance-based and bond-additive indices are frequently used in QSPR/QSAR studies. In this paper, the different versions of Szeged, Padmakar-Iven (PI) and Mostar indices for the molecular graphs of two types of dendrimers having phthalocyanines and porphyrins as their cores are illustrated through the numerical way. We have obtained exact analytical expressions of these indices by using the cut method.

The piezoelectric micro-positioning platform(PMP) has advantages in high displacement resolution and fast response, however, the serious hysteresis nonlinearity in the PMP restricts its positioning accuracy. This paper presents a discretization of Krasnosel’skii-Pokrovskii model to describe the hysteresis nonlinearity of the PMP. Then, the density function is identified online by the adaptive linear neural network. To compensate the intrinsic hysteresis nonlinearity in the PMP, a feedforward compensation control method is implemented by an innovative iterative learning control algorithm. Moreover, a hybrid control method based on iterative learning and fractional order PID control is proposed to further improve the control accuracy. A series of comparative experiments are carried out to validate the feasibility and effectiveness of the proposed control method.

As a second-generation fiber optic gyro (FOG), the resonator fiber optic gyro (RFOG) has great potential in miniaturization and high-precision application. At present, white noises in the system have become important factors restricting the performance improvement of RFOG. For the first time, the shot noise, thermal noise, and relative intensity noise in the RFOG are analyzed and modeled, which are used to establish the angular random walk (ARW) model of RFOG. Based on the ARW model, an improvement method of ARW is proposed by adjusting the modulation frequency. In order to make this method effective, we use the particle swarm optimization algorithm to optimize the multi-parameter involved in ARW. Then, the experiment of optimizing modulation frequency is executed, which verifies that this method can effectively improve ARW and reduce the influence of white noises in RFOG.

Reconfigurable photonic processors, which can be programmed to perform multiple photonic processing tasks by using the same hardware platform, own the advantages of higher flexibility and more cost-effectiveness compared with application-specific photonic integration circuits (ASPICs). In this paper, we present a novel programmable photonic processor based on two-dimensional meshes of self-coupled optical waveguide (SCOW) resonant structures. The proposed processor can be configured for realizing various basic optical components, as well as cascaded and coupled components. As a proof-of-principle, we experimentally demonstrate the concept with a 3 × 1 SCOW-based processor on the silicon platform, including tunable couplers, variable optical attenuators, and phase shifters. We implement eight different configurations using the chip, including ring resonators, Mach-Zehnder interferometers, Fabry-Perot resonators, and composite structures built of these basic components. These results demonstrate that the proposed processor can be a promising candidate for multi-functional photonic processors.

We experimentally demonstrate an integrated strictly non-blocking silicon 4 × 4 optical switch chip that can be operated in both thermo-optic (TO) and electro-optic (EO) switching modes. It is based on the double-layer network (DLN) architecture and consists of twelve 2 × 2 Mach-Zehnder interferometer (MZI) switch elements. TO phase shifters based on TiN microheaters and EO phase shifters based on p-i-n diodes are embedded in both waveguide arms of the MZI elements. The power consumption for TO and EO switching is 34 mW/π and 7 mW/π, respectively. The on-chip insertion losses are 1.74 ± 0.59 dB and 3.79 dB ± 1.32 dB for TO and EO switching, respectively. Due to the merits of the DLN architecture and the optimized performance of the switch elements, the chip possesses low crosstalk of -29.1 dB and -19.4 dB for TO and EO switching, respectively. Quadrature phase-shift keying (QPSK) optical signals with a data rate of 64 Gb/s are transmitted through the switch with no observable deteriorations. Such an optical switch is a promising candidate for both optical circuit switching and optical packet switching for a variety of applications.

Distance-based metrics has been practiced to various wide-ranging physicochemical applications, especially in the characterization and modeling of chemical structures. Computational techniques to acquire the precise analytic expressions for a number of distance-based topological indices of inorganic chemical networks and nano-materials are newly growing areas of reticular chemistry. Among these numerical invariants, the eccentricity-based indices are significant for their careful prediction of pharmaceutical properties. In this paper, we work out for the closed formulas of the various eccentricity-based indices, namely the eccentric-connectivity index, the total eccentric-connectivity index, the Zagreb eccentricity indices, the augmented eccentric-connectivity index, the modified eccentric-connectivity index, and the edge version of the eccentric-connectivity index for triazine- and porphyrin-based dendrimers. Furthermore, the expressions for the eccentric-connectivity, total eccentric-connectivity, and modified eccentric-connectivity polynomials are presented for these classes of dendrimers.

The 3D reconstruction is an important topic in computer vision with many applications, such as robotics and augmented reality. Since the raw depth images captured by consumer RGB-D cameras are often low resolution (LR), noisy, and incomplete. How to obtain high-quality 3D models with a consumer RGB-D camera is still a challenge for the existing systems. In this paper, we propose a new depth super-resolution and completion method implemented in a deep learning framework and build a high-quality 3D reconstruction system. We first improve the resolution of LR depth image with a depth super-resolution network and remove the outliers in high-resolution (HR) depth image based on gradient saliency. To further enhance the quality of HR depth image with the guide of HR color image, we learn surface normal and occlusion boundary images from the corresponding HR color image through two deep fully convolutional networks. In particular, the blurriness of HR color image is also detected and pixel-wise quantized. Finally, we obtain a completed HR depth image by optimizing the HR depth image with the surface normal, occlusion boundary, and color image blurriness. We have carried out qualitative and quantitative evaluations with baseline methods on public datasets. The experimental results demonstrate that our method has better performance both on single depth image enhancement and 3D reconstruction.

The present-day trend of the numerical coding of chemical structures with topological indices (TIs) has established quite successful in medicinal chemistry and bioinformatics. This strategy provides the annotation, comparison, rapid collection, mining, and retrieval of chemical structures within large databases. Afterward, TIs can be used to look for quantitative structure-activity relationships and quantitative structure-property relationships, which are models that associate chemical structure with biological activity. In these analyses, degree-based TIs have secured a significant place among the different types of descriptors because of the ease of generation and the momentum with which these computations can be executed. In this paper, we compute the lower and upper bounds of the first, second, and third Zagreb, the first and the second multiple Zagreb, the geometric-arithmetic, the general sum connectivity, the general Randic, the atom-bond connectivity, the augmented Zagreb, and the harmonic indices of the corona product of F-sum of connected graphs in the form of their factor graphs by applying combinatorial inequalities.

With the development of RGB-D sensors and mobile devices, 3D scanning has witnessed great progress in recent years. KinectFusion opens up an era of RGB-D 3D reconstruction, which integrates captured depth images into a voxel-based representation. A number of improvements have been applied to the KinectFusion to reduce large footprint and make the 3D reconstruction on mobile devices possible. However, these methods are designed to handle static scenes. In this paper, we propose a method which can perform the 3D scanning of high dynamic scenes using an RGB-D sensor and an inertial measurement unit (IMU) on a mobile device. We first introduce a novel method to segment the depth images into static and dynamic elements with the use of sparse optical flow and the rotational part of the IMU. Then, we determine the camera pose with pixels labeled as static. At last depth, images are integrated into a voxel-based representation. The truncated signed distance function values of static voxels are updated while dynamic voxels are set to free-space. The experiments show that compared with some state-of-the-art systems, our method has better results when scanning high dynamic scenes and has comparable results when scanning low dynamic and static scenes. Besides, our method processes eight frames/s on an Apple iPad Air 2.

As key electrical equipment in the power system, the normal operation of a high-voltage circuit breaker is related to the reliability and economy of the power supply. In this paper, a mechanical fault diagnostic method for a high-voltage circuit breaker via the hybrid feature extraction and the integrated extreme learning machine (IELM) is investigated. First, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is used to decompose the vibration signal to obtain intrinsic mode functions (IMF). Then, the sub-band reconstruction of each order IMF component is performed by combining the Hilbert transform and the band-pass filter in order to obtain the time-frequency matrix. Moreover, mechanical fault feature vectors can be formed by the time-frequency entropy and the singular entropy, which are extracted by transforming the time-frequency matrix into the energy matrix and normalizing the frequency bands via the normal cumulative distribution function (NCDF). In addition, an IELM is built for the fault classification. The advantages of the proposed CEEMDAN scheme in combination with band-pass filtering can eliminate the modal aliasing, reduce the number of auxiliary noise additions, and improve the decomposition efficiency. Besides, the performance of the singular entropy normalized by the NCDF is more stable, and the IELM composed of multiple weak classifiers can solve the shortcomings of the traditional extreme learning machine. The experimental results based on measured data show the proposed method can effectively diagnose the mechanical failure via small samples of high-voltage circuit breakers.

The atom-bond connectivity (ABC) index is one of the most actively studied degree-based graph invariants that are found in a vast variety of chemical applications. This paper is devoted to establishing some extremal results regarding the variant of the ABC index, the so-called multiplicative ABC index (ABC$\Pi $ ), which, for a graph $G$ , is defined as ${{ ABC}}\Pi (G)=\prod _{uv\in E(G)}\sqrt {\frac {d(u)+d(v)-2}{d(u)d(v)}}$ . We have shown that the complete graph $K_{n}$ has a minimum ABC$\Pi $ index among connected simple graphs with $n$ vertices, while the star graph $S_{n-1}$ has the maximum ABC$\Pi $ index. As $S_{n-1}$ attains the maximum amongst bipartite graphs on $n$ vertices, we additionally show that the bipartite complete balanced graph $K_{\left \lfloor{ n/2 }\right \rfloor, \left \lceil{ n/2 }\right \rceil }$ attains the minimum in this class of graphs. As an interesting problem, we propose to characterize the trees with the minimum value of this index, and, here, we have some structural properties of these trees. We conclude this paper with few conjectures for possible further work.

Although the sparse kernel adaptive filtering algorithms have been proposed to address the problem of redundant dictionary in non-stationary environments, there is few attempt of analyzing their stochastic convergence behaviors. In this paper, we briefly review the zero-attracting kernel leastmean-square (ZA-KLMS) algorithm with ℓ1-norm regularization from the perspective of nonlinear sparse system. Then, the theoretical transient convergence performance of ZA-KLMS algorithm using Gaussian kernel function with pre-tuned dictionary is analyzed in the mean and mean-square senses. The simulation results illustrate the accuracy of derived analytical models by the excellent consistency between the Monte Carlo simulations and the theoretical predictions, and the ZA-KLMS algorithm has better convergence performance than the KLMS algorithm for nonlinear sparse systems in stationary environment.

Internal overvoltage accidents in the distribution network are likely to cause an equipment insulation breakdown and result in system power outages and economic losses. Therefore, an internal overvoltage identification method based on the time-frequency atomic decomposition is investigated in this study. Firstly, the overvoltage waveforms are divided into four time periods. Then, the waveforms during these four time periods are decomposed by the atomic decomposition algorithm to obtain the effective atoms from the waveforms. Moreover, the root-mean-square (RMS) value of the zero-sequence voltage, the dominant atom frequency, the total relative matching degree, and the effective atom frequency are extracted as major features. In addition, the layered identification of the overvoltage types can be realized by combining the corresponding identification criteria. The salient advantage is the features with low dimension and a high degree of discrimination. Various overvoltage types can be identified just by the corresponding thresholds, and it is easier to deploy in the field than conventional methods based on classifier training. The effectiveness of the proposed method is verified by experimental results, and it concludes that the proposed algorithm has high accuracy and strong adaptability.

In the Body Sensor Networks (BSNs), a comprehensive and intuitive conclusion can be obtained by analyzing the physiological state in the mode of multi-heterogeneous source fusion. However, if a node in the sensing layer fails, the input-oriented analysis system is no longer reliable. In this paper, we propose a homogenous sensor weight calculation method for a single sensing node in the BSNs. Under the premise of valid sensing, a set of sequences will be randomly selected as a referee to make the rules of weight allocation. Weak competitiveness means that sensors have a high degree of antagonism, which indicates that the measuring results have errors or there may be abnormal sensors in the sensing module. Meanwhile, we introduce the trend factor of human body state to strengthen the connection between adjacent data in time series, so as to further improve the reliability of trust weight. The proposed algorithm is helpful for sensing nodes in BSNs to obtain more reliable original physiological data and ensure the accuracy of subsequent feature analysis and system decision. Specifically, it can (1) improve the robustness of sensing in BSNs applications, (2) suppress the interference of abnormal signals, (3) solve data conflict, and (4) be universal to the sensors in BSNs devices. The ECG simulations verify the validity of the algorithm in all extreme cases of valid sensing, and the PPG experiments show that the proposed algorithm is applicable to different types of biosensors. Finally, the evaluation and comparison with other four known data-level fusion methods show that the proposed algorithm has better fusion performance.

This paper studies the transient behavior of the diffusion least-mean-square (LMS) algorithm over the single-task network for the non-stationary system using diverse types of cyclostationary white non-Gaussian inputs for an individual node. The analytical models of the recursive mean-weight-error vector and mean-square-deviation are derived for the system with random walk varying parameters and the white random process with periodically deterministic time-varying input variance. In addition, the approximated steady-state mean-square-deviation of the diffusion LMS is presented for the slow varying input variance. Monte Carlo simulations show excellent agreement with the theoretical prediction of mean-square-deviation validating the accuracy of derived analytical models and the tracking ability for non-stationary system and cyclostationary inputs simultaneously.

The convolutional and deep neural networks are prevalent machine learning algorithms for real-world applications. As the neural network needs large computations, many artificial intelligence (AI) chips are designed to accelerate the computation. AI chips have achieved better energy efficiency and high computational capacity in the neural network implementation. The communication network in AI chips influences the data transformation and hardware efficiency. The network-on-chip (NoC) is one feasible solution to meet the data communication requirements in AI chips. This paper introduces the communication network in AI chips and the strategy of mapping neural network to chips with the extensible hierarchical architecture. We also conclude the opportunities for communication optimization in the design of AI chips. In this paper, we propose our processor architecture and optimize the performance and energy of intracommunication in chips from three aspects: data reuse, topology, and router architecture. The experimental results show that our optimization can totally achieve 25.31× latency reduction and 79.92× energy less than the baseline. The results show that our design can reduce the latency by 5.47× and save communication energy by 7.5× when compared with the state-of-the-art design DaDianNao. When compared with another design Eyeriss, our design can reduce latency by 7.57× and save communication energy by 3.03×.

Fault diagnosis of photovoltaic (PV) arrays is an essential task for improving the reliability and safety of a photovoltaic system (PVS). The PVS faults at the DC side are difficult to detect by traditional protective devices, which may reduce power conversion efficiency and even lead to safety matters and fire disaster. This study investigates a newly-designed fault diagnostic method for a PVS according to the following three steps. First, optimal fault features are extracted by analyzing I-V curves from different faults, including hybrid faults of the PVS under the standard test condition (STC). Moreover, the trust-region-reflective (TRR) deterministic algorithm combined with the particle-swarm-optimization (PSO) metaheuristic algorithm is proposed to standardize fault features into the ones under the STC. In addition, a multi-class adaptive boosting (AdaBoost) algorithm, which is the stage-wise additive modeling using multi-class exponential (SAMME) loss function based on the classification and regression tree (CART) as the weak classifier, is utilized to establish the fault diagnostic model. The effectiveness of the fault diagnostic model could long-term maintain by periodically updating the feature standardization equations to standardize the fault features into the ones under the STC. Various types of the PV modules are used to validate the generalization of the fault diagnostic method. Both the numerical simulations and experimental results show the accuracy and reliability of the proposed fault diagnostic method.

Vehicle platooning is of great importance in future autonomous driving and intelligent transportation systems, due to its advantages in road safety, traffic efficiency, energy consumption and exhaust emissions. This paper focuses on the scalability performance of platooning control, which aims to achieve long platoon size under the premise of ensuring consensus behavior of the platooning vehicles. However, in classical platooning schemes such as ACC (adaptive cruise control) and CACC (cooperative adaptive cruise control), as the number of platoon members increases, the communication range of the leader and the cascaded sensor delay affect the scalability of platoon. In this paper, a scalable platooning scheme, CACC-granulation, is proposed to improve the scalability of platooning based on a novel information flow topology. The granulating method is used to solve the problem of limited communication range of leader for CACC by forwarding their own information to platoon members through some vehicles. The CACC-granulation granulates platoon information flow topology and enhances the platoon scalability by reducing information flow topology matrix. Simulation experiments are conducted to verify the consensus and scalability performance of CACC-granulation. Compared with other two platooning schemes which can get long platoon size, i.e., ACC-cascade and ACC-CACC-integration, the simulation results indicate the performance advantages of the proposed CACC-granulation, which not only meets the consensus of platooning control, but also enhances the scalability of platooning control.

The modes' reconstruction into the stochastic and deterministic components is proposed for forecasting the crude oil prices with the concept of “divide and conquer” and modes reconstruction. It is to reduce the complexity in the computation and to enhance the forecasting accuracy of the decomposition ensemble technique. Under the framework of “divide and conquer”, the decomposition and ensemble methodologies of forecasting power successfully improves with the proposed model based on the modes' reconstruction. The corresponding reconstruction is using average mutual information (AMI). The proposed procedure is based on four layers i.e., complex data decomposition, reconstruction of modes into components, the prediction of each individual component and assembling the final prediction. In the proposed procedure, the modes of the stochastic component are analyzed thoroughly as it influences the prediction results significantly. For verification and illustration purposes, the case study of Brent and West Texas Intermediate (WTI) daily crude oil prices data are used, and the empirical study confirms that the outcomes outperform all the considered benchmark models, including auto-regressive integrated moving average (ARIMA) model, generalized autoregressive conditional heteroscedasticity (GARCH) model, NAÏVE model, ARIMA Kalman Filter model. This outcome is achieved, with the reconstruction decomposition ensemble (RDE) model along stochastic and deterministic components. Hence, it is concluded that the proposed model achieved higher forecasting accuracy and takes less computational time with the modes' reconstruction as opposed to using all the decompose modes.

We propose a novel Web services clustering framework by considering the word distribution of WSDL documents and tags. Typically, tags are annotated to Web services by users for organization. In this paper, four strategies are proposed to integrate the tagging data and WSDL documents in the process of service clustering. Tagging data is inherently uncontrolled, ambiguous, and overly personalized. Two tag recommendation approaches are proposed to improve the tagging data quality and service clustering performance. Comprehensive experiments demonstrate the effectiveness of the proposed framework using a real-world dataset.