Search

Inpainting images that have sizable missing blocks presents a considerable challenge in terms of preserving visual consistency and attaining a convincing result. In this study, the multi-scale generative adversarial network with multi-head external attention for image inpainting (denoted as MGDAN) is proposed. First, an adaptive weight style loss is designed into the generator of multi-scale generative adversarial networks (GAN) to guide the inpainting of the style and structure in the image inpainting, which improves the inpainting effect of image contour and emoticon. Second, the Adaptive Mix model is introduced to address the imbalance in generator and discriminator training by reducing the distance between the difficult and easy samples. This approach is intended to improve the overall performance of the network and drive high-quality image generation. Third, the generator and local discriminator in the image inpainting process are enhanced by the introduction of a multi-head external attention mechanism. This addition aims to capture the long-distance and multi-level dependent relationships between different areas of the image. It is proved beneficial for the generating clear geometric contours in the restored images and improving the overall global consistency of the inpainted results. By employing specific experimental methods, we find the optimal number of heads. The effectiveness of the proposed approach also is demonstrated through the extensive experiments conducted on public datasets.

This paper presents a comprehensive survey on interpretable intelligent fault diagnosis for rotating machinery, addressing the challenge of the “black box” nature of machine learning techniques that hampers reliability in automated diagnostic processes. It underscores the growing importance of interpretability in intelligent fault diagnosis (IFD), marking a shift from traditional signal processing methods to machine learning-based approaches that necessitate transparency for trustworthiness. Our review systematically collates and examines the spectrum of interpretability in IFD, distinguishing between post-hoc and ante-hoc strategies. We detail mainstream post-hoc methods, their applications, and critique their limitations, particularly the absence of physical significance. The survey then explores ante-hoc methods that incorporate physical knowledge upfront, enhancing interpretability. By categorizing and evaluating three distinct knowledge embedding approaches, we shed light on their unique applications. Conclusively, we highlight emerging research directions and challenges in the field, aiming to equip readers with a nuanced understanding of current methodologies and inspire future studies in making IFD more reliable and interpretable.

A novel adaptive robust hybrid force/position control (ARHFPC) strategy is proposed for robot manipulator systems subject to dynamic uncertainties and unknown matched and mismatched disturbances under input saturation. First, the position controller is designed based on the backstepping approach. The first-order low-pass filter and the auxiliary dynamic system are synthesized into the controller to overcome the complex derivative operation of virtual control and handle the effect of input saturation, respectively. Radial basis function neural networks (RBFNNs) are utilized to approximate the dynamic uncertainties and matched disturbances. Then, a disturbance observer is designed for the mismatched disturbances. To enhance control accuracy of the interaction force between the end-effector and the external environment, a fuzzy proportional-integral (FPI) control scheme is presented. Theoretical analysis proves that all signals in the closed-loop control system of robot manipulators are locally uniformly ultimately bounded (UUB). Simulation results demonstrate the effectiveness and robustness of the proposed control scheme.

In recent years, there has been extensive research on the routing problem of printed circuit boards (PCBs). Due to the increasing number of pins, high pin density, and unique physical constraints, manual PCB routing has become a time-consuming task to achieve design convergence. Previous work decomposed the problem into escape routing and area routing, focusing on these problems separately. However, there was always a gap between these two problems, requiring significant human effort for iterative algorithm adjustments. Furthermore, previous area routing work mainly focused on routing between ball grid array (BGA) packages in escape routing. However, in practice, many components are not in the form of BGA packages, such as passive devices, decoupling capacitors, and through-hole pin arrays. Therefore, it is necessary to study a unified routing approach. The current unified routing approach adopts the A* algorithm, but there is still room for improvement in routing speed. This paper proposes a new algorithm called Unet-Astar, which accelerates the routing efficiency by employing deep learning algorithms in a simulated environment. Additionally, a Deeper Unet is proposed for generating recommended regions for the routing algorithm. The new network structure can provide more contextual information, thereby improving routing efficiency. Experimental results demonstrate the effectiveness and efficiency of the proposed algorithm. Specifically, for all given test cases, our router achieves approximately a 70% improvement in runtime speed compared to the old router. Another major contribution of this work is the development of a routing problem set generator, which can generate parameterized routing problem sets with different sizes and constraints. This enables the evaluation of different routing algorithms and the generation of training datasets for future data-driven routing methods. All the code has been open-sourced and can be found at https://github.com/Firesuiry/Unet-Astar-For-PCB-Routing.

Azimuth electromagnetic (EM) Logging While Drilling (LWD) tools play an important role in geological steering and reservoir evaluation. Its measuring response is amplitude ratio (ATT) and phase difference (PS), which can’t directly reflect formation information. To obtain direct formation information such as resistivity and formation boundary, accurate and efficient inversion method is essential. However, the existing inversion methods (i.e., iterative method) have some problems, such as slow computation speed, dependence on initial value selection and easy to be trapped by local minimum. Therefore, this paper proposes an intelligent inversion method for azimuthal EM LWD measurements based on the U-net deep learning network framework. Firstly, an efficient analytical solution is used to generate amounts of EM LWD data. Those samples are divided into training and test sets in a 9:1 ratio, which are used for training and testing of the network, respectively. Then, the network parameters are constantly adjusted during training to ensure its inversion performance. Finally, the trained network is utilized to invert the samples of the test sets, and the inversion results are compared with the forward formation model. The study’s results demonstrate that the network is capable of efficiently and precisely inverting both isotropic and anisotropic formations, with a single sample being inverted less than 0.05 seconds. Its noise layer parameters can be improved to successfully invert noisy data, leading to good robustness. In addition, the network has good applicability for the inversion of complex formations. These consequences highlight the significant potential of this method in azimuthal EM LWD inversion applications.

As one of the most commonly used metal cutting machines in modern manufacturing industry, the drilling and tapping center machine tool has an irreplaceable position in modern manufacturing industry due to its high operating speed and fast tool changing speed. However, in the process of machining, the thermal deformation of the x-axis lead screw is caused due to the changes of the ambient end temperature, the wire mother end temperature and the bearing temperature of the motor seat end, resulting in the error of the workpiece. In this paper, through the above three kinds of thermal deformation simulation and experimental data measurement, and then through the analysis and modeling of the experimental data by MATLAB, a scheme of real-time compensation for the thermal error of the x-axis lead screw is proposed. After the compensation, the products produced by an enterprise are verified. The verification results show that the accuracy after the compensation can be improved by 56.25% compared with that before the compensation.

In this paper, we propose an enhanced end-to-end object detector based on Sparse R-CNN (EnSparse R-CNN), which aims at backbone, neck and head of object detector. For the backbone network, we constructure an aggregate residual network (ARNet) with the aggregate connection, which can achieve global features to improve the ability of feature extraction by extracting the salient feature from the feature maps. For the neck network, we build an attention context feature pyramid network (ACFPN) to fuse the global features with the context features, which can integrate multi-layer features and attention features to make full use of global context features. For the detection head, we design a gradually refined detection head (ReHead) to detect objects, which adopts the sparse detection with the refined bounding box regression to improve the confidence of the regression. The experimental results on COCO dataset show that EnSparse R-CNN outperforms Sparse R-CNN by 2.5 % and the anchor-free one-stage detector AutoAssign by 1.1%.

This paper is devoted to the study of delay-dependent stability of time-varying delay systems. A delay-derivative-partitioning approach is proposed. By constructing an augmented Lyapunov functional that contains two delay-product-dependent terms and using the delay-derivative-partitioning approach, an improved stability condition is established. The derived condition is described as a set of linear matrix inequalities, which can be readily implemented. Finally, four examples are carried out so as to attest the effect and merit of the presented approach.

Government hotline is closely related to people’s lives and plays an important role in solving social problems and maintaining social stability in China. However, the event text of the hotline is inconsistent in length and unclear in elements, so it is a challenge for the operator to manually complete the assignment tasks of hotline event. To address these problems, we propose a joint learning method for event text classification and event assignment for Chinese government hotline. Firstly, graph convolution network (GCN) and BERT are used to process the event text respectively to obtain the corresponding representation vector. Then, the obtained two representation vectors are fused by the dynamic fusion gate to get fusion vector and classified the fusion vector through the text classification. Secondly, we use multi-attention mechanism to process the GCN result vector, BERT result vector and the “sanding” vector to obtain attentive “event-sanding” representation vector and calculate the corresponding department probability distribution. Finally, the historical prior knowledge based reorder model is used to sort the results of the “event-sanding” matching module and output the optimal assignment department of government hotline event. Experimental results show that our method can achieve better performance compared with several baseline approaches. The ablation experiments also demonstrate the validity of each proposed module in our model.

Hematoxylin and eosin (H&E) stained colors is a critical step in the digitized pathological diagnosis of cancer. However, differences in section preparations, staining protocols and scanner specifications may result in the variations of stain colors in pathological images, which can potentially hamper the effectiveness of pathologist's diagnosis and the robustness. To alleviate this problem, several color normalization methods have been proposed. Most previous approaches map color information between images highly dependent on a reference template. However, due to the problem that pathological images are usually unpaired, these methods cannot produce satisfactory results. In this work, we propose an unsupervised color normalization method based on channel attention and long-range residual, using a technology called invertible neural networks (INN) to transfer the stain style while preserving the tissue semantics between different hospitals or centers, resulting in a virtual stained sample in the sense that no actual stains are used. In our method, the expert does not need to choose a template image. More specifically, we have developed a new unsupervised stain style transfer framework based on INN that is different from state-of-the-art methods. Meanwhile, we propose a new generator and a discriminator to further improve the performance. Our approach outperforms state-of-the-art methods both in objective metrics and subjective evaluations, yielding an improvement of 1.0 dB in terms of PSNR. Moreover, the amount of computation of the proposed network has been reduced by 33 %. This indicates that the inference speed is almost one third faster while the performance is better.

Post-stroke care encounters challenges, including high cost, lack of professionals, and insufficient rehabilitation state evaluation. Computer technology can alleviate these issues, as it allows health care professionals (HCP) to quantify the workload and thus enhance rehabilitation care quality. In this paper, a novel multi-model fusion method, terms as pose dual-stream network (PDSN), is devised, aiming to test the feasibility of monitoring the training actions of rehabilitating stroke patients in care management. In particular, this deep-learning-based algorithm combines human pose estimation and dual-stream networks in an innovative way. We utilize an improved OpenPose to estimate human pose from videos obtained by the low-cost monocular camera. In dual-stream networks, the spatial and motion streams are flexibly integrated. The spatial stream network combines the Gated Recurrent Unit (GRU) and attention mechanism to extract spatiotemporal data, while the motion stream network is composed of improved multi-layer 1D Convolutional Neural Networks (CNN), which enhanced by causal and dilated convolution skillfully. Additionally, an adaptive weight fusion strategy is used to fuse the two networks for the final action classification. Results show high accuracy on two public datasets and a dataset created by us, which validate the superiority and feasibility of our method.

With the support of the second phase of the Chinese Meridian Project, a new mesosphere-stratosphere-troposphere observation radar (MST radar), called Qinzhou MST radar, located in Qinzhou City, Guangxi Province, China (22°7'55''N,108°16'20''E) is under construction. The Qinzhou MST radar is a large VHF monostatic pulse Doppler radar with an active phased array antenna system, mainly using Doppler beam swinging technique for observation. The capability of multibeam observation, space antenna observation, and frequency- and spatial-domain interferometry are also employed. The main scientific objective of this radar is to explore the atmosphere dynamical processes at low latitudes of China. The observation of ionospheric field-aligned irregularities (FAIs) is also an important function of the radar. The radar applies a circular array composed of 1261 3-element crossed Yagi antennas with a diameter of 153 m and an aperture of near 2×104 m2. Thereinto, 997 elements connect both transmitting and receiving channels and 264 elements distributed near the edge connect only receiving channels. The peak power of the transmitter in each transmitting/receiving (T/R) module is 2 kW. The central frequency and bandwidth of the output signal are 50 and 4 MHz, respectively. The peak power of the whole system will be close to 2 MW. This article presents the initial design of the Qinzhou MST radar, and the construction will be completed at the end of 2022.

Ship detection by an airborne passive interferometric microwave sensor has been proposed as a powerful complementary tool for ship detection, especially for large ships in low visibility conditions and high sea clutter. In this article, the brightness temperature (TB) models of the sea background and the ship on the sea are established. Given the TB models, a ship detection and tracking algorithm with four steps is proposed, which is devoted to ship tracking. In addition, the impacts of the noises in the passive interferometric microwave images (PIMI) and the algorithm parameters are analyzed and discussed. Numerical simulations and airborne experiments are performed to assess the practicability of the algorithm. Both numerical and experimental results have demonstrated the practicability of the proposed algorithm with dealing with the additional systematic errors and estimating the ship heading, and the results also exhibit that the additional noises in the PIMI caused by the stochastic fluctuations of the aircraft and the algorithm parameters have a non-negligible impact on the algorithm detection performance. As a result, the “ship-free” observed TB maps should be carefully considered as the sea background TB map to improve the algorithm detection performance by reducing the noises caused by the stochastic fluctuations of the aircraft. In addition, the algorithm parameter should also be carefully considered to get a high ship detection probability with a low ship false-alarm rate.

Under the perspective of carbon neutrality, the green electricity absorption target constrained by the quota system policy plays a crucial role in reducing the carbon emission of the power industry. However, the current green certificate policy has not achieved good results. On the premise of reducing the additional market burden as much as possible, the policy parameters should take into account the influence of market behavior to formulate better policy parameters in line with China’s carbon emission peak goal. This paper constructs a combined hierarchical reinforcement learning with off-policy correction and multi-agent deep deterministic policy gradient algorithm (HIRO-MADDPG). It realizes the benefit analysis of the existing policy parameters joint with the solution of the optimal policy parameters. The algorithm solves the problem that benefit analysis and parameter formulation cannot be jointly trained and improves the precision. The results indicate: 1) HIRO-MADDPG algorithm can reach the highest policy benefits on the premise of maintaining market fairness; 2) under the new optimal policy parameters, the income per kilowatt hour of thermal power generator(TPG) and renewable power generator(RPG) can be maintained at 10% under the condition of abolishing subsidies; 3) with the help of the new policy parameters, China’s power sector will reach the peak of carbon emissions from coal-fired power plants in 2026 ahead of schedule, and reduce carbon emissions by a further 11% by 2030.

To protect image data privacy, a secure and fast image encryption algorithm using hyper-chaos based key generator and vector operation is proposed. Firstly, we introduce a novel post-process method to create a key matrix for significantly reducing the iterations of the hyperchaotic system from $W\,\,\times \,\,H$ /4 to $2W$ for an image size of $W\,\,\times \,\,H$ . And within the post-process, a random selector driven by the piecewise linear chaotic map ensures the good randomness and unpredictability of the generated key matrix. Secondly, the vector operation is employed to accelerate the cipher block chaining based diffusion process, so the plain image can be parallel encrypted row by row and column by column with high efficiency. Thirdly, we utilize a Logistic map to quickly produce an initial vector for the vectorized diffusion process. Finally, the proposed algorithm is evaluated by some common security and performance tests. Experimental results show that the cipher image can pass all tests of NIST SP 800–22 with $P$ -values $\gg ~0.01$ , its correlation coefficient between pixels is close to 0, and the entropy is greater than 7.999. Combined with other results of security tests, we can safely conclude that the proposed image encryption algorithm provides adequate protection against statistical, brute-force, chosen-plaintext, and other common types of attacks. In addition, the time complexity is in the order of ${O}$ ( $W+H$ ) and the average encryption time of 512 $\times $ 512 images is only 0.023s. The results indicate that our algorithm with high security and fast speed can meet the requirements of real-time confidential transmission of massive image data.

This paper analyzes an optimal dynamic production control problem of a remanufacturing system in stochastic environment. The system consists of one inventory for serviceable items and one virtual inventory for the items used by customers which will be either remanufactured to be serviceable items or discarded. Demands will be lost if there is no serviceable items available. At any decision epoch, the decision maker has to decide whether to produce items or not by raw materials, so as to minimize the long run average cost. The optimization problem is constructed as a Markov decision process (MDP). We show that the optimal policy has a multi-threshold structure (switching curve) and it is monotonic in system parameters. Furthermore, based on the structure of the optimal policy, we construct a performance evaluation model for computing efficiently the optimal thresholds. The expression of the average cost is given by a quasi-birth-death (QBD) process. Finally, we provide some numerical results to show the effectiveness of our approach and the impact of different parameters on the optimal policy and average cost.

Coastal land cover mapping is a significant yet challenging pixel-level segmentation task. Domain shift between optical remote sensing imagery will give rise to remarkable performance degradation for deep supervised methods. Besides, the ground objects characterized with interclass variance and class imbalance may further aggravate the adverse effect. Traditional adversary-based domain adaptation algorithms always leverage a binary discriminator to conduct global adaptation, ignoring the detailed class information. In this article, we develop a novel class-aware domain adaptation method to address these issues. Unlike the naive single one, we propose a joint local and global adversarial adaptation framework to separately execute class-specific and global domain alignment on feature and output spaces. For the former, the introduced classwise discriminator possesses different strategies to extract labels for both data domains. Meanwhile, we restore to entropy minimization to produce high-confident target prediction rather than using the early generated pseudo label with high confidence. Furthermore, we additionally adopt comprehensive reweighting on the supervised segmentation loss to track the class imbalance problem. This manner mainly comprises the sample-based median frequency balancing and the focal loss function for the minority and hard classes. We measure the proposed method on two typical coastal datasets and compare it with other state-of-the-art models. The experimental results confirm its excellent and competitive performance on cross-domain land cover mapping.

This article studies the model, control and simulation on transition flight of a novel tail-sitter with vertical takeoff and landing capability. The proposed tail-sitter adopts an innovative varying fuselage shape design to break the technical bottleneck in the balance of efficient horizontal flight and agile vertical flight. Overset grids and computational fluid dynamic methods are used to explore the vehicle’s prestall aerodynamics, which are relative to not only the angle of attack and sideslip angle but also to the varying angle between two rear fuselage’s parts. High angle of attack aerodynamics based on the improved Viterna and Corrigan stall model is also established for this novel tail-sitter. Meanwhile, an accurate model of the propeller is tested in a wind tunnel. Combining the forces and moments generated by propellers, aerodynamics and gravity, a 6DoF nonlinear time-varying dynamic model is built. A robust controller based on incremental dynamic inversion method is designed for this tail-sitter, which is good at dealing with uncertainties and external forces. Nominal and model mismatch conditions are simulated to verify the controller’s performance. Different varying strategies for the mechanism are analyzed during the transition flight. Simulation results show that this novel tail-sitter can transform between vertical and horizontal flight mode easily and the varying strategy related to pitch angle is a prior choice for transition flight.

The unmanned tilt-wing cargo aircraft using distributed propulsors is an emerging aircraft significantly different from traditional types. This paper proposes an aerodynamic, propulsion, noise, weight integrated optimization design method for this new aircraft. The method consists of several functional modules specially developed or adjusted targeting the aircraft's characteristics, such as the boundary state analysis, propeller/rotor oblique inflow analysis, waked wing analysis, propeller/rotor noise evaluation, multi-state wing mass analysis, multi-objective genetic algorithm optimization. It comprehensively considers the impact of various complex factors on the optimization results, such as the impact of distributed propulsors on the wing aerodynamics, the effect of wingtip propellers on the induced drag reduction, the coupling between wing aerodynamics and structure, the propeller/rotor aerodynamics optimization, and noise control. With the proposed method, it is possible to directly translate the top-level design requirements into the design scheme with the optimal specific system performance (such as the lowest delivery cost and highest delivery efficiency) at the very initial aircraft design stage, thereby greatly shortening the development cycle. A case study was presented. The results show that the introduction of distributed propulsors can increase the delivery efficiency by 28.2% and reduce the delivery cost by 15%; suppressing the wingtip vortices using propellers can increase the wing lift-drag ratio by 5.43%-6.65%; the slipstream generation efficiency and thrust efficiency are significantly different between different distributed propulsor schemes. To maximize the overall efficiency, it is necessary to balance between the slipstream generation efficiency and the overall thrust efficiency when optimizing the tilt-wing cargo aircraft.

Long time coherent integration technique is one of the most important methods to improve the radar detection ability of a weak maneuvering target. However, the integration performance may be greatly influenced by the range migration and Doppler migration effects caused by the complex motions of maneuvering targets when long time integration is applied. Besides, the frequency hopping characteristic of the transmitted signal will also result in Doppler migration which cannot be eliminated by the conventional coherent integration method. To deal with this problem, a novel coherent integration method based on phase compensation is proposed. By using the frequency hopping pattern, the Doppler migration is eliminated by adding an extra phase in the twiddle factor of Fourier Transform in this proposed method. Since the phase error is compensated, long time coherent integration can be achieved. The effectiveness of the proposed method is verified by the simulation results and the performance analysis.

The technology of dynamic wireless power transfer (DWPT) has been recognized as an effective way to recharge battery electric bus and overcome some drawbacks (e.g. high battery cost and limited service range) with opportunity charging. This study develops a mixed integer non-linear model to optimize a feeder bus transit powered by DWPT. The decision variables consist of bus route networks, service frequency, locations of DWPT devices and battery capacity. The objective is to minimize total cost, including the costs of charging devices, battery, operation and travel time. A tangible nested genetic algorithm (NGA) is developed to find the optimal solution. The computational efficiency of NGA is demonstrated through numerical comparisons to the solutions founded by LINGO and GA. It was found that with NGA the solution converges to an acceptable level faster than using LINGO and GA. A real-world bus network is employed to explore the relation between the minimized costs and decision variables. The result suggested that DWPT outperforms terminal charging technology in terms of the least total cost, and that the yielded total infrastructure cost with DWPT is 16.6% less than that with terminal charging technology.

In recent years, the ultra-low frequency oscillation of power system is one of the most concerned frequency stability problem. The governor deadband are considered as one of the main influences. However, the existing methods can only deal with single deadband segment. As for a multi-machine system with multiple deadband segments, it lacks the analysis method for frequency oscillation. To solve this problem, this paper first define the extended description function applied to nonlinear composite system according to the principle of traditional description function method. The extended description function can quantify the relationship of input and output of a system with multiple nonlinear segments. Then, based on the uniform frequency model of multiple governors, the turbines and governors of hydro and thermal power units with different deadband widths are equivalent to one extended description function. The locus of minus reciprocal of the extended description function with oscillation amplitude and the Nyquist curve of the generator are plotted on the complex plane for the analysis of system stability. Next, the critical oscillation amplitude of disturbance to trigger the system frequency oscillation is calculated according to the point of intersection of both curves. Finally, the simulation results show that the critical oscillation amplitude can reflect the stability of system. The change of deadband width is positively correlated with the critical oscillation amplitude. The appropriate setting of deadband and configuration of hydro-thermal generation are effective measures to suppress the ultra-low frequency oscillation while retain the ability of frequency regulation.

In order to solve the problem when the target detection of passive bistatic radar is seriously affected by direct signal and multi-path, a clutter cancellation algorithm based on frequency domain analysis is proposed in this paper. By analyzing and processing the frequency domain correlation between the reference signal and the echo signal, the time delay and the amplitude of the direct signal and multipath in the echo signal can be estimated for clutter cancellation. This method can not only reduce the computation greatly, but also eliminate the influence of clutter echo residue on target detection. The effectiveness of the proposed algorithm is verified by simulation and real data processing.

The ultra-low frequency oscillation (ULFO) problem reappears in China recently. Different from the classical low frequency oscillation (LFO), the generators in ULFO have the same mode shapes, namely oscillating in the same pace. To reveal the reason why ULFO generators do not oscillate inversely as classical LFO does, this paper investigates the mechanism of ULFO based on improved Philips-Heffron model of two-machine system. Through this model, the phase angle relationship between different generators is obtained. Based on the bode diagram, it is found that the phase angle difference of the two machines is around zero in ultra-low frequency band however the generator parameters change. On the contrary, the situation is different in classical low frequency band, which reveals the mechanism that generators in ULFO always oscillate in the same pace. Finally, the analysis above is validated through a 2-area 4-machine system and a 6-machine system established in PSASP software.

The customized bus (CB) transit is recognized as an effective transportation mode offering more flexible and demand-responsive service than traditional bus transit with fixed route and schedule, especially during the peak hours. The novelty of this study is the development of a mixed integer non-linear model for optimizing multi-terminal CB service in an urban setting. According to the estimated spatiotemporal passenger demand, the objective total cost, consisting of supplier's and users' costs, is minimized subject to capacity and time window constraints. A mixed bus fleet with various bus sizes is employed to accommodate passenger demand, which increases vehicle utilization and reduces supplier's cost. The inconvenience of passengers caused by early arrival at the destination is treated as penalty and considered in users' cost. The study optimization problem is combinatorial with many decision variables including trip assignment, bus routing and associated timetables, and fleet size. A hybrid genetic algorithm (HGA) which integrates the features of genetic algorithm (GA) and simulated annealing (SA) is developed to effectively search for the optimal solution. A real-world CB network is employed to demonstrate the applicability of the developed model and explore the relation between the model parameters and optimized results. It was found that the total cost can be reduced by 16.5% after employing multiple terminals and a mixed bus fleet.

The joined-wing configuration SensorCraft, presents a more feasible way to meet tough design requirements such as long- endurance and 360-deg radar coverage compared to conventional and flying wing configurations. Proper assessment of the interactions between aerodynamics, structure, and embedded antenna in the optimization of the joined-wing SensorCraft is essential to obtain superior performances. This paper establishes the embedded radar performance estimation model and describes an integrated design process of aerodynamic, structure, and embedded radar performance of such an aircraft. Results show that multi-objective optimization leads to an increase of 14.76% in lift-to-drag ratio, 3.96% and 8.75% in the forward-looking and backward-looking radar detection ranges respectively. Moreover, the structural weight is reduced by 6.0% compared to that of the baseline design. Moreover, sensitivities of the rear wing dihedral angle, endplate height ratio, and joint location of the joined-wing configuration are analyzed in detail.

In the domain of speech emotion recognition (SER), generally there is an unbalanced data distribution of emotional samples in existing emotional datasets. Moreover, different fragment areas in an utterance contribute diversely to SER. To address these two issues, this paper proposes a new SER method by combining a unified first-order attention network with data balance. The proposed method firstly utilizes the strategy of data balance to augment and balance the training data. Then, a pre-trained convolutional neural network (CNN) model (i.e., VGGish) is fine-tuned on target emotional datasets to learn segment-level speech features from the extracted Log Mel-spectrograms. Next, the unified first-order attention mechanism, including different feature-pooling strategies such as sum, min, max, mean, and standard deviation (std), is embedded into the output of a bi-directional long short-term memory (Bi-LSTM) network. This is used for learning high-level discriminative segment-level features, and simultaneously aggregating the learned segment-level features into fixed-length utterance-level features for SER. Finally, based on utterance-level features, the softmax layer in a Bi-LSTM network is adopted to conduct final emotion classification task. Extensive experiments are implemented on three public datasets such as BAUM-1s, AFEW5.0, and CHEAVD2.0, demonstrate the advantage of the proposed method.

This paper addresses the distributed tracking control problem of pure-feedback multi-agent systems with full state constraints under a directed graph in finite time. By introducing the nonlinear mapping technique, the system with full state constraints is converted into the form without state constraints. Furthermore, by combining fractional dynamic surface and radial basis function neural networks, a novel finite-time adaptive tracking controller is conducted recursively. In light of Lyapunov stability theory, it is proven that all signals of multi-agent systems are semi-globally uniformly ultimately bounded in finite time and the full states satisfy the constraints. Lastly, numerical simulations are supplied to demonstrate the effectiveness of the proposed control strategy.

Predicting enzymes function is an important and difficult problem, particularly when enzymes may have the multiplex character, i.e., some enzymes simultaneously have two or three function classes. Most of the existing enzyme function predictor can only be used to deal with the mono-functional enzymes. Actually, multi-functional enzymes should not be ignored because they usually possess diverse biological functions worthy of our special notice. By introducing the “improved Hybrid Multi-label Classifier” and “neighbor score”, a new predictor, called MF-EFP, has been developed that can be used to deal with the systems containing both mono-functional and multi-functional enzymes. As demonstration, the jackknife cross-validation was performed with MF-EFP on a benchmark dataset of enzymes classified into the following 7 functional classes: (1) EC 1 Oxidoreductase, (2) EC 2 Transferase, (3) EC 3 Hydrolase, (4) EC 4 Lyase, (5) EC 5 Isomerase, (6) EC 6 Ligase, (7) EC7 Translocases, where none of enzymes included has ≥90% pairwise sequence identity to any other in a same subset. The subset accuracy and average precision thus obtained by MF-EFP was 85.62% and 94.16% respectively. Extensive experiments also show that MF-EFP can outperform the existing predictors that also have the capacity to deal with such a complicated and stringent system. As a user-friendly web-server, MF-EFP is freely accessible to the public at the web-site http://www.jci-bioinfo.cn/MF-EFP.

Path planning is necessary for free-floating space manipulators to perform space tasks. However, the coupled motion between manipulator and base directly acts on path planning procedure, which results in current manipulator motion state is dependent on previous motion state, so we must consider the control of base-coupled motion in path planning. In this paper, we propose a general path planning strategy for free-floating space manipulators based on the representation space (RS) analytical method. RS consists of the representation variables which are connected to task attributes and base-coupled motion simultaneously, and it can be utilized to check task realizability and construct a path planning domain for free-floating space manipulators. To analyze the effects of base-coupled motion on path planning, we classify RS into five types according to different base control modes: fixed vehicle RS (FRS), attitude constrained RS (CRS), maximum reachable RS (MRS), guaranteed RS (GRS), and partly guaranteed RS (PGRS). CRS tells us the variation of base attitude (instead of base centroid position) directly influences task realizability and path planning, so we should particularly consider the control of motion of base attitude in path planning. With appropriately considering the control of base attitude and satisfying the base attitude deflection limitation, only PGRS can be utilized to check task realizability and construct a path planning domain for free-floating space manipulators. Then, we design a path planning strategy in PGRS based on A* algorithm. Finally, we apply the RS analytical method to the 3-DOF free-floating space manipulator to visually show RS, and the effectiveness of the method is verified by a simulation experiment. The RS analytical method is appropriate to path planning for any kinds of manipulators and various tasks. We only need to select the suitable representation variables, which can faithfully reflect task and manipulator attributes.

Kinematic calibration performance is heavily dependent on two factors-the ability of calibration configurations mapping kinematic parameter errors, and the un-modeled errors including joint clearance, thermal expansion, and measurement noise. Therefore, this paper deals with the calibration configuration optimization to reduce the impact of the two factors on calibration performance. We pay particular attention to establish an index for evaluating calibration configuration's quality. Different from other works, the proposed comprehensive quality index can simultaneously reflect configurations' observability and globality. Furthermore, the numerical methods are used to analyze the relationships between the comprehensive quality index and configuration number, and the relationships between calibration performance and configuration number. Based on the above relationships, we provide a feasible solution for determining the calibration configuration number of a specific manipulator. Based on the above work, configuration optimization model is established and solved by particle swarm optimization. The simulation of an eight degree-of-freedom manipulator illustrates the advantages of the proposed method. In 100 calibration simulations, optimized configurations perform better than random configurations, with the position accuracy increased by 43.86% and the attitude accuracy increased by 14.29%.

This paper investigates the distributed convex optimization problem with coupled inequality constraints over unbalanced digraphs depicted as row-stochastic matrices, where each agent in the network only has access to its local information, while the local constraint functions of all the agents are coupled in inequality constraints. To solve this kind of problems, a novel distributed iterative algorithm is proposed via consensus scheme and the projected primal-dual subgradient method. Different from the previous related results, our algorithm can not only deal with coupling inequality constraints but also conquer the unbalanced topology caused by directed graphs. Moreover, it is proved that under certain assumptions, the optimal solution of the problem can be asymptotically obtained by performing the designed algorithm. Finally, the numerical examples are presented to further illustrate the efficacy of the proposed approach.

This paper proposed a dynamic traffic assignment method based on the connected transportation system to express the time-varied traffic flows caused by uncertain traffic demand and supply in the real traffic network accurately. In dynamic traffic assignment, because of the cumulative volumes of successive time intervals, a route-based traffic assignment algorithm is used for dynamic traffic assignment in this paper. Besides, a novel shortest path algorithm based actual and artificial link is proposed for the time-expanded network that consists of two kinds of arcs: the actual arc and the artificial one. Then, a simple example is used to illustrate the performance of the proposed method. The computational results show that the novel shortest path algorithm based actual and artificial link could be applied in the time-expanded networks with two types of arcs, and the route-based assignment method could achieve optimal balance solutions under different demand modes.

Considering the wear in mechanical bearings and the requirement for sensors or electrical power in active magnetic bearings, a novel nutation blood pump using a passive magnetic spherical bearing was developed in this paper. A mathematical model was derived to calculate the magnetic forces of the bearing between two pairs of magnetic sleeves in the nutation process. The calculation results demonstrate that the fluctuations of magnetic forces enlarge with the increase in the nutation angle; the magnetic forces obviously increase with the decrease in the air gap, especially along the z-axis. The dynamic magnetic finite element simulation was carried out to validate the mathematical model. The simulation and calculation results of the magnetic forces show consistent trends and provide a theoretical basis for the parameter design. To validate the performance of the pump, computational fluid dynamics (CFD) analyses and in vitro experiments were conducted. The predicted values of the flow velocity vector through the pump, and the wall shear stress, demonstrate that the pump has an antithrombotic property and would not cause serious blood damage. The hydraulic experiment shows that a pressure rise can be achieved in the range of 60-140 mmHg, at a rotational speed of 600-1600 rpm and a flow rate of 0.4-6.7 L/min. The normalized index of hemolysis (NIH) of the nutation pump was 0.0043 ± 0.0008 g/100L. The in vitro tests indicate the feasibility of a magnetically levitated ventricular-assist nutation blood pump for further suspension stability and animal trials.

The secondary control of microgrids (MGs) normally relies on the communication network to exchange information. Communication delay is an inherent characteristic during the signal transmission process, which may deteriorate the system dynamic performance and even destabilize the system. This paper studies the impact of communication delay on the small-signal stability of MGs with distributed secondary frequency and voltage control, and further investigates the corresponding delay compensation method. First, a time-delayed small-signal dynamic model of MGs considering both distributed secondary frequency and voltage controllers is developed. Based on the model, a comprehensive delay-dependent stability analysis is performed to reveal how the amount of communication delay, the topologies of communication networks, and the values of control parameters affect the system stability. Subsequently, a delay compensator based on lead-lag compensation along with gain regulation block is designed to enhance the system stability against the communication delay. Finally, time-domain simulation and experimental results are presented to validate the effectiveness of the proposed methods.

In order to solve the problem of lack of effective methods for ontology inconsistency, a user preferences-oriented ontology alignment repair model is proposed. This model uses 0-1 linear programming method to minimize the remove cost; the structure and source of ontology are used to measure the axiom of importance; Finally, by choosing the minimal conflict sets and the user preferences limit repair strategy, the purpose of eliminating inconsistency, reducing semantic loss and guaranteeing credibility is achieved. The experimental results show that this method can effectively solve the problem of ontology inconsistency, and the restored ontology is more suitable for user preferences than traditional methods.

The electron temperature of each subband is the key to understanding the characteristics of quantum cascade lasers (QCLs). We show herein a method, based on the particle swarm optimization algorithm, to obtain the electron temperature of the conduction band in the active region of QCLs. Accurate subband electron temperatures at room temperature, low temperature, and under various external electric fields are obtained. Furthermore, we discuss the variation and distribution of subband electron temperature that occurs during the operation of QCLs. The laser gain and I-V curves are obtained from the subband electron temperature. A comparison of these results with experimental results confirms that consideration of the subband electron temperature can improve our understanding of QCLs.

This paper proposes a trajectory optimization method considering gravitational orbit-attitude coupling (GOAC) for asteroid landing. First, by modeling the spacecraft as a rigid body rather than a point mass, and using the polyhedral method to describe the irregular gravitational field of the asteroid, GOAC is fully embedded into the 6 degree-of-freedom (DOF) dynamic model of the spacecraft. Second, attitude control, which is assumed to consume electric energy, is introduced into the optimization, the indirect method is adopted to formulate the 6-DOF fuel-optimal control problem, and the boundary conditions are modified to expand the search space of the problem. Third, a two-phase homotopic approach is proposed to realize a smooth continuation from the 6-DOF energy-optimal problem to the fuel-optimal problem. Finally, a numerical example is presented to demonstrate the feasibility of the proposed method, and the comparison results show that the obtained 6-DOF optimal trajectory can guarantee the fuel-optimality of the landing mission under GOAC.

This paper proposes a novel dynamic thermal management approach named Adaptive Periodic Thermal Management (APTM) for multi-core pipelined processors with hard real-time constraints. During online execution, the approach optimizes the peak temperature of the processor by dynamically calculating and deploying new adaptive periodic thermal management schemes, which periodically switch the stages to sleep state until next adapting. At each such instant of adapting, new APTM schemes which optimize the peak temperature under real-time constraints are calculated according to the extended pay-burst-only-once principle. To reduce the overhead of online computation, our approach utilizes thermal properties obtained from offline experiments in searching the ATPM scheme minimizing the temperature. We evaluate our APTM with simulation on the Intel SCC chip as well as real experiments on an Intel I7 processor. Compared to existing approaches, our approach reduces the peak temperature by up to 8 K for the Intel I7 processor, and 7.5 K for the Intel SCC using four to sixteen processor cores.

The modular multilevel converter (MMC) is now the most attractive topology for medium and high voltage power conversion applications with several advantages over the traditional voltage source converter (VSC). However, due to a large number of sub-modules (SMs) in the MMC, system reliability is a big challenge in its practical application, where each SM may be considered as a potential point of failure. In this paper, a reliability evaluation based on the Markov model is proposed for the MMC. The failure rates of the power electronic devices and SMs are firstly analyzed. Then, the Markov model and the state transition equation of the system are built in detail. A general reliability evaluation function is established, in which the mean time to failure and reliability evaluation of the MMC with redundant SMs are also discussed. Finally, a practical direct current (DC) distribution example for reliability evaluation is analyzed, and the results verify that the reliability evaluation based on the Markov model could provide a useful reference for project design.

Renewable energy is widely used as a clean energy in the world, yet the intermittent of the power generation from renewable energy power plants, resulting in poor quality of power supply. However, the hybrid energy system and some energy storage devices can be installed to mitigate the power fluctuation, to achieve power smoothing and maximize the profit, a double-objective optimization model of hybrid system integrated with micro-hydro power (MHP), photovoltaic (PV), and pumped hydro storage (PHS) is established in this study. The day-ahead optimized operation strategy of the PHS based-on hour-level data of seven days is obtained, which aims to maximize the similarity value (SIM) between the power generation curve and load profile, and the economic revenue (ER) of the PHS. Optimized schedule is obtained by means of chance-constrained programming (CCP) and modified multi-objective particle swarm optimization (MOPSO). The performance of the model is evaluated by using the meteorological, the historical MHP power generation and the load data obtained in Xiaojin County, Sichuan province, China. The simulation results reveal that the best trade-off value between SIM and ER can be obtained on the Pareto front with comprehensive consideration of the system operation strategy and target. Furthermore, the generation/load similarity can be increased 7.89% under the participating of the PHS.

All-optical modulation in a quantum cascade laser (QCL) can be rapidly evaluated using a photoluminescence (PL) spectrum. The PL spectrum reflects the electron transition rate of the sub-bands of the active region of a QCL. The peak position, amplitude, and displacement of the PL spectrum can be used to determine the optimal wavelength, spot position, the laser incident angle, optical power, and displacement in the electric field of the near-infrared laser. This PL spectrum approach is rapid, accurate, and simple. Furthermore, it can be used to optimize high-speed all-optical modulation of QCLs for use in free-space optical communication and molecular-detection applications.

To solve the problem that seriously degraded the kinematics performance may not meet the requirements of on-orbit operation tasks, a coping strategy for a multi-joint multi-type asynchronous failure of a space manipulator is proposed in this paper. Based on the entropy method, a comprehensive kinematics performance index (CKPI) is constructed, which can be regarded as the criterion for solving the artificial joint limits and the optimal locked angle of fault joint. And by solving them, a prevention strategy of serious kinematics performance degradation and a treatment strategy of joint free-swinging failure are established. Through synthesizing the above strategies, a coping strategy for a multi-joint multi-type asynchronous failure is constructed. The contributions of this paper are as follows. The constructed CKPI can make the kinematics performances optimal comprehensively in both operation space and joint space of a space manipulator, the method for calculating the artificial joint limits can solve the coupling problem during calculation, and the constructed coping strategy can ensure that the kinematics performance could meet the requirements of on-orbit operation tasks after multi-joint multi-type asynchronous failure. The correctness and effectiveness of the proposed strategy in this paper are verified by a seven-degree-of-freedom space manipulator.

This paper proposes a semi-slack scheduling framework for mixed-criticality systems with arbitrary task activation patterns. In particular, based on the schedulability test of arbitrary task activation model, we present a tight slack-reclaim scheme that can adaptively make use of system slack to improve the quality of service to low critical tasks. This scheme works at the moment that low critical tasks are supposed to be degraded by the offline schedulability test, while at other moments the system is scheduled by the earliest deadline first with virtual deadlines. The semi-slack scheduling scheme is first discussed in uniprocessor and later extended to multiprocessor. Extensive simulation results demonstrate that compared with some state-of-the-art scheduling approaches, the proposed semi-slack scheduling scheme efficiently reduces deadline misses of low critical tasks without jeopardizing the guarantee to critical tasks.