Search

• A high-order model considering shear deformation and thickness stretching effect is established to investigate the bending and buckling responses of porous FG curved nanobeam. • The governing equation is derived based on the consistent two-phase local/non-local stress gradient theory, and the analytical solution of curved nanobeams is obtained by Navier's method. • The difference of buckling mode sensitivity between curved characteristics, scale effect and thickness stretching effect of the curved beam is analyzed. In this work, we provide an analysis of the bending and buckling properties of functionally graded (FG) nanobeams with trigonometric function-dependent porosity distribution. Herein, a high-order curved nanobeam model considering both shear deformations and thickness stretching effect is established. Two-phase local/nonlocal stress-driven gradient theory is utilized to capture small-scale effects. Hamilton's principle is adopted to obtain the governing equation and boundary conditions, and the analytical solutions, such as exact formulas for lateral displacement and critical buckling load are obtained by employing Navier's method for simply supported boundary conditions. Furthermore, the effects of local volume fraction, nonlocal parameter, material length scale parameter, shear deformation, material distribution, and porosity distribution pattern in stress-driven models are analyzed in detail. The parametric studies indicate that the thickness stretching effect has a different influence on bending and buckling. In addition, the thickness stretching effect and the opening angle show different coupling mechanisms during bending and buckling. On the other hand, the higher-order buckling mode leads to a more significant scale effect and thickness tensile effect, but reduces the influence of the opening angle. Finally, we find that due to the coupling effect between the pore distribution pattern and the orientation bias of the FG nanobeam, the variation of the material gradient index leads to a different order of the dimensionless critical buckling loads predicted by the three pore distributions. [ABSTRACT FROM AUTHOR]

As an emerging research topic, online class imbalance learning often combines the challenges of both class imbalance and concept drift. It deals with data streams having very skewed class distributions, where concept drift may occur. It has recently received increased research attention; however, very little work addresses the combined problem where both class imbalance and concept drift coexist. As the first systematic study of handling concept drift in class-imbalanced data streams, this paper first provides a comprehensive review of current research progress in this field, including current research focuses and open challenges. Then, an in-depth experimental study is performed, with the goal of understanding how to best overcome concept drift in online learning with class imbalance. [ABSTRACT FROM AUTHOR]

The article focuses on the Songtsam Shangri-La Lodge built by Baima Dorji in China where he converted his home into a boutique travel lodge. Topics mentioned include the opening of the hotel in 2001 with six lodges in the Diqing Tibetan Autonomous Prefecture of Yunnan, the aim to introduce Tibetan culture and make the hotels a spiritual haven, and the creating of a place where people can escape from the busy life in the cities.

• A unified size-dependent beam model incorporating various higher-order beam theories as well as Euler–Bernoulli and Timoshenko beam theories is developed for vibration analysis of nanobeams. • The governing equations are derived based on consistent two phase local/nonlocal strain- and stress-gradient theory incorporating surface effects. • The coupling effects of scale parameters and surface elasticity in aluminum and silicon materials are studied. In this work, a unified high-order nanobeam model considering various high-order shear deformation beam theories is established to investigate the vibration response of nanobeam on the basis of two-phase local/nonlocal strain and stress gradient theory, as well as surface elasticity theory. The unified model also includes the Euler-Bernoulli beam model and Timoshenko beam model. Herein, the elastic dynamics governing equations and boundary conditions are derived using Hamilton's principle, and the analytical solutions, such as exact formulas for natural frequencies, are obtained by employing the Navier method for simply supported boundary conditions. The effects of local volume fraction, nonlocal parameter, material length scale parameter, shear deformation and surface energy in stress and strain-driven models are analyzed in detail, respectively. The parametric studies reveal that the two scale parameters (nonlocal parameters and material length characteristic parameters) have opposite effects on the stiffness of the nanobeams in the two driving models, while the surface parameters have the same effect on the stiffness of the two driving models. The influence of the slenderness ratio on the surface effect and scale effect is opposite, meaning that the increase of the slenderness ratio deepens the influence of the surface effect but weakens the influence of the scale effect. There are also differences in the effects of higher-order modes on the two effects. Higher modes lead to more significant scale effects, but the effect of higher modes on surface effects depends on the surface elastic properties of the material. We also find that the introduction of surface elasticity increases the gap between the TBT and other higher-order beams, which indicates the prediction results of the higher-order beam model are more accurate when both surface and nonlocal effects are considered. In addition, it is represented that the surface elasticity makes aluminum nanobeams exhibit a stiffness softening effect, while the effect of surface elasticity on the stiffness of silicon nanobeams is significantly dependent on the slenderness ratio and the number of modes. [ABSTRACT FROM AUTHOR]

In this article, the underwater target tracking control problem of a biomimetic underwater vehicle (BUV) is addressed. Since it is difficult to build an effective mathematic model of a BUV due to the uncertainty of hydrodynamics, target tracking control is converted into the Markov decision process and is further achieved via deep reinforcement learning. The system state and reward function of underwater target tracking control are described. Based on the actor–critic reinforcement learning framework, the deep deterministic policy gradient actor–critic algorithm with supervision controller is proposed. The training tricks, including prioritized experience replay, actor network indirect supervision training, target network updating with different periods, and expansion of exploration space by applying random noise, are presented. Indirect supervision training is designed to address the issues of low stability and slow convergence of reinforcement learning in the continuous state and action space. Comparative simulations are performed to show the effectiveness of the training tricks. Finally, the proposed actor–critic reinforcement learning algorithm with supervision controller is applied to the physical BUV. Swimming pool experiments of underwater object tracking of the BUV are conducted in multiple scenarios to verify the effectiveness and robustness of the proposed method. [ABSTRACT FROM AUTHOR]

The article deals with the challenges of wetlands protection in China. Chen Kelin, director of Wetlands International, has raised a concern on the change in focus in the No. 1 Document of 2011 to water conservancy infrastructure construction. Threats to wetlands that are discussed include the growing population, rapid economic development, and over reclamation and cultivation of wastelands. Indications of the Chinese government's increased emphasis on the restoration of wetlands' ecological functions are identified.

Few-shot learning, aiming to learn novel concepts from one or a few labeled examples, is an interesting and very challenging problem with many practical advantages. Existing few-shot methods usually utilize data of the same classes to train the feature embedding module and in a row, which is unable to learn adapting to new tasks. Besides, traditional few-shot models fail to take advantage of the valuable relations of the support-query pairs, leading to performance degradation. In this article, we propose a transductive relation-propagation graph neural network (GNN) with a decoupling training strategy (TRPN-D) to explicitly model and propagate such relations across support-query pairs, and empower the few-shot module the ability of transferring past knowledge to new tasks via the decoupling training. Our few-shot module, namely TRPN, treats the relation of each support-query pair as a graph node, named relational node, and resorts to the known relations between support samples, including both intraclass commonality and interclass uniqueness. Through relation propagation, the model could generate the discriminative relation embeddings for support-query pairs. To the best of our knowledge, this is the first work that decouples the training of the embedding network and the few-shot graph module with different tasks, which might offer a new way to solve the few-shot learning problem. Extensive experiments conducted on several benchmark datasets demonstrate that our method can significantly outperform a variety of state-of-the-art few-shot learning methods. [ABSTRACT FROM AUTHOR]

Accurate wind speed forecasting is of great significance to ensure the safe utilization of wind power. However, the randomness and volatility nature of wind speed give rise to an enormous challenge to the precision of wind speed forecasting. Combining the data preprocess technology, feature selection method, forecasting model, optimization algorithm and data postprocessing technology, the complete ensemble empirical mode decomposition with adaptive noise-least absolute shrinkage and selection operator-quantile regression neural network (CEEMDAN-LASSO-QRNN) model is developed to preform multistep wind speed probabilistic forecasting. Within the proposed model, CEEMDAN technology is firstly employed to decompose original wind speed timeseries into several subsequences. For each subsequence, the explanatory variables constructed by a hybrid multistep forecasting strategy are selected by LASSO regression. Subsequently, QRNN forecasting models are established to obtain multistep conditional quantiles predictions for entire subsequences. Ultimately, the aggregated quantiles are served as the samples to fit approximate distribution through kernel density estimation (KDE), thus obtaining the probability density function, further achieving probabilistic predictions, interval predictions and point predictions. The case studies including four real datasets are provided to validate the dependability and feasibility of the proposed model. Experimental results indicate higher accuracy and robustness of the proposed model occur in multistep wind speed probabilistic forecasting. • The cooperative ensemble method is an ingenious integration of several single methods. • CEEMDAN is used to decompose intricate original data into several sample subseries. • Each subseries with the optimal parameters promises a more superior performance. • Multistep wind speed probabilistic forecasting is realized by DR strategy and KDE. • A case including four datasets verifies the performance of CEEMDAN-LASSO-QRNN model. [ABSTRACT FROM AUTHOR]

In this paper, a new numerical scheme based on weighted and shifted Grünwald formula and compact difference operate is proposed. The proposed numerical scheme is used to solve time fractional partial integro-differential equation with a weakly singular kernel. Meanwhile the time fractional derivative is denoted by the Riemann-Liouville sense. Subsequently, we prove the stability and convergence of the mentioned numerical scheme and show that the numerical solution converges to the analytical solution with order O (τ 2 + h 4), where τ and h are time step size and space step size, respectively. The advantage is that the accuracy of the suggested schemes is not dependent on the fractional α. Furthermore, the numerical example shows that the method proposed in this paper is effective, and the calculation results are consistent with the theoretical analysis. • A new numerical scheme based on compact finite difference for solving partial integro-differential equation are obtained. • The advantage is that the accuracy of the suggested schemes is not dependent on α. • The proposed method can be used in many fields of differential equations, such as fractional nonlinear problems. [ABSTRACT FROM AUTHOR]

Infrared and visible image fusion methods aim to combine salient target instances and abundant texture details into fused images. However, due to the interference of harsh conditions, such as dense smoke, fog, and intense light, it is feasible for external interference information to be integrated into the fused image, which seriously affects the image quality. To this end, we propose an infrared and visible image fusion network with visual perception and cross-scale attention modules, termed VCAFusion, which integrates critical information from source images in harsh conditions more efficiently. Specifically, considering that the human eye can identify key information under adverse conditions, we design a visual perception module (VPM), guiding information integration from the perspective of human visual perception. In addition, we propose a cross-scale attention module (CSAM) based on shifted window cross-attention, which aims to capture the long-distance correlation between adjacent scale features, providing more accurate image information for image restoration. Experiments with a variety of datasets reveal that the VCAFusion can adaptively retain image information and further improve image generation ability in harsh conditions. [ABSTRACT FROM AUTHOR]

Low-rank Multiview Subspace Learning (LMvSL) has shown great potential in cross-view classification in recent years. Despite their empirical success, existing LMvSL-based methods are incapable of handling well view discrepancy and discriminancy simultaneously, which, thus, leads to performance degradation when there is a large discrepancy among multiview data. To circumvent this drawback, motivated by the block-diagonal representation learning, we propose structured low-rank matrix recovery (SLMR), a unique method of effectively removing view discrepancy and improving discriminancy through the recovery of the structured low-rank matrix. Furthermore, recent low-rank modeling provides a satisfactory solution to address the data contaminated by the predefined assumptions of noise distribution, such as Gaussian or Laplacian distribution. However, these models are not practical, since complicated noise in practice may violate those assumptions and the distribution is generally unknown in advance. To alleviate such a limitation, modal regression is elegantly incorporated into the framework of SLMR (termed MR-SLMR). Different from previous LMvSL-based methods, our MR-SLMR can handle any zero-mode noise variable that contains a wide range of noise, such as Gaussian noise, random noise, and outliers. The alternating direction method of multipliers (ADMM) framework and half-quadratic theory are used to optimize efficiently MR-SLMR. Experimental results on four public databases demonstrate the superiority of MR-SLMR and its robustness to complicated noise. [ABSTRACT FROM AUTHOR]

Abstract: Capacitated Arc Routing Problem (CARP) has drawn much attention during the last few years. In addition to the goal of minimizing the total cost of all the routes, many real-world applications of CARP also need to balance these routes. The Multi-objective CARP (MO-CARP) commonly exists in practical applications. In order to solve MO-CARP efficiently and accurately, this paper presents a Multi-population Cooperative Coevolutionary Algorithm (MPCCA) for MO-CARP. Firstly, MPCCA applies the divide-and-conquer method to decompose the whole population into multiple subpopulations according to their different direction vectors. These subpopulations evolve separately in each generation and the adjacent subpopulations can share their individuals in the form of cooperative subpopulations. Secondly, multiple subpopulations are used to search different objective subregions simultaneously, so individuals in each subpopulation have a different fitness function, which can be modeled as a Single Objective CARP (SO-CARP). The advanced MAENS approach for single-objective CARP can be used to search each objective subregion. Thirdly, the internal elitism archive is used to construct evolutionary pool for each subregion, which greatly speeds up the convergence. Lastly, the fast nondominated ranking and crowding distance of NSGA-II are used for selecting the offspring and keeping the diversity. MPCCA is tested on 91 CARP benchmarks. The experimental results show that MPCCA obtains better generalization performance over the compared algorithms. [Copyright &y& Elsevier]

Abstract: This paper focuses on the modeling of undulation propulsion and a neural network (NN) yaw controller based on this model. First, because of the importance of the first link’s oscillation in yaw control, a motion model of free swimming was built based on the Lagrangian function, and the coupled dynamic and kinematic functions were calculated based on the relation between the generalized force and the fluid force. Second, a neural network was trained through the data generated from this model to obtain a predictive yaw controller that could control the orientation by the different offsets of each link. Finally, simulations were conducted to demonstrate the performance of the controller. [Copyright &y& Elsevier]