The authors propose a new defined tensile stress distribution for steel fiber reinforced concrete (SFRC) to evaluate the flexural strength of SFRC beams. It is a simplified formula based on the SFRC tensile stress – ...
详细信息
Backward Monte Carlo method of the complicated and exact three-dimensional turbine with the spectral emission and reflection characteristics of the turbine blades materials and the spectral absorption and emission cha...
详细信息
Backward Monte Carlo method of the complicated and exact three-dimensional turbine with the spectral emission and reflection characteristics of the turbine blades materials and the spectral absorption and emission characteristics of combustion gas is *** factors affecting the accuracy of the radiation temperature measurement are *** results show that reducing the distance from the probe to the target surface can reduce the effect of the environment on the measurement *** the temperature and emissivity of the target surface can improve the measurement *** reflection characteristics of the surfaces have little influence on the radiation temperature measurement,so the blades can be considered as diffuse reflectors in order to improve the calculation *** temperature measurement accuracy decreases rapidly as the temperature of the combustion gas *** temperature measurement accuracy decreases with the increase of total gas pressure and H_(2)O *** measuring the temperature of rotating blades,the apparent emissivity of the target surface is inversely proportional to the measurement accuracy.
In this work, we investigate the issue of inadequate solution quality and insufficient convergence in multi-sensor scheduling algorithms for tracking aerial moving targets, and enhance the traditional Non-dominated So...
详细信息
Deep reinforcement learning(DRL) achieves success through the representational capabilities of deep neural networks(DNNs). Compared to DNNs, spiking neural networks(SNNs),known for their binary spike information proce...
详细信息
Deep reinforcement learning(DRL) achieves success through the representational capabilities of deep neural networks(DNNs). Compared to DNNs, spiking neural networks(SNNs),known for their binary spike information processing, exhibit more biological characteristics. However, the challenge of using SNNs to simulate more biologically characteristic neuronal dynamics to optimize decision-making tasks remains, directly related to the information integration and transmission in *** by the advanced computational power of dendrites in biological neurons, we propose a multi-dendrite spiking neuron(MDSN) model based on Multi-compartment spiking neurons(MCN), expanding dendrite types from two to multiple and deriving the analytical solution of somatic membrane *** apply the MDSN to deep distributional reinforcement learning to enhance its performance in executing complex decisionmaking tasks. The proposed model can effectively and adaptively integrate and transmit meaningful information from different sources. Our model uses a bioinspired event-enhanced dendrite structure to emphasize features. Meanwhile, by utilizing dynamic membrane potential thresholds, it adaptively maintains the homeostasis of MDSN. Extensive experiments on Atari games show that the proposed model outperforms some state-of-the-art spiking distributional RL models by a significant margin.
作者:
Han, XinhuiPan, HaoyuanWang, ZhaoruiLi, JianqiangShenzhen University
College of Computer Science and Software Engineering Shenzhen China
Future Network of Intelligence Institute The School of Science and Engineering Shenzhen China Shenzhen University
National Engineering Laboratory for Big Data System Computing Technology The College of Computer Science and Software Engineering Shenzhen China
We investigate the timely status update in linear multi-hop wireless networks, where a source tries to deliver status update packets to a destination through a sequence of half-duplex relays. Timeliness is measured by...
详细信息
Feature selection is a crucial step in data preprocessing because feature selection reduces the dimensionality of data by eliminating irrelevant and redundant features. Since manual labeling is expensive, unsupervised...
详细信息
Feature selection is a crucial step in data preprocessing because feature selection reduces the dimensionality of data by eliminating irrelevant and redundant features. Since manual labeling is expensive, unsupervised feature selection has received increasing attention in recent years. However, existing unsupervised feature selection methods tend to prioritize selecting highly correlated features over exploring feature diversity. Thus, a regularized fractal autoencoder(RFAE) method is proposed to select informative features in an unsupervised way. Specifically, the fractal autoencoder network extends autoencoders to construct a correspondence neural network and a selection neural network. The correspondence neural network exploits interfeature correlations and the selection neural network selects the informative features. A redundancy regularization strategy consists of a redundancy elimination regularization term based on the dependency between features and a sparse regularization term based on the group lasso. The redundancy regularization strategy eliminates feature subset redundancy and enhances network generalization ability. Extensive experimental results on six publicly available datasets show that the proposed RFAE outperforms the compared methods regarding clustering accuracy and classification accuracy. Moreover, the proposed RFAE achieves acceptable computation efficiency.
In this work, La,Cu,Si-rich eutectic interfaces consisting of the refined α-Fe and LaCuSi phases are primarily constructed via adding Cu by selective laser melting to rapidly acquire a combination of strength and duc...
详细信息
Recently,the Third Pole(TP)region has experienced rapid environmental *** data are essential for hydrometeorological and ecological applications but still have large uncertainties on the TP owing to the heterogeneous ...
详细信息
Recently,the Third Pole(TP)region has experienced rapid environmental *** data are essential for hydrometeorological and ecological applications but still have large uncertainties on the TP owing to the heterogeneous land surface,complex terrain,and sparse weather *** this study,a long-term(1979–2020)high-resolution(1/30°)meteorological forcing dataset for the TP(TPMFD)was developed,as a sister to the widely used China Meteorological Forcing Dataset(CMFD).The TPMFD comprises seven components necessary for driving land surface *** have previously contributed precipitation and downward shortwave radiation data for the TPMFD,and this study presents the development of five other components and focuses on validations for all ***,2-meter air temperature,2-meter specific humidity,10-meter wind speed,and surface air pressure were generated by combining the fifth-generation atmospheric reanalysis for European Center for Medium-Range Weather Forecasts(ERA5),a short-term high-resolution atmospheric simulation,and in situ observations,and the downward longwave radiation was calculated using semi-physical *** cross-validation and independent-validation demonstrated that most variables in the developed dataset outperformed those in widely used reanalysis datasets,including ERA5,ERA5-Land,and the Global Land Data Assimilation system(GLDAS).This dataset is expected to be beneficial for climate analyses and modeling applications of land-surface processes on the TP.
Laser communication provides a promising solution for solving the communication blackout problem in reentry and hypersonic flight. Application of the orbital angular momentum (OAM) of laser has advantages in communica...
详细信息
Laser communication has potential to solve the communication blackout problem during hypersonic flight of reentry capsule or hypersonic vehicles. The influence of ablation particles from thermal protection materials o...
详细信息
暂无评论