基于CNN-GRU-Attention的民机重着陆预测模型

重着陆事件是民用飞机着陆阶段最容易出现的事故之一,对其进行预测对于提升飞行安全具有重大意义。因此提出了一种基于卷积神经网络(Convolution Neural Network,CNN)和门控循环单元(Gated Recurrent Unit,GRU)并融合了注意力机制的重着陆预测模型,以飞机实际运行过程中采集的快速存取记录器(Quick Access Recorder,QAR)数据作为数据集。对QAR数据进行预处理并通过轮载信号确定数据集区间;通过Spearman相关系数对与飞机着陆相关的参数进行相关性分析,从中提取了24个特征参数作为预测模型输入,以重着陆判定指标垂直加速度作为输出,建立基于CNN-GRU-Attention的重着陆预测模型。实验结果表明,所提模型与其他预测模型相比具有更好的预测效果。

Pitch-derived soft carbon composite with hard carbon fibers for high-rate and long-cycling K+storage

Potassium-ion batteries(KIBs)have been seen as a competitive alternative to lithium-ion batteries(LIBs)due to their natural abundance,low cost and rocking chair-like operating mechanism similar to LIBs.Soft carbon has a lower voltage plateau compared to hard carbon and an easily modulated lattice structure compared to graphite,which provides particular advantages in KIBs anodes.Pitch has attracted much attention as a simple,readily available and inexpensive precursor for soft carbon,but its structure is easily damaged during cycling.Herein,the flexible film Pitch@CNF are prepared by uniformly winding reticulated carbon fibers on the surface of pitch-soft carbon via electrostatic spinning technique,which not only enables the pitch to maintain its structure well during cycling and withstand the volume expansion upon K+insertion,but also is conducive to ionic transport of the three-dimensional reticulated structure.Meanwhile,the abundant pores on the carbon fibers can provide more K+active sites.The prepared flexible self-supporting films can be used directly as electrodes without the addition of binders and conductive agents.The reversible capacity is 290 mAh·g^(-1)at a current density of 0.1 A·g^(-1),and the capacity retention rate is 83%after 500 cycles.

Deformation characteristics and damage ontologies of soft and hard composite rock masses under impact loading

As one of the most common occurring geological landforms in deep rock formations, the dynamic mechanical properties of layered composite rock bodies under impact loading have been widely studied by scholars. To study the dynamic properties of soft and hard composite rocks with different thickness ratios, this paper utilizes cement, quartz sand and gypsum powder to construct soft and hard composite rock specimens and utilizes a combination of indoor tests, numerical calculations, and theoretical analyses to investigate the mechanical properties of soft and hard composite rock bodies. The test results reveal that:(1) When the proportion of hard rock increases from 20% to 50%, the strength of the combined rock body increases by 69.14 MPa and 87 MPa when the hard rock face and soft rock face are loaded, respectively;however, when the proportion of hard rock is the same, the compressive strength of the hard rock face impact is 9%-17% greater than that of the soft rock face impact;(2) When a specimen of soft and hard combined rock body is subjected to impact loading, the damage mode involves mixed tension and shear damage, and the cracks generally first appear at the ends of the specimen, then develop on the laminar surface from the impact surface, and finally end in the overall damage of the soft rock part. The development rate and the total number of cracks in the same specimen when the hard rock face is impacted are significantly greater than those when the soft rock face is impacted;(3) By introducing Weibull’s statistical strength theory to establish the damage variables of soft-hard combined rock bodies, combined with the DP strength criterion, the damage model and the Kelvin body are concatenated to obtain a statistical damage constitutive model, which can better fit the full stress-strain curve of soft-hard combined rock body specimens under a single impact load.

Integrated simulation and monitoring to analyze failure mechanism of the anti-dip layered slope with soft and hard rock interbedding

The significant difference between the mechanical properties of soft rock and hard rock results in the complexity of the failure mode of the anti-dip layered slope with soft and hard rock interbedding.In order to reveal the landslide mechanism,taking the north slope of Fushun West Open-pit Mine as an example,this paper analyzed the failure mechanism of different landslides with monitoring and field surveys,and simulated the evolution of landslides.The study indicated that when the green mudstone(hard rock)of the anti-dip slope contains siltized intercalations(soft rock),the existence of weak layers not only aggravates the toppling deformation of anti-dip layered slope with high dip,but also causes the shear failure of anti-dip layered slope with stable low dip.The shear failure including subsidence induced sliding and wedge failure mainly exists in the unloading zone of the slope.Its failure depth and failure time were far less than that of toppling failure.In terms of the development characteristics of deformation,toppling deformation has the long-term and progressive characteristics,but shear failure deformation has the abrupt and transient characteristics.This study has deepened the understanding of such slope landslide mechanism,and can provide reference for similar engineering.

Theoretical study on stability evolution of soft and hard interbedded bedding reservoir slopes

Soft and hard interbedded bedding rock slopes,which is prone to failure,are widely distributed in the Three Gorges Reservoir,China.Limit equilibrium method(LEM)is commonly used to analyze the stability of bedding rock slopes that have a single failure plane.However,this method cannot accurately estimate the stability of soft and hard interbedded bedding reservoir slopes because the strength parameters of a soft and hard interbedded rock mass vary spatially along the bedding plane and deteriorate with time due to periodic fluctuations of reservoir level.A modified LEM is proposed to evaluate the stability evolution of soft and hard interbedded bedding reservoir slopes considering the spatial variation and temporal deterioration of shear strength parameters of rock masses and bedding planes.In the modified LEM,the S-curve model is used to define the spatial variation of shear strength parameters,and general deterioration equations of shear strength parameters with the increasing number of wettingdrying cycles(WDC)are proposed to describe the temporal deterioration.Also,this method is applied to evaluate the stability evolution of a soft and hard interbedded bedding reservoir slope,located at the Three Gorges Reservoir.The results show that neglecting the spatial variation and temporal deterioration of shear strength parameters may overestimate slope stability.Finally,the modified LEM provides useful guidance to reasonably evaluate the long-term stability of soft and hard interbedded bedding reservoir slopes in reservoir area.

基于PSO-Soft attention双向LSTM算法的光伏发电量预测研究

为解决传统的光伏发电量预测中数据特征方程较为粗糙、预测精度不高等问题,提出一种基于PSO-Soft attention双向LSTM算法的光伏发电功率预测算法,在前向传播过程中通过自适应学习策略的PSO算法对LSTM模型的关键参数进行寻优,同时优化了双向LSTM网络的反向传播过程,加入了Soft attention机制使模型收敛性更强,重新构造该模型为八层网络结构且采用两层卷积层,同时解析了PSO-Soft attention双向LSTM算法的光伏发电功率预测过程。实验结果表明,相较于SVM模型以及LSTM模型,所提预测模型的预测精度提升了2%左右。

Chinese Named Entity Recognition with Character-Level BLSTM and Soft Attention Model

Unlike named entity recognition(NER)for English,the absence of word boundaries reduces the final accuracy for Chinese NER.To avoid accumulated error introduced by word segmentation,a deep model extracting character-level features is carefully built and becomes a basis for a new Chinese NER method,which is proposed in this paper.This method converts the raw text to a character vector sequence,extracts global text features with a bidirectional long short-term memory and extracts local text features with a soft attention model.A linear chain conditional random field is also used to label all the characters with the help of the global and local text features.Experiments based on the Microsoft Research Asia(MSRA)dataset are designed and implemented.Results show that the proposed method has good performance compared to other methods,which proves that the global and local text features extracted have a positive influence on Chinese NER.For more variety in the test domains,a resume dataset from Sina Finance is also used to prove the effectiveness of the proposed method.

Defining soft skills in nursing:a concept analysis

Objective:To define soft skills in nursing.Methods:This is a concept analysis using the Walker and Avant approach.Multiple electronic databases in the English language including CINAHL,Science Direct,MEDLINE on EBSCOhost,and Scopus were consulted.Results:Soft skills in nursing include the intrapersonal attributes,interpersonal skills,and creativity of the nurse,which,when combined with professionalism,teamwork,and effective communication skills,improve the quality of nursing care and client satisfaction.Conclusions:Although hard and soft skills are compatible and synergetic,soft skills are to be prioritized for the success of nursing care in the healthcare industry.This concept analysis adds to the body of knowledge by identifying the attributes of soft skills in nursing and revealing that a good nurse needs more than just good grades to be successful.

Constitutive model for methane desorption and diffusion based on pore structure differences between soft and hard coal

This paper aims to improve the accuracy and applicability of gas diffusion mathematical models from coal particles. Firstly, a new constitutive model for gas diffusion from coal particles with tri-disperse pore structure is constructed by considering the difference in characteristics between soft coal and hard coal.The analytical solution is then derived, that is, the quantitative relationship between gas diffusion rate(Qt/Q_∞) and diffusion time(t), The pore structure parameters of soft coal and hard coal from Juji coal mine are determined. Gas diffusion rules are numerically calculated and investigated by physical simulation methods. Lastly, the applicability of this model is verified. The results show that the homogeneous model only applies to the gas diffusion process of hard coal during the initial 10 min. The calculation results from this model and the physical experimental results of soft coal and hard coal are nearly identical during the initial 30 min.

Joint training with local soft attention and dual cross-neighbor label smoothing for unsupervised person re-identification

Existing unsupervised person re-identification approaches fail to fully capture thefine-grained features of local regions,which can result in people with similar appearances and different identities being assigned the same label after clustering.The identity-independent information contained in different local regions leads to different levels of local noise.To address these challenges,joint training with local soft attention and dual cross-neighbor label smoothing(DCLS)is proposed in this study.First,the joint training is divided into global and local parts,whereby a soft attention mechanism is proposed for the local branch to accurately capture the subtle differences in local regions,which improves the ability of the re-identification model in identifying a person’s local significant features.Second,DCLS is designed to progressively mitigate label noise in different local regions.The DCLS uses global and local similarity metrics to semantically align the global and local regions of the person and further determines the proximity association between local regions through the cross information of neighboring regions,thereby achieving label smoothing of the global and local regions throughout the training process.In extensive experiments,the proposed method outperformed existing methods under unsupervised settings on several standard person re-identification datasets.

Comparison of Starch Granule Size Distribution Between Hard and Soft Wheat Cultivars in Eastern China

Granule size distribution of wheat starch is an important characteristic that can affect its chemical composition and functionality. Two types of wheat cultivars, the hard and soft wheat cultivars, grown at Tai＇an Experimental Station of Shandong Agricultural University, Taian, Shandong, China, were examined in this study. The granule size distribution and amylose contents in wheat grains were studied and compared, and relationships between the properties were identified. A clear bimodal distribution of granule size was shown in all wheat cultivars. Volume distribution of starch granules shows the typical bimodal with peak values in the ranges of 5.6-6.1μm and 20.7-24.9μm, respectively. Also, granule surface area distribution was bimodal with peak values in the ranges of 2.4-3.2μm and 20.7-24.9μm, respectively. Number distribution of granules was a typical population with a peak value in the range of 0.54-1.05μm. Contributions from the granules 〈 2.8μm and 〈 9.9μm to the total volume were in the ranges of 94.2-95.1% and 99.7-99.9% of total number, respectively. Proportions of granules〈2.8μm, 2.8-9.9μm, 9.9-22.8μm, and 22.8-42.8μm were in the ranges of 12.9-14.3%, 28.4-31.1%, 33.5-35.6%, and 19.7-22.7% for hard wheat, and 10.3-13.9%, 26.6-28.1%, 32.7-34.6%, and 24.2-27% for soft wheat. Hard wheat had greater B-type granules （ 〈 9.9μm）, and had fewer granules of 22.8-42.8μm than soft wheat. Amylose content was positively related to volume percentage of granules 22.8-42.8μm, and negatively related to volume percentage of granules 2.8-22.8μm.

Excavation-induced microseismicity and rockburst occurrence:Similarities and differences between deep parallel tunnels with alternating soft-hard strata

Excavation-induced microseismicity and rockburst occurrence in deep underground projects provide invaluable information that can be used to warn rockburst occurrence,facilitate rockburst mitigation procedures,and analyze the mechanisms responsible for their occurrence.Based on the deep parallel tunnels with the maximum depth of 1890 m created as part of the Neelum–Jhelum hydropower project in Pakistan,similarities and differences on excavation-induced microseismicity and rockburst occurrence between parallel tunnels with soft and hard alternant strata are studied.Results show that a large number of microseismic(MS)events occurred in each of the parallel tunnels during excavation.Rockbursts occurred most frequently in certain local sections of the two tunnels.Significant differences are found in the excavation-induced microseismicity(spatial distribution and number of MS events,distribution of MS energy,and pattern of microseismicity variation)and rockbursts characteristics(the number and the spatial distribution)between the parallel tunnels.Attempting to predict the microseismicity and rockburst intensities likely to be encountered in subsequent tunnel based on the activity encountered when the parallel tunnel was previously excavated will not be an easy or accurate procedure in deep tunnel projects involving complex lithological conditions.

Mixed-Weights Least-Squares Stable Predictive Control Algorithm with Soft and Hard Constraints

Mixed-weight least-squares (MWLS) predictive control algorithm, compared with quadratic programming (QP) method, has the advantages of reducing the computer burden, quick calculation speed and dealing with the case in which the optimization is infeasible. But it can only deal with soft constraints. In order to deal with hard constraints and guarantee feasibility, an improved algorithm is proposed by recalculating the setpoint according to the hard constraints before calculating the manipulated variable and MWLS algorithm is used to satisfy the requirement of soft constraints for the system with the input constraints and output constraints. The algorithm can not only guarantee stability of the system and zero steady state error, but also satisfy the hard constraints of input and output variables. The simulation results show the improved algorithm is feasible and effective.

基于Attention-LSTM-Kalman建模的风洞动态流量软测量

针对风洞流量测量中传统静态软测量模型估计精度低、鲁棒性差等问题,提出了注意力机制(Attention mechanism,Attention)、长短时记忆神经网络(Long short-term memory,LSTM)和卡尔曼滤波(Kalman filtering,Kalman)结合的Attention-LSTM-Kalman软测量模型:通过LSTM网络建立静态软测量模型,在此基础上,提出一种基于注意力机制的改进方案,考虑到系统的动态特性,使用卡尔曼滤波动态调整软测量模型输出序列。实验结果表明,静态预测模型LSTM的预测效果优于循环神经网络(Recurrent neural network,RNN)和门控循环单元(Gated recurrent unit,GRU)等模型;基于LSTM、Attention-LSTM和Attention-LSTM-Kalman的3种模型的对比预测测量结果表明,注意力机制能有效提高模型精准度,引入卡尔曼滤波改善了模型的动态测量特性。该模型方案在风洞系统的流量测量验证了其可行性和有效性。

Influence of confining stress on fracture characteristics and cutting efficiency of TBM cutters conducted on soft and hard rock

Combined with numerical simulation, the influence of confining stress on cutting process, fracture conditions and cutting efficiencies of soft and hard rock has been conducted on the triaxial testing machine(TRW-3000) designed and manufactured in Central South University(China). Results are obtained by performing analysis on the fracture scopes of cement and granite plates,the characteristics of cutting force in cutting processes and the cutting efficiency. Firstly, the increase of latitude fracture scope and the decrease of longitude fracture scope are both more notable in the tests conducted on cement plates subjected to the increasing confining stresses; secondly, the increase tendency of peak penetration forces obtained from tests conducted on granite plates is more obvious, however, the increase tendencies of average penetration forces achieved from cement and granite plates are close to each other; thirdly, the cutting efficiency could be improved by increasing the spacing between cutters when the confining stress which acts on soft and hard rock increases in a certain degree, and the cutting efficiency of soft rock is more sensitive to the varying confining stresses.

Experimental analysis of pore structure and fractal characteristics of soft and hard coals with same coalifcation

Accurate and quantitative investigation of the physical structure and fractal geometry of coal has important theoretical and practical signifcance for coal bed methane(CBM)development and the prevention of dynamic disasters such as coal and gas outbursts.This study investigates the pore structure and fractal characteristics of soft and hard coals using nitrogen and carbon dioxide(N_(2)/CO_(2))adsorption.Coal samples from Pingdingshan Mine in Henan province of China were collected and pulverized to the required size(0.20–0.25 mm).N_(2)/CO_(2)adsorption tests were performed to evaluate the specifc surface area(SSA),pore size distribution(PSD),and pore volume(PV)using Braunuer-Emmett-Teller(BET),Barrett-Joyner-Halenda(BJH),and Density Functional Theory(DFT).The pore structure was characterized based on the theory of fractal dimensions.The results unveiled that the strength of coal has a signifcant infuence on pore structure and fractal dimensions.There are signifcant diferences in SSA and PV between both coals.The BJH-PV and BET-SSA obtained by N_(2)-adsorption for soft coal are 0.029–0.032 cm^(3)/g and 3.523–4.783 m^(2)/g.While the values of PV and SSA obtained by CO_(2)-adsorption are 0.037–0.039 cm^(3)/g and 106.016–111.870 m^(2)/g.Soft coal shows greater SSA and PV than hard coal,which is consistent with the adsorption capacity(VL).The fractal dimensions of soft and hard coal are respectively diferent.The Ding coal exhibits larger D1 and smaller D_(2),and the reverse for the Wu coal seam is observed.The greater the value of D1(complexity of pore surface)of soft coal is,the larger the pore surface roughness and gas adsorption capacity is.The results enable us to conclude that the characterization of pores and fractal dimensions of soft and hard coals is diferent,tending to diferent adsorption/desorption characteristics.In this regard,the results provide a reference for formulating corresponding coal and gas outburst prevention and control measures.

Numerical simulation for rheological characteristics of alternating distribution of soft and hard rock layers

The difficulty of selecting appropriate rheological model and parameters for the alternating distribution of soft and hard rock layers was often encountered due to the unhomogeneity, discontinuity and anisotropy of rock mass. The Burgers and generalized Kelvin models were applied to the soft and hard rock respectively and the rheological parameters were obtained based on the method of optimum separation. By using the simulated code FLAC^(3D), the stability of surrounding rocks of a certain underground plant was analyzed. The effect of surface load and weakening the parameters intensity of argillaceous and bioclastic interlayers between soft and hard rock on rheological behavior of layer composite rock mass was investigated. The results indicate that the rheological characteristics of soft and hard rock layer in composite rock mass can be described well with above two rheological models.

Contrastive Study of Metadiscourse in the Conclusion of Academic Papers between Soft Science and Hard Science

Academic paper as a genre is communicative.Interactional metadiscourse is one of the linguistic features that can significantly show the communicative feature of texts.With different research methodologies and characteristics,academic papers of hard and soft science present different degree of communicative features.The corpus of this study consists of 16 academic papers among which 8 are from hard science and 8 from soft science.This article aims to explore the different ways of scholars from soft and hard science to use metadiscourse in the conclusion part of their academic papers to exert lights to the pedagogical development of academic papers.

Magnetic Properties Hard-Soft SmCo₅-FeNi and SmCo₅-FeCo Composites Prepared by Electroless Coating Technique

Composites of SmCo5-FeNi and SmCo5-FeCo, hard-soft magnetic materials, have been synthesized via electroless plating of magnetically hard SmCo5 powder particles with magnetically soft FeNi and FeCo, respectively. The influence of coating thickness of soft magnetic layers on the structure and magnetic properties of the composite has been studied. Overall FeNi coating was found to be less dense compared to FeCo for the same plating duration. Structurally the coat ing was found to be nodular in morphology. These coating have dramatic effect on the overall magnetic property of the composite. As compared to FeNi coated SmCo5 composite, two-fold increase in the saturation magnetization has been observed upon coating SmCo5 (Ms^28 emu/g) with FeCo to a value 56 emu/g. The coercivity of composite powder was found to decrease with increasing the coating layer thickness. The absence of exchange spring behavior in the hard-soft composite is attributed to magnetically soft layer thickness exceeding the theoretical length limit for exchange-spring coupling.

Proteomic analysis reveals the differential proteins of endosperm between soft and hard wheat varieties

In this study,we revealed the differential proteins from the wheat endosperms using proteomic analysis and investigated their surface properties.The pattern of the polypeptides obtained from the Yangmai-15 and Yangmai-16 wheat varieties were compared using two-dimensional polyacrylamide gels.In addition,we compared the characteristics of the grain such as grain hardness,protein content,wet gluten,dough development time,dough stability,gliadin and glutenin contents between Yangmai-15 and Yangmai-16,and the results were significantly different.Notably,216 and 197 protein spots were separated from Yangmai-15 and Yangmai-16,respectively.The isoelectric points of the identified proteins ranged from 4 to 10 and the molecular weights of proteins varied from 10 to 100 kDa.Further,21 and 8 specific differential protein spots were identified fromthe flour of Yangmai-15 and Yangmai-16,respectively.The surface properties of identified peptides consisted of hydrophobic or hydrophilic residues,as well as randomly scattered residues.The proteomic analysis of the wheat endosperms provides a novel insight into the biochemical basis for the differences in physicochemical properties between the soft and hard wheat varieties.