Railway passenger flow forecasting can help to develop sensible railway schedules,make full use of railway resources,and meet the travel demand of passengers.The structure of passenger flow in railway networks and the...Railway passenger flow forecasting can help to develop sensible railway schedules,make full use of railway resources,and meet the travel demand of passengers.The structure of passenger flow in railway networks and the spatiotemporal relationship of passenger flow among stations are two distinctive features of railway passenger flow.Most of the previous studies used only a single feature for prediction and lacked correlations,resulting in suboptimal performance.To address the above-mentioned problem,we proposed the railway passenger flow prediction model called Flow-Similarity Attention Graph Convolutional Network(F-SAGCN).First,we constructed the passenger flow relations graph(RG)based on the Origin-Destination(OD).Second,the Passenger Flow Fluctuation Similarity(PFFS)algorithm is used to measure the similarity of passenger flow between stations,which helps construct the spatiotemporal similarity graph(SG).Then,we determine the weights of the mutual influence of different stations at different times through an attention mechanism and extract spatiotemporal features through graph convolution on the RG and SG.Finally,we fused the spatiotemporal features and the original temporal features of stations for prediction.The comparison experiments on a railway bureau’s accurate railway passenger flow data show that the proposed F-SAGCN method improved the prediction accuracy and reduced the mean absolute percentage error(MAPE)of 46 stations to 7.93%.展开更多
The aim of our present study was to construct genetic structure and relationships among Chinese fine-wool sheep breeds. 46 individuals from 25 breeds or strains were genotyped based on the Illumina Ovine 50K SNP array...The aim of our present study was to construct genetic structure and relationships among Chinese fine-wool sheep breeds. 46 individuals from 25 breeds or strains were genotyped based on the Illumina Ovine 50K SNP array. Meanwhile, genetic variations among 482 individuals from 9 populations were genotyped with 10 microsatellites. In this study, we found high genetic polymorphisms for the microsatellites, while 7 loci in the Chinese superfine Merino strain （Xinjiang types） （CMS） and 5 loci in Gansu alpine superfine-wool sheep strain （GSS） groups were found deviated from Hardy-Weinberg equilibrium （HWE）. Genetic drift FsT=0.019 （P〈0.001） and high gene flows were detected in all the 7 fine-wool sheep populations. Phylogenetic analysis showed fine-wool sheep populations were clustered in a group independent from the Chinese indigenous breeds such that the 7 fine-wool sheep clustered distinct from Liangshan semifine-wool sheep （LS） and Hu sheep （HY） reflected by different population differentiation analyses. Overall, our findings suggested that all fine-wool sheep populations have close genetic relationship, which is consistent with their breeding progress. These populations, therefore, can be regarded as open-breeding populations with high levels of gene flows. Furthermore, the two superfine-wool strains, viz., CMS and GSS, might be formed by strong artificial selection and with frequent introduction of Australian Merino. Our results can assist in breeding of superfine-wool sheep and provide guidance for the cultivation of new fine-wool sheep breeds with different breeding objectives.展开更多
The hot compression experiments were performed to investigate the effects of hot deformation parameters on the flow stress of BT20(Ti-6Al-2Zr-1Mo-1V) titanium alloy. The results show that the flow stress decreases wit...The hot compression experiments were performed to investigate the effects of hot deformation parameters on the flow stress of BT20(Ti-6Al-2Zr-1Mo-1V) titanium alloy. The results show that the flow stress decreases with the increment of deformation temperature and increases with the growth of strain rate. The peak stress moves toward the direction of strain reducing and the strain rate sensitivity increases with the rising deformation temperature. There is obvious deformation heating created during hot deformation under relatively higher strain rate and lower deformation temperature. The improved back propagation(BP) neural network with 3-20-16-1 architecture has been employed to establish the prediction model of flow stress using deformation degree, deformation temperature and strain rate as input variables. The predicted values obtained by BP network agree well with the measured values, the relative error is within 6.5% for the sample data and not bigger than 9% for the non-sample data, which indicates that the ANNs adopted can predict the flow stress of BT20 alloy effectively and can be used as constitutive relationship system applied to FEM simulation of plastic deformation.展开更多
In this paper, long interfacial waves of finite amplitude in uniform basic flows are considered with the assumption that the aspect ratio between wavelength and water depth is small. A new model is derived using the v...In this paper, long interfacial waves of finite amplitude in uniform basic flows are considered with the assumption that the aspect ratio between wavelength and water depth is small. A new model is derived using the velocities at arbitrary distances from the still water level as the velocity variables instead of the commonly used depth-averaged velocities. This significantly improves the dispersion properties and makes them applicable to a wider range of water depths. Since its derivation requires no assumption on wave amplitude, the model thus can be used to describe waves with arbitrary amplitude.展开更多
Objective To evaluate the effect of wound closure tension on the blood flow of the expanded pedicled fasciocutaneous flap,so as to find the best tension for the blood supply of the flap.Methods 8 piglets,aged 9-12 mon...Objective To evaluate the effect of wound closure tension on the blood flow of the expanded pedicled fasciocutaneous flap,so as to find the best tension for the blood supply of the flap.Methods 8 piglets,aged 9-12 months,were used.On展开更多
文摘Railway passenger flow forecasting can help to develop sensible railway schedules,make full use of railway resources,and meet the travel demand of passengers.The structure of passenger flow in railway networks and the spatiotemporal relationship of passenger flow among stations are two distinctive features of railway passenger flow.Most of the previous studies used only a single feature for prediction and lacked correlations,resulting in suboptimal performance.To address the above-mentioned problem,we proposed the railway passenger flow prediction model called Flow-Similarity Attention Graph Convolutional Network(F-SAGCN).First,we constructed the passenger flow relations graph(RG)based on the Origin-Destination(OD).Second,the Passenger Flow Fluctuation Similarity(PFFS)algorithm is used to measure the similarity of passenger flow between stations,which helps construct the spatiotemporal similarity graph(SG).Then,we determine the weights of the mutual influence of different stations at different times through an attention mechanism and extract spatiotemporal features through graph convolution on the RG and SG.Finally,we fused the spatiotemporal features and the original temporal features of stations for prediction.The comparison experiments on a railway bureau’s accurate railway passenger flow data show that the proposed F-SAGCN method improved the prediction accuracy and reduced the mean absolute percentage error(MAPE)of 46 stations to 7.93%.
基金sponsored by the Earmarked Fund for Modern China Wool & Cashmere Technology Research System (CARS-40-03)the National Natural Science Foundation for Young Scholars of China (31402057)Project support was provided by the ASTIP (Agricultural Science and Technology Innovation Program) for Genetic Resource and Breeding of Fine-Wool Sheep, Chinese Academy of Agricultural Sciences
文摘The aim of our present study was to construct genetic structure and relationships among Chinese fine-wool sheep breeds. 46 individuals from 25 breeds or strains were genotyped based on the Illumina Ovine 50K SNP array. Meanwhile, genetic variations among 482 individuals from 9 populations were genotyped with 10 microsatellites. In this study, we found high genetic polymorphisms for the microsatellites, while 7 loci in the Chinese superfine Merino strain （Xinjiang types） （CMS） and 5 loci in Gansu alpine superfine-wool sheep strain （GSS） groups were found deviated from Hardy-Weinberg equilibrium （HWE）. Genetic drift FsT=0.019 （P〈0.001） and high gene flows were detected in all the 7 fine-wool sheep populations. Phylogenetic analysis showed fine-wool sheep populations were clustered in a group independent from the Chinese indigenous breeds such that the 7 fine-wool sheep clustered distinct from Liangshan semifine-wool sheep （LS） and Hu sheep （HY） reflected by different population differentiation analyses. Overall, our findings suggested that all fine-wool sheep populations have close genetic relationship, which is consistent with their breeding progress. These populations, therefore, can be regarded as open-breeding populations with high levels of gene flows. Furthermore, the two superfine-wool strains, viz., CMS and GSS, might be formed by strong artificial selection and with frequent introduction of Australian Merino. Our results can assist in breeding of superfine-wool sheep and provide guidance for the cultivation of new fine-wool sheep breeds with different breeding objectives.
文摘The hot compression experiments were performed to investigate the effects of hot deformation parameters on the flow stress of BT20(Ti-6Al-2Zr-1Mo-1V) titanium alloy. The results show that the flow stress decreases with the increment of deformation temperature and increases with the growth of strain rate. The peak stress moves toward the direction of strain reducing and the strain rate sensitivity increases with the rising deformation temperature. There is obvious deformation heating created during hot deformation under relatively higher strain rate and lower deformation temperature. The improved back propagation(BP) neural network with 3-20-16-1 architecture has been employed to establish the prediction model of flow stress using deformation degree, deformation temperature and strain rate as input variables. The predicted values obtained by BP network agree well with the measured values, the relative error is within 6.5% for the sample data and not bigger than 9% for the non-sample data, which indicates that the ANNs adopted can predict the flow stress of BT20 alloy effectively and can be used as constitutive relationship system applied to FEM simulation of plastic deformation.
基金Supported by the Knowledge Innovation Programs of the Chinese Academy of Sciences (Nos. KZCX2-YW-201 and KZCX1-YW-12)Natural Science Fund of the Educational Department, Inner Mongolia (No.NJzy08005)the Science Fund for Young Scholars of Inner Mongolia University (No. ND0801)
文摘In this paper, long interfacial waves of finite amplitude in uniform basic flows are considered with the assumption that the aspect ratio between wavelength and water depth is small. A new model is derived using the velocities at arbitrary distances from the still water level as the velocity variables instead of the commonly used depth-averaged velocities. This significantly improves the dispersion properties and makes them applicable to a wider range of water depths. Since its derivation requires no assumption on wave amplitude, the model thus can be used to describe waves with arbitrary amplitude.
文摘Objective To evaluate the effect of wound closure tension on the blood flow of the expanded pedicled fasciocutaneous flap,so as to find the best tension for the blood supply of the flap.Methods 8 piglets,aged 9-12 months,were used.On
