Top Read Articles

Published in last 1 year |

In last 2 years |

In last 3 years |

All

Please wait a minute...


For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails

Select

Uyghur-Chinese neural machine translation method based on back translation and ensemble learning FENG Xiao, YANG Ya-ting, DONG Rui, AZMAT Anwar, MA Bo Journal of Lanzhou University of Technology    2022, 48 (5): 99-106.   Abstract （1520）      PDF（pc） （863KB）（336）       Knowledge map    Save From the perspective of efficient utilization of existing resources, a method based on back-translation and ensemble learning is proposed to solve the problem of the poor performance of Uyghur-Chinese neural machine translation caused by the lack of parallel corpus. Firstly, Uyghur and Chinese pseudo-parallel corpora are constructed by using back translation and large-scale Chinese monolingual corpora, and the intermediate model is obtained by using pseudo parallel corpora training. Secondly, the bootstrap is used to resample the original parallel corpus for N times, and N sub-datasets with similar distribution but different characteristics are obtained. The intermediate model were fine-tuned based on N sub-data sets, and N sub-models with differences were obtained. Finally, integrate these sub-models. Experiments on the test sets of CWMT2015 and CWMT2017 show that theBLEU(Bilingual Evaluation Understudy) value of this method are 2.37 and 1.63 higher than that of the baseline system, respectively. Reference | Related Articles | Metrics

Select

Nitrogen doping of tubular hard carbon as conjugate microporous polymer and its lithium storage performance ZHANG Qing-tang, BAI Yong-bao, LI Meng Journal of Lanzhou University of Technology    2020, 46 (1): 71-76.   Abstract （974）      PDF（pc） （4508KB）（747）       Knowledge map    Save Conjugate microporous polymer (CMP) and nitrogen-contained conjugate microporous polymer (NCMP) were synthesized with chosen different monomers and the tubular hard carbon(THC) and N-doped tubular hard carbon(NTHC) were prepared by means of pyrolysis of those two kinds of CMP. The morphology and structure of the two tubular hard carbons were characterized with SEM, TEM and BET and it is shown that the materials of THC and NTHC have similar morphology and specific surface area. Further, The electrochemical property of THC and NTHC then compared with that of common carbon nanotube(CNT) was analyzed by constant current charge and discharge method and the result showed that the lithium storage performance of NTHC would be the best and it would have a high reversible lithium storage capacity of 541.2 mAh/g at 0.1 A/g and its specific capacity would retain around 458 mAh/g at 0.6 A/g after 500 cycles. Reference | Related Articles | Metrics

Select

Load-structure model of side pile deformation based on pile-tunnel method YUAN Yang-kou, JIAO You-quan, LUO Mu-chi, ZHANG HaoLIU Cheng-bin, FAN Zhi-yuan Journal of Lanzhou University of Technology    2020, 46 (1): 123-128.   Abstract （904）      PDF（pc） （5523KB）（282）       Knowledge map    Save Grasping the stress and deformation regularity of side pile is very important for the safety of pile-tunnel approach-based station. Based on the modified load-structure model, the deformation regularity of the side pile is studied and compared with the field-measured result. The result shows that the lateral displacement of the side pile from the top to the bottom plate will exhibit parabolic form and its lateral displacement below the bottom plate will basically be zero, and the initial branch system of the guide tunnel will be able to withstand the stress of largeportion of its surrounding rock. However, the lateral pressure coefficient near the bottom of the guide tunnel should be enlarged by 20 percent if the surrounding rock stress of the guide tunnel was ignored in the calculation model; and the lateral displacement of the side pile calculated with conventional load-structure model will be relatively small as a whole. The lateral displacement of the side pile calculated with modified load-structure model will be in better agreement with the field measured data. The axial force on side pile calculated with modified load-structure model will be greatly reduced and the shearing force in side pile and bending moment on it will bebasically unchanged. Reference | Related Articles | Metrics

Select

Study on magnetic properties of Fe-based amorphous composites ZHAO Yan-chun, WANG Bo, LI Ze-feng, SUN Hao, KOU Sheng-zhong Journal of Lanzhou University of Technology    2021, 47 (2): 1-6.   Abstract （819）      PDF（pc） （3215KB）（533）       Knowledge map    Save The Fe-15Mn-5Si- xCr-0.2C(wt.%) amorphous composite material was prepared by using water-cooled copper crucible smelting and copper mold suction casting. XRD was used to characterize the phase composition of the sample. Vibrating Sample Magnetometercombined with M-H and B-H curves were used to analyze the effect of the change of the Cr element content on the magnetic properties for amorphous composite material. The results show that the Fe-15Mn-5Si- xCr-0.2C ( x=1, 2, 4, 6, 8, 10, 12, 13, 14) samples are composed of α-Fe phase, γ-Fe phase and the amorphous phase. As thecontent of Cr element increases, the mixing enthalpy decrease while the mixing entropy increase. The synergistic effect of α-Fe phase and the amorphous phase made the Fe-15Mn-5Si-14Cr-0.2C exhibit the best soft magnetic properties, which the coercive force and saturation magnetization are 16.12A/m and 5.1T, respectively. Reference | Related Articles | Metrics

Select

Effect of impeller exit width on performance and pressure fluctuation of centrifugal pump ZHAO Wei-guo, ZHAI Li-jing, XIAN Li-xia, MA Liang-liang Journal of Lanzhou University of Technology    2020, 46 (2): 62-68.   Abstract （814）      PDF（pc） （6499KB）（853）       Knowledge map    Save Taking a single-stage end-suction centrifugal pump as study object and assuring the inlet and outlet setting angle and diameter and as well as other parameters unchanged, five sets of pump models with different outlet width of impeller were designed respectively, their numerical simulation was carried out one after another and the influence of impeller outlet width on the performance and pressure fluctuation was analyzed. It was shown by the study result that with the increase of impeller outlet width, the lift and shaft power would increase to different extent, the efficiency curve would be hump shaped, indicating that there would be an optimum impeller outlet width to make the flow loss minimum; increasing the width of the impeller outlet, flow separation phenomenon in flow channel would be enhanced, blocking phenomenon there would decrease hydraulic loss would decrease thereby, indicating that an appropriate width of impeller outlet would be effective for reducing the energy loss of centrifugal pump. With progressive decrease of impeller outlet width, the shaft frequency peak would change obviously, assuming progressive increase trend, indicating that an excessive narrow impeller outlet width would easily lead to flow blocking in flow passage and hinder the development of the flow mode, At the same time, the impeller outlet width would also have a greater effect on pressure pulsation in centrifugal pump. Reference | Related Articles | Metrics

Select

Effect of a kind of low-level jet on aerodynamic performance of horizontal-axis wind turbine YANG Cong-xin, LUO Song, LI Shou-tu, ZHANG Xu-yao Journal of Lanzhou University of Technology    2020, 46 (1): 53-59.   Abstract （792）      PDF（pc） （4922KB）（184）       Knowledge map    Save An 1.5 MW variable-pitch variable-speed three-blade horizontal-axiswind turbine was taken as research object. The similarities and differences of its aerodynamic performance were investigated under the three kinds of inflows and the effect of a typical low-level jet on its aerodynamic performance was studied emphatically. The results showed that the power and thrust of the rotor under low-level jet are larger than those under the uniform flow and the flow with wind shear. Surface pressure of the blade under low-level jetis generally larger than those of the other two inflows and there is a more noticeable flash-down zone of pressureon the section side of airfoil in the middle section due to jet attenuation.Under the condition of low-level jet, the normal and tangential force coefficients of the blade are generally larger than ones underthe other two inflows. The attack angles of each airfoil section under low-level jet are larger than ones unders shear flow by 0.05%~36.66%, the stall region on the blade surface are larger than those under other inflows, especially near blade root, the separation line moves close to the leading edge, and the separation vortexes on airfoil sections due to jet attenuation are more obvious. The study of effects of different kinds of low-level jets on aerodynamic performance of horizontal-axis wind turbine will be done in the future. Reference | Related Articles | Metrics

Select

Preparation and adsorption properties of corncob based biochar WEI Hui-ge, LI Dai-yuan, MA Jun-hui, ZHOU Deng-feng, CUI Da-peng Journal of Lanzhou University of Technology    2020, 46 (6): 77-84.   Abstract （760）      PDF（pc） （2593KB）（673）       Knowledge map    Save Using agricultural waste corncob as raw material, biochar with low cost and high performance was prepared by hydrothermal carbonization (HTC) at different temperatures. The biochar has mesoporous structure and abundant oxygen-containing functional groups on its surface, such as —OH, C ==O, C—O and so on. Methylene blue (MB) was used as model adsorbent to further study the adsorption properties of the biochar. Results from the study of adsorption kinetics show that the adsorption behavior is in accordance with the pseudo second-order kinetics model, and the biochar obtained under the hydrothermal condition of 225 ℃ has the maximum adsorption capacity (41.37 mg/g) and the maximum adsorption rate. The adsorption equilibrium data agree well with the Langmuir isotherm model, indicating that the adsorption of MB on biochar belongs to monolayer adsorption, and the interaction between oxygen-containing functional groups on the surface of the biochar and the MB molecule is helpful to the adsorption process. Reference | Related Articles | Metrics

Select

Dynamic behavior of stochastic predator-prey system with Lévy jumps HUANG Can-yun, LIANG Hong-yang, MENG Xin-you Journal of Lanzhou University of Technology    2020, 46 (1): 152-157.   Abstract （695）      PDF（pc） （2636KB）（193）       Knowledge map    Save Dynamic behavior of stochastic predator-prey model with Lévy jumps is investigated. By constructing an appropriate Lyapunov function and using It formula, the existence uniqueness of the global positive solution of the system is proven and the sufficient conditions that guarantee the stochastic permanence or extinction of the system are given out. Finally, numerical simulation is conducted to verify the validity of the theoretic result obtained. Reference | Related Articles | Metrics

Select

Analysis of thrust and velocity variation of permanent magnet synchronous linear motor under variable load SHEN Hao, ZHANG Wang, TAN Wei-rong, DANG Sheng-hui Journal of Lanzhou University of Technology    2021, 47 (2): 54-59.   Abstract （681）      PDF（pc） （1152KB）（1210）       Knowledge map    Save Taking the permanent magnetic synchronous linear motor as the research object, its force was analyzed; mathematical model was established, and the simulation model of the linear motor by using Matlab/Simulink in order to study the thrust and speed change of the linear motor when different loads are applied. The results show that with the increase of the workpiece’s mass, the state response time and maximum deviation of the linear motor increased; the response time curve of the speed is flat; the response time curve of thrust changed obviously; the overshoot curve of thrust increase flatly; the overshoot curve of velocity increased stably, the overshoot curve of speed is obvious.When the quality of the workpiece is 240 kg, the over-adjusting amount of motor speed exceeds 30%, which seriously affects the operation of the motor. In order to reduce the effect of load change on linear motor stability, the maximum load quality and maximum motor thrust are calculated. Reference | Related Articles | Metrics

Select

Anti-sway design of overhead crane based on model predictive control TANG Wei-qiang, HUANG Xiao-li, LONG Wen-kun, SUN Li-juan Journal of Lanzhou University of Technology    2020, 46 (2): 92-96.   Abstract （606）      PDF（pc） （1787KB）（729）       Knowledge map    Save Aimed at the sway problem of lifted weight in the working process of overhead crane, an electronic anti-swing design method is proposed based on predictive control strategy. Firstly, dynamic analysis is carried out for the overhead crane, and a control-oriented mathematical model is established by using Lagrange equation, Secondly, according to the historical information of the object and the future input, the future output is predicted and a control rule is obtained by optimizing the performance function combined with prediction error compensation. Finally, it is shown by simulation result that the designed anti-sway system can be used to realize rapid positioning of the trolley and effective restriction of the swing angle of lifted weight. Compared with an optimal anti-sway control, the system overshoot and swing angle amplitude will be greatly reduced, manifesting the superiority of this anti-sway method; When a parameter perturbation or measurement noise is added onto this system, the anti-sway system can still achieve the control purpose very well, showing that this anti-swing method has a better robustness. Reference | Related Articles | Metrics

Select

Simulation of dendrite free growth with phase-field model based on adaptive finite element method ZHU Chang-sheng, XU Sheng, LEI Peng, HAN Dan Journal of Lanzhou University of Technology    2020, 46 (1): 24-31.   Abstract （592）      PDF（pc） （8402KB）（125）       Knowledge map    Save Phase-field model is used to simulate the growth and morphological evolution of single intact equiaxed dendrite during its solidification in supercooled melt, the governing equations of the phase-field model are solved with adaptive finite element method, and the influence of physical parameters such as anisotropy and supercooling degree on the dendrite shape and growth under condition of larger computational domain and less interfacial layer thickness. The result shows that a competition will occur more intensively among the dendrites and their growth speed will be less when the supercooling degree is greater. However, the greater the anisotropy coefficient is, the stronger the dendrite growth tendency along a chosen direction will be, the faster the growth speed will be, the less the secondary dendrite spacing will be, and the more developed the lateral branch will be. Compared with the finite difference method (FDM), the adaptive finite element method (AFM) will reduce CPU time and storage space by one order of magnitude, and the larger the system size is, the more obvious the superiority of adaptive finite element method will be, which will facilitate the numeric simulation of greater dimensional and multi-field coupled phase field. By means of comparing to FDM and previous numerical simulation of crystal growth, the AFM is proved to be accurate, efficient and robust. Reference | Related Articles | Metrics

Select

Preparation of mesoporous rGO/TiO 2hollow blocks with wide spectral response and photocatalytic performance related with them LI Cui-xia, SHI Xiao, SUN Hui-zhen, WANG Hui-kang, KANG Wei-chao Journal of Lanzhou University of Technology    2020, 46 (1): 32-36.   Abstract （568）      PDF（pc） （3545KB）（181）       Knowledge map    Save Mesoporous graphene oxide/TiO 2 hollow bulk materials were synthesized in-situ with Sol-Gel method, by taking tetrabutyl titanate (TBT) and graphene oxide(GO) as raw materials, yeast as cavity template agent and citric acid as hydrolysis inhibitor and mesoporous template agent, and then the reduced graphene oxide/TiO 2 (rGO/TiO 2) bulk materials were obtained by means of ultraviolet radiation reduction. The samples were analyzed by means of various characterization means such as X-ray diffraction (XRD), scanning electron microscopy (SEM), Brunauer-Emmett-Teller (BET) and photoluminescence (PL) and the effect of yeast addition on degradation of tetracycline hydrochloride with rGO/TiO 2 was studied. It was shown by experimental result that rGO/TiO 2 would be mesoporous bulk materials with specific surface area of 50.66 m 2/g and mesopores size of 4.3 nm in majority. The addition of yeast could effectively inhibit the recombination of the photo-generated electron-hole pair of catalyst, broaden the spectral response range and improve the photocatalytic performance. When the mass fraction of yeast was 5%, the catalytic performance of rGO/TiO 2 would reach optimal level. Under visible light and ultraviolet light, the degradation rate of tetracycline hydrochloride would reach 84.97% and 96.96%, respectively. Reference | Related Articles | Metrics

Select

General model for kinematic analysis of swash plate axial piston pump WANG Jian-sen, CAO Wei-dong, HE Xin-long, WANG Zheng-rong, ZHANG Zhi-wei Journal of Lanzhou University of Technology    2021, 47 (5): 59-64.   Abstract （546）      PDF（pc） （3701KB）（775）       Knowledge map    Save Based on the comprehensive consideration of swash plate angle, cross angle and piston inclination angle of the swash plate axial piston pump, the general piston kinematics model is established by coordinate transformation method. The analysis results show that the cross angle of swash plate and the inclination angle of piston will increase the maximum amplitudes of displacement, velocity and acceleration of piston, and then increase the displacement of pump, the maximum transient flow velocity in the piston chamber and the inertia force of piston; the cross angle of swash plate will change the phase of piston movement and make the internal and external dead points deflect, which will change the pre-compression and pre-expansion angle of valve plate; the inclination angle of piston will affect the movement of the piston, the smaller the inclination angle of the plunger, the closer the movement of the plunger is to the simple harmonic motion, on the contrary, the larger the deviation is, the maximum deflection angle of internal and external dead center is 90°, the smaller the absolute value of the cross angle is, the smaller the deflection angle of internal and external dead center changes in the range of big swash plate angle, and the greater the gradient of change when swash plate angle is close to zero displacement. Reference | Related Articles | Metrics

Select

Study on turbulence effects on the wake and torque characteristics of horizontal axis wind turbine LI De-shun, SU Jin-cheng, LI Yin-ran, GUO Tao Journal of Lanzhou University of Technology    2021, 47 (1): 47-53.   Abstract （540）      PDF（pc） （5089KB）（397）       Knowledge map    Save Based on a horizontal axis wind turbine of 33 kW, the wake flow field of thewind turbine under condition of uniform flow or turbulent flow of different intensities is simulated by using the method of large eddy current to simulate coupled actuator line, and the effect of turbulence on the wake and torque characteristics of horizontal axis wind turbine is also investigated. Simulated results indicate that: the energy exchange between the wake and the surrounding flow field of the wind turbine in the turbulent flow is stronger than that in the uniform flow. The turbulent velocity is faster therefore, so that velocity recovery in the wake area becomes faster as a result. The greater the turbulence is, the faster the velocity recovery may be. Compared with the uniform flow, the maximum loss rate of the wake velocity of the wind turbine decreases during turbulent flowing, but the degree of turbulence has less effect on the maximum loss rate of the wake velocity. The expanding rate of wake radius decreases gradually in the range from the plane of the wind turbine to twice the diameter of the wind turbine, and remains basically unchanged after twice the diameter. Turbulence will increase expansion of the wake, and the greater the turbulence is, the stronger the expansion will be. Under the condition of 11 m/s wind speed, the increase of turbulence leads to not only the decrease of mean value of the wind turbine torque, and the increase of fluctuation amplitude, but also the increase of energy consumption of the wind turbine torque in high frequency and low frequency ranges. The increase of energy consumption in the range of low frequency is faster. Reference | Related Articles | Metrics

Select

The effect of hot isostatic pressing temperature on the microstructure and tensile properties of TC4 alloy CHENG Jian-hong, XIAO Le, CHEN Yong-qing, YAN Ying-jie, CAO Rui Journal of Lanzhou University of Technology    2021, 47 (1): 1-4.   Abstract （532）      PDF（pc） （9525KB）（348）       Knowledge map    Save The preparation of TC4 alloy by powder NNS-HIP has received widespread attentions in recent years due to its unique advantages. In this research, effects of hot isostatic pressing temperature on the microstructure and tensile properties of TC4 alloy were systematically investigated. The experimental results show that the alloy is completely densified when the holding temperature reaches 800 ℃ under the holding pressure condition of 120 MPa for 2 hours; As the holding temperature increases from 800 ℃ to 920 ℃, the tensile strength of the alloy decreases from 942 MPa to 920 MPa, and the tensile strain increased from 13.7% to 18.1%; The volume fraction of β phase in the alloy microstructure increases with the rise of the holding temperature, and the distribution of β phase tends to be uniform as a result. The addition in β phase is a key factor for the decline in strength and ascent in plasticity of the alloy. Reference | Related Articles | Metrics

Select

A secure face recognition scheme based on CKKS homomorphic encryption and neural network SU Yun-xuan, TU Zheng, WANG Xu-an, LIN Chuan Journal of Lanzhou University of Technology    2023, 49 (2): 103-109.   Abstract （510）      PDF（pc） （2072KB）（393）       Knowledge map    Save As big data technology has been evolving over the years, facial recognition is widely used in different fields. However, users are then troubled by safety problems, such as privacy disclosure and data leakage. A face recognition scheme based on homomorphic encryption and neural network under the cloud server is proposed in this paper. The face image is encrypted through the CKKS homomorphic encryption scheme, and the LFW data set is trained through the ResNet50 model and Arcface Loss function in the cloud server, then the vector correlation between the encrypted images is calculated, followed by the decryption of the ciphertext result and comparison with the threshold to judge whether it is the same person, to realize face recognition. The experiment result shows that on the LFW dataset, the scheme of this paper can recognize the cryptograph with an accuracy rate of 99.398% when the threshold value is about 0.25. This result proves the proposed scheme has high recognition accuracy. Reference | Related Articles | Metrics

Select

Realization of face feature point recognition based on cascaded convolutional neural network ZHANG Yun, LI Lan Journal of Lanzhou University of Technology    2020, 46 (3): 105-109.   Abstract （500）      PDF（pc） （862KB）（1754）       Knowledge map    Save If a face image is interfered by redundant information, the accuracy of the extracted effective feature points from the image is not high enough. To solve this problem, a face feature point detection algorithm based on a cascade convolutional neural network is proposed in this paper. The algorithm reads in information of an original image via the input-function first, then extracts local features of the image through neurons in a receptive domain, and inputs all local features into a pooling domain. After doing that, the algorithm is able to average all captured local features stored in the pooling domain, and further do down-sampling to the pooling domain, and reduce the dimension of the convolution results, and finally output detection results of those feature points by virtual of iterative training. In this study, Python language is used to program the algorithm, and simulation experiments are carried out with the help of face data set. Simulation results show that the algorithm has a high recognition rate for face feature points. Reference | Related Articles | Metrics

Select

Improved calculation formula of fundamental frequency for the fabricated concrete simple supported skew girder bridges YU Zhen-gang, LI Yan, YANG Ting-ting, LU Kai-yuan Journal of Lanzhou University of Technology    2020, 46 (5): 144-148.   Abstract （499）      PDF（pc） （1526KB）（438）       Knowledge map    Save Taking a prestressed concrete prefabricated simply supported girder bridge asengineering background, the influence of oblique angle, span, bridge width and transverse torsional stiffness on the fundamental frequency of the bridge is analyzed by numerical method. The results show that as the skew angle becomes larger, the first three orders of the vertical natural vibration frequencies increase. With the increase of the span, the whole vertical stiffness becomes lower, the fundamental frequency reduce obviously, but the rate to deduction decreased gradually. The fundamental frequency affected by the mechanical characteristic of coupling of bending and torsion increases with the increase of torsional stiffness. Based on the influence of key factors above on the base frequency of the skew beam bridge, an improved formula, taking the torsional stiffness and the skew angle into account, for estimating the base frequency of the concrete simply supported skew beam bridge is therefore proposed. The precision and the applicability of the formula are verified by all measured data of several real bridges. Reference | Related Articles | Metrics

Select

A small sample reliability assessment method based on Bootstrap ZHANG Zhen, LIU Jian-hui, ZHAO Cheng, YAN Chang-feng Journal of Lanzhou University of Technology    2022, 48 (1): 39-44.   Abstract （471）      PDF（pc） （2389KB）（442）       Knowledge map    Save In order to solve the problem that the maximum likelihood estimation method (Mle) may produce large errors in solving the distributed parameters in the case of small data samples, the B-MLE method is proposed based on Bootstrap data expansion. Firstly, the Bootstrap method was used to resample the small sample data to generate multiple groups of regenerated samples, so as to expand the data sample. Secondly, the maximum likelihood estimation is used to solve the distribution parameters of the regenerated samples, and the maximum likelihood estimation of multiple parameters is obtained. The probability density function is obtained directly from the parameter estimation by using the kernel density estimation method. Finally, at a given confidence level, the confidence interval of parameters is determined to obtain the confidence interval of reliability. The feasibility and credibility of the proposed method are verified by Monte Carlo method. The results show that the proposed method can reduce the error of maximum likelihood estimation. Reference | Related Articles | Metrics

Select

Cleaning and modeling of abnormal data of wind farm power curve CAO Li-xin, LIU Wei-min, GUO Hu-quan Journal of Lanzhou University of Technology    2022, 48 (4): 64-70.   Abstract （469）      PDF（pc） （3178KB）（764）       Knowledge map    Save The wind speed and power data of wind farms usually contain a large number of abnormal data, which is difficult to reflect the real working conditions of wind turbines, affect the accuracy of wind power prediction, and then cause certain economic losses. To solve this problem, the characteristics of abnormal data are analyzed, a slip-quartile abnormal data elimination method is proposed, and the wind speed power curve of the eliminated data is modeled by using high-order polynomial and Logistic function. Finally, the applicability and effectiveness of this method are verified by variance, root means square error and determination coefficient. The analysis in the example shows that the method is simple, efficient and versatile, and can significantly improve the accuracy of wind turbine power characteristic analysis. Reference | Related Articles | Metrics