Loading...

Archive

    For Selected: Toggle Thumbnails
    Information & Control
    Sensor Fault Estimation of Uncertain Vehicle Electronic Stability Control System
    WANG Hong-wei, ZHANG Hao-tian, HAN Jie, LIU Chen-yu
    2023, 44 (1):  1-8.  DOI: 10.12068/j.issn.1005-3026.2023.01.001
    Abstract ( 772 )   HTML ( 32)   PDF (835KB) ( 406 )  
    An observer-based fault estimation strategy is proposed to solve the fault problem of yaw rate sensor. Firstly, considering the uncertainty caused by tire cornering stiffness characteristics and nonlinear disturbance in the system, the mathematical model of vehicle four-wheel steering system is established. Then, an observer-based sensor fault estimation strategy is designed, and the observer gain is solved by Lyapunov function and linear matrix inequality method. Finally, the effectiveness of the sensor fault estimation strategy is verified by Carsim and Simulink simulation. The results show that the proposed strategy can estimate the system state and sensor fault well when the sensor fault occurs. It can provide key information for vehicle safety and stability system, and provide technical support for the next step of sensor fault-tolerant control.
    References | Related Articles | Metrics
    An Superheat Identification Method in Aluminium Electrolysis Based on Residual Convolutional Self-Attention Neural Network
    LIN Qing-yang, CHEN Xiao-fang, XIE Yong-fang
    2023, 44 (1):  8-17.  DOI: 10.12068/j.issn.1005-3026.2023.01.002
    Abstract ( 666 )   HTML ( 10)   PDF (909KB) ( 361 )  
    Superheat is an important indicator to reflect the current production efficiency of aluminium electrolytic cells. Due to the difficulty of superheat online real-time measurement, this paper proposes a superheat identification method based on residual convolution self-attention neural network (RCSANN). As the production data in aluminium electrolysis process is time series data and featured with multi-source heterogeneous characteristics, the isomorphic representation method is designed for heterogeneous data. On this basis, the RCSANN superheat model is proposed to extract the global and local features of the isomorphic time series data. Aiming at the problem of few labels and uneven category distribution of superheat data, the unsupervised pre-training method based on auto-encoder and the weighted cross-entropy loss function are used to improve the performance of the superheat identification task. The validity of the proposed method is verified by simulation and comparison experiments on the benchmark dataset. Then, experiments are carried out on the dataset of superheat in aluminium electrolysis with only a few unbalanced labels. The results show that not only the accuracy of superheat identification is improved compared with other existing models, but also the generalization ability can be guaranteed under few training labeled-samples.
    References | Related Articles | Metrics
    Early Wear Diagnosis of Gears Based on Spectrum Correlation Analysis
    WANG Hong-min, CHAN Liang
    2023, 44 (1):  18-25.  DOI: 10.12068/j.issn.1005-3026.2023.01.003
    Abstract ( 480 )   HTML ( 8)   PDF (1791KB) ( 204 )  
    Gear fault detection based on vibration analysis has been proven to be effective in fault identification. However, the extraction and identification of vibration signals characterizing early wear have not been solved well. This paper proposes a method for early wear diagnosis of gears, combining variational mode decomposition (VMD) based on spectrum correlation analysis with kernel support vector machine (SVM). For weak gear vibration signals that can characterize early wear, the modal numbers are initialized by an approximate complete reconstruction criterion. Meanwhile, the frequency corresponding to the maximum value of the signal power spectral density is used to initialize the center frequency of the VMD method. It is used to effectively extract gear wear information from gear vibration signals and then be combined with kernel support vector machines for early wear diagnosis of gears. The experimental results show that the proposed method can effectively overcome the problem that the modal numbers cannot be preset with large background noise, be with better robustness to noisy situations, and achieve a diagnostic accuracy of 94.4%, which provides a solution for early wear detection of gears.
    References | Related Articles | Metrics
    Improved Two-Branch Person Re-identification Algorithm Based on Transformer
    LIU Yang, YAN Dong-mei, MENG Fan-wei
    2023, 44 (1):  26-32.  DOI: 10.12068/j.issn.1005-3026.2023.01.004
    Abstract ( 633 )   HTML ( 14)   PDF (1323KB) ( 424 )  
    In order to solve the problem of insufficient global information modeling of person re-recognition algorithm based on convolutional neural network, the limitation of convolution operation is analyzed, and an improved global-local two-branch person re-recognition algorithm based on Transformer is proposed. Firstly, the multi-headed self-attention mechanism which is embedded in the Resnet50 backbone network is optimized by relative position-coding. After that, the processed image is split into two parts on the global branch geometrically, and the ability of extracting the abstract features is enhanced by the Transformer’s global receptive field. On the local branch, the Layer_3 output is under the supervision of dimensionality reduction while the multi-scale pooling obtains richer local features. The experimental result shows that, on the Market-1501 and DukeMTMC-reID datasets, mAP/Rank-1 of the algorithm reaches 93.45%/95.61% and 88.79%/90.35%, respectively. Compared with the algorithm which is only based on convolutional neural network, higher accuracy is achieved.
    References | Related Articles | Metrics
    Named Entity Recognition in Threat Intelligence Domain Based on Deep Learning
    WANG Ying, WANG Ze-hao, LI Hong, HUANG Wen-jun
    2023, 44 (1):  33-39.  DOI: 10.12068/j.issn.1005-3026.2023.01.005
    Abstract ( 817 )   HTML ( 24)   PDF (1821KB) ( 548 )  
    In order to extract key information of threat intelligence from different sources and facilitate the government regulatory authorities to carry out security risk assessment, to reduce the difficulty identification caused by the serious mixing of Chinese and English threat intelligence texts and the lack of professional vocabulary, based on BiGRU-CRF model, a threat intelligence named entity recognition(NER)method integrating boundary features and iterated dilated convolution neural network (IDCNN) is proposed. Firstly, entities with clear boundaries, such as English words, are transformed according to the artificially constructed rule dictionary to reduce the loss of information easily caused by the model when processing long texts. The local feature information and the context global feature information are obtained through IDCNN and bidirectional gated recurrent unit (BiGRU), respectively. The results of experiments on threat intelligence corpus show that the proposed model is better than other models in relevant evaluation indexes, and the F-score reaches 87.4%.
    References | Related Articles | Metrics
    Materials & Metallurgy
    Effect of Zr Content on Microstructure and Performance of Al-0.36Mg-1.23Si Alloy Sheet for Automobile Body
    TIAN Ni , LI Yan-jie , WANG Wen-ze , FENG Zhen
    2023, 44 (1):  40-48.  DOI: 10.12068/j.issn.1005-3026.2023.01.006
    Abstract ( 587 )   HTML ( 2)   PDF (2794KB) ( 107 )  
    The effect of Zr content (≤ 0.29%) on the microstructure, formability and baking hardenability of Al-0.36Mg-1.23Si aluminum alloy sheet after T4P tempering was investigated through the optical microscope (OM) and transmission electron microscope (TEM), tensile and cupping tests. The results show that when the mass fraction of Zr was less than 0.15%, the element in the alloy mainly existed in the form of 40~50nm (AlSi)3(ZrxTi1-○x○) dispersoid second-phase particles, and the amount of the particles increased with the Zr addition. The 3~23 μm (AlSi)3(ZrxTi1-x) primary particles were observed when the Zr content was over 0.22%. The recrystallization grains were refined from 55 μm to 20 μm with the Zr fraction increasing from 0% to 0.15%, and there was no significant change of the grain size after a further increase of Zr. With the Zr increase, the strength of the alloy increased, the elongation firstly remained unchanged and then decreased, and the n10~20, r10 and IE values remained unchanged. The increments of yield strength after the Zr addition were in excess of 86 MPa after the simulated paint baking treatment and the increase of Zr content had no significant effects on their baking hardenability.
    References | Related Articles | Metrics
    Microstructure Evolution of High Carbon Galvanized Steel Wire and Effect on Torsional Property
    ZHI Wei-jun , YAO Zan, JIANG Zhou-hua
    2023, 44 (1):  49-54.  DOI: 10.12068/j.issn.1005-3026.2023.01.007
    Abstract ( 579 )   HTML ( 6)   PDF (2997KB) ( 183 )  
    Focusing on the fluctuation problem of torsion property of steel wire after galvanizing in practice, microstructure evolution of the wire in the drawing, hot-galvanizing and torsion processes was systematically characterized and analyzed in this work. The results show that in the processes of drawing and torsion of the wire, the sorbite sheet is distorted, the lamellar spacing is reduced, the cementite sheet is broken, the small cementite particles are dissolved, and dislocation density greatly increases. The hot-galvanizing process can dissolve the small-sized cementite particles, eliminate the internal stress and reduce the dislocation density in the wire. The comprehensive effects of cementite sheet breakage, cementite particle dissolution, sorbite lamellar spacing and dislocation density lead to the performance difference in the process of steel wire torsion.
    References | Related Articles | Metrics
    Effect of Low Vacuum on Electroslag Remelting Process and Ingot Quality
    HUANG Xue-chi, LI Bao-kuan, LIU Cheng-jun, LIU Zhong-qiu
    2023, 44 (1):  55-62.  DOI: 10.12068/j.issn.1005-3026.2023.01.008
    Abstract ( 561 )   HTML ( 5)   PDF (2071KB) ( 282 )  
    Deoxidation and inclusions control are the key to improve the quality of ingots produced by electroslag remelting (ESR). The effect of low vacuum on the ESR process and ingot quality was experimentally studied in combination with thermodynamic calculations. H13 electrodes were remelted under argon protection at atmospheric and low vacuum (10kPa) conditions, respectively. The results show that, compared with atmospheric pressure, a higher gas content in the slag pool, more drastic fluctuations of the electrode melting rate, and a rougher surface of the ESR ingot were observed under low vacuum. The total wO in the ESR ingots remelted under atmospheric and low vacuum conditions are 24×10-6 and 18×10-6, respectively. The carbon deoxidation reaction is greatly enhanced under low vacuum and gradually replaces aluminum deoxidation as the pressure decreases. Compared with Al-O equilibrium, the dissolved wO in molten steel equilibrated with wC of the electrode at 10kPa is much lower. Thus, lowering pressure can further reduce the wO of the ingot. Carbon deoxidation, of which the product is CO gas, will not introduce new inclusions. The pressure has little effect on the type of inclusions in ESR ingots. The biggest inclusions in the ESR ingots remelted under atmospheric and low vacuum conditions decrease from 14.9μm to 10.5μm and 8.3μm, respectively. The ratio of the detected inclusions with equivalent diameters less than 3μm are 63.85% and 75.91%, respectively. ESR under low vacuum condition can further refine the inclusions and improve the cleanliness of ESR ingots.
    References | Related Articles | Metrics
    Inversion of Optical Parameters of Semitransparent Media Based on Quantum Particle Swarm Optimization Algorithm
    WEI Lin-yang, GUO Xin, WANG Cun-hai, LI Guo-jun
    2023, 44 (1):  63-69.  DOI: 10.12068/j.issn.1005-3026.2023.01.009
    Abstract ( 502 )   HTML ( 7)   PDF (968KB) ( 118 )  
    Aiming at the problem of optical parameter estimation for semitransparent media, an inversion model for optical parameters of semitransparent media exposed to pulse laser irradiation is established. The quantum particle swarm optimization (QPSO) algorithm is adopted to estimate the refractive index and absorption coefficient. The effects of measurement error and thermophysical parameters on the inversion results are analyzed. The relationship between the inversion accuracy and measurement error is revealed by sensitivity analysis. The calculation results show that the established inversion model and the QPSO algorithm can accurately estimate the refractive index and absorption coefficient. Even with measurement error of 10%, the retrieval results still have strong robustness and high accuracy. This study can provide technical reference for obtaining the physical parameters of semitransparent media.
    References | Related Articles | Metrics
    Hydraulic Modelling and Scheduling Scheme of Blast Furnace Gas Pipeline Network
    FANG Xiao-qing, LIU Shu-han, SUN Wen-qiang
    2023, 44 (1):  69-75.  DOI: 10.12068/j.issn.1005-3026.2023.01.010
    Abstract ( 513 )   HTML ( 6)   PDF (913KB) ( 206 )  
    To develop a set of accurate scheduling scheme of blast furnace gas (BFG) for the whole plant, the structure and operation characteristics of the BFG pipeline network of an integrated steelworks are analyzed. Considering the correlation between the pressure and flow, a hydraulic model of BFG pipeline network is established to make scheduling strategy of BFG pipeline network. The results of case study show that the pressure of the pipeline network increases with the decrease of the flow rate in the 1700 hot rolling mill. When the pressure rises to the upper limit of 20 kPa, it is necessary to open the 1# and 2# flare towers and adjust the volume of discarding BFG. The pipeline network pressure can be maintained within the allowable range if the discarding reaches 84400m3/h. When 4# and 6# blast furnaces take blowing-down, the pipeline network pressure drops sharply. Therefore, it is necessary to shut down the auxiliary boilers and 2# 80 MW gas generator unit to restore the pipeline network pressure. The pipeline network pressure is stable at 13 kPa by setting a gasholder. When the pressure relative error is less than 0.01% and the flow relative error is less than 0.8%, the calculation time of the model for each working condition is shorter, ranging from 3.59 to 4.46 s.
    References | Related Articles | Metrics
    Mechanical Engineering
    Parallel Adaptive Sampling Strategy for Structural Reliability Analysis
    ZHA Cong-yi, SUN Zhi-li, PAN Chen-rong, WANG Jian
    2023, 44 (1):  76-82.  DOI: 10.12068/j.issn.1005-3026.2023.01.011
    Abstract ( 516 )   HTML ( 16)   PDF (855KB) ( 214 )  
    Many existing adaptive sampling strategies are limited to the Kriging models, or are with low efficiency for only selecting one best sample point at each iteration. To solve the above issues, a general parallel adaptive sampling strategy CF-K is proposed. The proposed method considers the local uncertainty of the sample points and ensures that the selected sample points distribute around the limit-state function. Furthermore, the k-means algorithm is incorporated to achieve parallel computation, which means that several sample points can be simulated simultaneously at each iteration on several computers. The numerical cases show that the proposed method has less iterations and saves more time than other methods under the condition of satisfying the precision. The structural reliability analysis based on the proposed method not only achieves a good balance between computational efficiency and accuracy, but also be available for any existing surrogate models in principle.
    References | Related Articles | Metrics
    Wear Behavior of Polyurethane Polishing Pads Used in BK7 Ultrasonic Vibration-Assisted Polishing
    LIANG Ying-dong, NIU Jun-kai, ZHANG Chao, YU Tian-biao
    2023, 44 (1):  82-88.  DOI: 10.12068/j.issn.1005-3026.2023.01.012
    Abstract ( 522 )   HTML ( 5)   PDF (1603KB) ( 264 )  
    Polyurethane polishing pads are often used for polishing hard and brittle materials such as optical glass. The morphology and the wear of polishing pads directly influence the polishing accuracy and efficiency. Experimental studies were conducted on the micro-morphology and wear behavior of polyurethane polishing pads during different polishing time and their effects on material removal rate and surface roughness were analyzed. The results showed that the material removal rate and surface roughness are 0.977μm/min and 153.67nm respectively when the spindle speed is 8000r/min, the feed speed is 0.0150mm/s and the axial ultrasonic amplitude is 5μm. Polyurethane polishing pads have relatively little wear in the first 30 min, the surface of the polishing pads is filled with abrasive particles and optical glass debris, as the polishing time increases, damaging the porous structure of polishing pads. The surface of the polishing pads gradually hardens and loses its elasticity. Meanwhile, the abrasive particles and the debris invading the surface of the polishing pads will prevent the renewal of the abrasive particles in the contact area during the polishing, resulting in a reduction in the polishing quality.
    References | Related Articles | Metrics
    Grinding Control Method of Robotic Active Compliance Constant-Force
    GUO Wan-jin, YU Su-yang, ZHAO Wu-duan, CHEN Jie
    2023, 44 (1):  89-100.  DOI: 10.12068/j.issn.1005-3026.2023.01.013
    Abstract ( 1008 )   HTML ( 16)   PDF (2605KB) ( 696 )  
    In order to solve the problem of disturbance in grinding control system, a robotic force control end-effector is designed, and a grinding control method of robotic active compliance constant-force is proposed, based on the active disturbance rejection control and the fuzzy variable impedance control. A fuzzy variable impedance controller is adopted by the inner-loop control of the proposed method, and an active disturbance rejection grinding controller is adopted by the outer-loop control. Convergence of the tracking error to zero for the proposed method is guaranteed by the Lyapunov stability theory. The effectiveness and applicability of the proposed method are verified by the co-simulation and experiment. By comparing the proposed method and the PID controller, which is used to replace the outer-loop control, when the inner-loop control is same, the former has better ability to reduce force tracking error and position overshoot during simulation experiments. Meanwhile, using the proposed method, the control effects and the robustness of the force control system for the robotic grinding are improved, and the robotic compliant constant-force control is realized.
    References | Related Articles | Metrics
    Resources & Civil Engineering
    Rational Layout of Roadway for Downward Cross-Pillar Mining in Close Distance Coal Seams
    WANG Hong-bo, ZHANG Yong, PANG Yi-hui
    2023, 44 (1):  100-109.  DOI: 10.12068/j.issn.1005-3026.2023.01.014
    Abstract ( 525 )   HTML ( 1)   PDF (2623KB) ( 168 )  
    Meiyukou coal mine was taken as engineering case to explore the influence of remnant coal pillar on the layout of underlying roadway. The maximum principal stress concentration coefficient was used to identify the concentrated stress zone in the floor. The evolution characteristics of concentrated stress zone in the floor at different mining stages were studied by means of theoretical analysis, programming calculation and numerical simulation. The results show that the width and the buried depth of pillar, the distance from edge of pillar to peak zone and the peak value of vertical stress control the development of concentrated stress zone in the floor. The influence of stress recovery characteristics of goaf on the concentrated stress zone can be ignored. Comparing the peak value of the influence range of concentrated stress zone on the lower coal seam and the staggered distance between coal pillar and roadway in each simulation scheme, and combining with the gradual recovery of stress in the goaf and the regionalization characteristics of ground pressure in longwall face, it is considered that the reasonable staggered distance range is 10~15m. Choosing staggered distance of 12m for industrial test has achieved good application effect.
    References | Related Articles | Metrics
    Mechanism Study of Enhanced Hematite Reverse Flotation Desilication by Flocculation of Micro Fine Kaolinite with Polyoxyethylene Oxide
    WANG Li-ying, HAO Xiao-yu, YAO Jin, YANG Bin
    2023, 44 (1):  110-116.  DOI: 10.12068/j.issn.1005-3026.2023.01.015
    Abstract ( 570 )   HTML ( 3)   PDF (1638KB) ( 195 )  
    Mineral flotation test, combined with laser particle size test, scanning electron microscope (SEM), mineral surface potential and XPS (X-ray photoelectron spectroscopy) were used to study the behavior and mechanism of polyoxyethylene oxide (PEO) flocculation of fine kaolinite in the process of hematite reverse flotation. The flotation test showed that the addition of PEO improves the mineral recovery of kaolinite and the separation efficiency of artificial mixed ore, which is conducive to promoting the reverse flotation separation of hematite and kaolinite. The results of laser particle size test and SEM showed that PEO does not flocculate hematite, but flocculates kaolinite particles to increase their apparent particle size. Zeta-potential tests and XPS analysis indicated that chemical adsorption of PEO occurs on the surface of kaolinite particles, and its Zeta-potential is positively shifted. Therefore, PEO has certain guiding significance in the study of reverse flotation separation of hematite and kaolinite.
    References | Related Articles | Metrics
    Study on Mechanism of Weak Interaction Between Biochar and Phenol Based on Quantum Chemical Calculation
    LI Hai-bo, ZHANG Wen-xin, LI Ying-hua, LIU Wen-gang
    2023, 44 (1):  117-123.  DOI: 10.12068/j.issn.1005-3026.2023.01.016
    Abstract ( 744 )   HTML ( 7)   PDF (1642KB) ( 218 )  
    The molecular behavior of aromatic compounds on the biochar surface, especially the mechanism of weak interaction, is not yet clear. In order to reveal the mechanism of weak interaction between biochar and phenol, adsorption isotherm and adsorption kinetics experiments, as well as FT-IR characterization were used to analyze the binding energy, reduced density gradient (RDG) and electron density topological relationship between biochar and phenol of different adsorption configurations with density functional theory (DFT). Results showed that the adsorption isotherm reaches the equilibrium after 72h, pH significantly affects the saturated adsorption capacity, and the adsorption capacity is the largest at pH=6. The functional groups on the biochar surface are the adsorption centers. The adsorption enthalpy (ΔH) is always in the range of hydrogen bond energy. The weak interaction of the four adsorption configurations (A/B/C/D) is dominated by van der Waals forces and weak hydrogen bonds. The B configuration has the lowest binding energy and the most stable. The C configuration has the strongest weak interaction, and the D configuration is the weakest.
    References | Related Articles | Metrics
    Study on Drying Shrinkage of Recycled Tyre Polymer Fiber Reinforced Concrete
    CHEN Meng, WANG Yu-ting, CAO Yu-xin
    2023, 44 (1):  123-129.  DOI: 10.12068/j.issn.1005-3026.2023.01.017
    Abstract ( 560 )   HTML ( 4)   PDF (1674KB) ( 184 )  
    To investigate the effect of recycled tyre polymer fiber (RTPF) on the drying shrinkage properties of concrete, dry shrinkage tests and analysis of the mechanism of fiber action were carried out on plain concrete, RTPF (volume fraction is 0.1%, 0.2%, 0.4% and 0.8%,respectively) concrete and 0.1%polypropylene fiber (PPF) concrete. Results indicated that the slump of RTPF concrete with different contents decreases by 8.1%~62.2%, and the air fraction increases by 11.2%~47.9% as compared with plain concrete. The drying shrinkage rate of RTPF concrete increases rapidly at 0~7d and then tends to be stabilized. The addition of RTPF reduces the early growth rate of the drying shrinkage of concrete. The drying shrinkage rate of concrete decreases initially and then increases with the increase of RTPF content. When the volume fraction of RTPF is 0.2%, the concrete shows the minimum drying shrinkage rate. The drying shrinkage rates of RTPF concrete with different contents at 7d and 28d are 14.2%~36.0% and 2.9%~27.2%, respectively, which are lower than those of plain concrete. The drying shrinkage rate of PPF concrete is lower than that of RTPF concrete at the same volume fraction (0.1%). In terms of the inhibiting concrete drying shrinkage, RTPF with a volume fraction of 0.2% could replace PPF with a volume fraction of 0.1%. The scanning electron microscope showed that RTPF could bond well with the concrete matrix and transfers the shrinkage stress of the concrete matrix. Finally, by comparing the test data and fitting value of drying shrinkage rate, a drying shrinkage calculation model suitable for RTPF concrete was obtained.
    References | Related Articles | Metrics
    Seismic Behavior and Restoring Force Characteristic of Concrete Columns Reinforced with HRB600E High Strength Steel Bars
    LI Yan-yan, PAN Jin, YUE Yuan-yuan, ZHANG Ya-long
    2023, 44 (1):  130-137.  DOI: 10.12068/j.issn.1005-3026.2023.01.018
    Abstract ( 685 )   HTML ( 4)   PDF (3138KB) ( 105 )  
    Square section concrete columns with six HRB600E high strength steel bars and one HRB400E steel bars were tested under cyclic loading to investigate the seismic behavior of concrete columns with HRB600E high strength steel bars. The influence of axial compression ratio, stirrup spacing, the strength of longitudinal reinforcement and longitudinal reinforcement ratio on the seismic behavior of high strength reinforced concrete column was studied. Based on the test results, a restoring force model for concrete columns reinforced with HRB600E high strength steel bars was established. The test results indicated that the specimens had excellent performance in terms of hysteretic behavior, deformability and energy dissipation capacity. With the increase of axial compression ratio, the ductility decreased, but the bearing capacity and energy dissipation capacity increased. As the stirrup spacing decreased, the deformability and energy dissipation capacity increased. When the longitudinal reinforcement ratio increased, the bearing capacity increased, but the ductility and energy dissipation capacity decreased. The restoring force model of HRB600E high strength reinforced concrete column was in good agreement with the test results, which provides a reference for the seismic elastic-plastic analysis of engineering structures.
    References | Related Articles | Metrics
    Management Science
    Latent Profile Analysis on the Association of Energy Management Strategies with Work Outcomes
    ZHANG Lan-xia, JIA Yuan-yuan, DAI Guang-song, MAO Meng-yu
    2023, 44 (1):  138-144.  DOI: 10.12068/j.issn.1005-3026.2023.01.019
    Abstract ( 648 )   HTML ( 4)   PDF (587KB) ( 215 )  
    Based on the effort-recovery model and ego depletion theory, latent profile analysis (LPA) was used to classify 522 employees into different types of energy management strategies at work, and further analyze the differences of employees adopting different types of energy management strategies in terms of workplace well-being and work performance. It was shown that employees’ energy management strategies can be divided into four types, i.e., avoidant-energy users, work-energy users, casual-energy users and positive-energy users. Furthermore, employees classified as work-energy users and positive-energy users have the highest level of workplace well-being and work performance, and there is no significant difference in the outcomes of the two types, casual-energy users have the middle level of workplace well-being and work performance, and avoidant-energy users have the lowest level of workplace well-being and work performance among the four types.
    References | Related Articles | Metrics
    Analyzing the Causes of Traffic Accidents of Online Ride-Hailing Cars Using the Bayesian Network
    PENG Zhi-peng, PAN Heng-yan, WANG Yong-gang
    2023, 44 (1):  145-152.  DOI: 10.12068/j.issn.1005-3026.2023.01.020
    Abstract ( 708 )   HTML ( 10)   PDF (1765KB) ( 270 )  
    To mitigate the traffic accidents of online ride-hailing cars, the accident causes were studied by taking the ride-hailing drivers as the research object. A self-reported questionnaire survey was conducted to collect information about self characteristics, work intensity, work stress, risky driving behaviors, and accident history for 2458 ride-hailing drivers. After sorting of data, the Bayesian network method was used to establish the prediction model of accident frequency. The accuracy of the model was calibrated using the confusion matrix and receiver operating characteristic curve based on the ten-fold cross-validation. The results show that the model has a good prediction ability. The model found 11 influencing factors directly related to accident frequency and identified 16 categories of unfavorable states leading to an increased probability of high-frequency accidents. Also, nonlinear amplification and superposition effects of the combination of multiple unfavorable states on accident frequency were confirmed. The conclusions of the study help the management department to make prevention countermeasures to reduce the accident frequency of online ride-hailing cars.
    References | Related Articles | Metrics