Chinese  |  English

Most Download

  • Published in last 1 year
  • In last 2 years
  • In last 3 years
  • All

Please wait a minute...
  • Select all
    |
  • Orginal Article
    Yi LIU,Long LIU,Wangfeng LI,Yebin DONG,Xiuqing ZHANG
    Journal of Tsinghua University(Science and Technology). 2015, 55(1): 80-86.
    Abstract (48275) PDF (20529) HTML (116)   Knowledge map   Save

    An integrated method combining risk analysis, a CALPUFF dispersion model, and spatial analyses, were used to assess the regional atmospheric risk caused by major risk sources in petrochemical park plans. A typical petrochemical park in the planning stage was selected as the case study. The risks of six major sources, involving chlorine, phosgene, hydrogen sulfide, ammonia and acrylonitrile, were simulated in different seasons. The results show that the atmospheric risk is increased in 98.7% of the region 1141.5 km2. In 28.2% of the area, i.e. 326.5 km2, human health may be adversely affected after some accidents. The meteorological conditions in the fall are not conducive to pollution dispersion. The highest risks come from the diphenyl methane diisocyanate producing(MDI) project that contributes more than 90% of the increment in the regional atmospheric risk.

  • Orginal Article
    Xiaodong LI,Shu SU,Tianjian HUANG
    Journal of Tsinghua University(Science and Technology). 2015, 55(1): 50-55.
    Abstract (44475) PDF (18910) HTML (68)   Knowledge map   Save

    Construction dust is one of the most significant occupational health threats for field workers. Thus, the impact of dust on human health must be accurately evaluated to improve building health impact assessments and occupational health management. A health risk evaluation system and building health impact assessment system were used to develop a health damage assessment model to evaluate the effects of dust pollution from various construction activities at different stages and to monetize the evaluation results. A typical residential building in Beijing is used as a case sample with major activities at the main structure construction stage as evaluated objectives to illustrate the evaluation steps and demonstrate the model's application and operation. The health effect analysis among worker groups shows obvious differences in the dust pollution among various construction activities, with carpenters suffering the most health damage from dust during the main structure construction stage.

  • Orginal Article
    Qiujuan DAI
    Journal of Tsinghua University(Science and Technology). 2014, 54(1): 114-120.
    Abstract (44033) PDF (18736) HTML (70)   Knowledge map   Save

    Through administrating questionnaires to managers in Japanese foreign affiliates who used to work in state-owned enterprises (SOE), this paper intends to find out the effectiveness of these managers' SOE experiences in their current employment. Empirical evidence shows that the respondents' educational background, on-the-job training and work experiences in the state-owned enterprises in general help them climb the career ladder in the Japanese companies, thereby proving the effectiveness of their SOE experiences or skills. China's reform and opening up policy provides these managers with opportunities to start a new career in foreign companies, and the human capital accumulated in the planned economy is brought into full play after the economic transition.

  • Orginal Article
    Feng XIE, Yong PENG, Wei ZHAO, Yang GAO, Dejing WANG, Xuefeng HAN
    Journal of Tsinghua University(Science and Technology). 2014, 54(1): 29-34.
    Abstract (41076) PDF (17470) HTML (74)   Knowledge map   Save

    Industrial control devices (ICDs) are widely used in many industrial facilities such as petrochemical factories, power generation plants, water treatment plants, and transportation systems. The cyber security of such devices directly affects industrial production, which affects our economic security as well as our national security. This paper describes common cyber security issues in these devices and the root causes of these issues. The effects of cyber attacks are also analyzed for attacks from the information space and the physical space connected to the ICDs. Finally, security evaluation techniques are given including standards, evaluation methods and certifications along with future research challenges.

  • AUTOMOTIVE ENGINEERING
    ZHANG Xinyu, GAO Hongbo, ZHAO Jianhui, ZHOU Mo
    Journal of Tsinghua University(Science and Technology). 2018, 58(4): 438-444. https://doi.org/10.16511/j.cnki.qhdxxb.2018.21.010
    Abstract (9492) PDF (4031)   Knowledge map   Save CSCD(42)
    This paper introduces target recognition and detection methods based on the convolutional neural network (CNN) model, the improved regions with convolutional neural network (R-CNN) and the task-assistant convolutional neural network (TA-CNN) model for pedestrian detection. This paper also describes stereo matching based on a deep learning model for stereo matching using the Siamese network. Multi-source data fusion is also introduced based on a vision sensor, a radar sensor and a camera using a deep learning network. The CNN is used for end-to-end horizontal and vertical control of autonomous vehicles. Deep learning is widely used in the perception level, decision-making level and control level in automatic driving systems to continuously improve the perception, detection, decision-making and control accuracy. Analyses show that deep learning will improve of autonomous driving systems.
  • ELECTRONIC ENGINEERING
    TIAN Yao, CAI Meng, HE Liang, LIU Jia
    Journal of Tsinghua University(Science and Technology). 2016, 56(11): 1143-1148. https://doi.org/10.16511/j.cnki.qhdxxb.2016.26.002
    Abstract (8283) PDF (3512)   Knowledge map   Save CSCD(5)
    A hybrid model combining the deep neural network (DNN) for speech recognition and the i-vector model for speaker recognition has been shown effective for speaker recognition. The system performance is further improved by using the DNN with speaker labels to extract bottleneck features to replace the original short-term spectral features for statistics extractions to make the statistics contain more speaker-specific information to improve the speaker recognition. Tests on the NIST SRE 2008 female telephone-telephone-English task demonstrate the effectiveness of this method. The relative improvements of the bottleneck features are 7.65% for the equal error rate(EER) and 5.71% for the minium detection function(minDCF) compared with the short-term spectral features.
  • Orginal Article
    Xiao LUO, Yuan YAO, Jinhuan ZHANG
    Journal of Tsinghua University(Science and Technology). 2014, 54(3): 289-293.
    Abstract (7744) PDF (3292) HTML (13)   Knowledge map   Save

    Unified automotive safety technology is the tendency of developing automotive safety technologies. The occupant protection performance of conventional automotive passive safety technology can be further improved by cooperating with active sensors such as millimeter-wave radars and cameras. The view of interest in automotive passive safety technology is nearer than that in active safety technology. A calibration method for radars and cameras was developed using vehicle longitudinal symmetry plane as a datum to more accurately detect and locate in automotive passive safety application. Calibration parameters were obtained, with the coordinate transformation relationship between the two sensors being developed. Validation tests show that the radar and the camera in this method have good measurement accuracy, and the target location can be transformed into picture coordinate accurately. This method can provide reliable data for automotive safety systems.

  • Intelligent Construction
    LIU Tianyun
    Journal of Tsinghua University(Science and Technology). 2022, 62(8): 1281-1291. https://doi.org/10.16511/j.cnki.qhdxxb.2022.25.045
    Abstract (6656) PDF (2813) HTML (5)   Knowledge map   Save CSCD(1)
    This paper presents a 3D printing system using intelligence robots for rapid, efficient filling of large construction projects. The 3D printing system includes a construction scheduling system and a 3D assembly line that uses intelligence robots. The 3D scheduling system cuts the 3D digital design model into slices to calculate the filling material information and then plans the transport roads on the site map model for each step in the construction process. The construction robots collect the fill materials when needed and transport the fill materials with intelligent paving and rolling of the fill materials. Once each construction layer is finished, the robots send construction state information to the scheduling system. The complete filling process is then printed step by step under the control of the 3D printing scheduling system.
  • AUTOMOTIVE ENGINEERING
    XU Huachi, PEI Pucheng, WU Ziyao
    Journal of Tsinghua University(Science and Technology). 2016, 56(6): 587-591. https://doi.org/10.16511/j.cnki.qhdxxb.2016.22.016
    Abstract (6473) PDF (2733)   Knowledge map   Save CSCD(3)
    The response characteristics of linear sweep voltammetry (LSV) were used to develop an equivalent circuit model of a proton exchange membrane (PEM) fuel cell to distinguish the various electrochemical processes, including hydrogen desorption on the platinum, charging of the double-layer capacitance, electron internal short circuits and hydrogen crossover. This model eliminated the effect of the scan rate on the LSV results so that the hydrogen crossover current and membrane electronic resistance could be measured. A galvanostatic method was also used to measure the hydrogen crossover current and the electronic resistance. Measurements on a single cell with an active area of 34 cm2 with the LSV method show that the hydrogen crossover current is 1.19 mA·cm-2 and the electronic resistance is 479 Ω·cm2 while the galvanostatic measurements give the hydrogen crossover current of 1.25 mA·cm-2 and the electronic resistance of 413 Ω·cm2. This model can be used to analyze various electrochemical measurements in PEM fuel cells with the two methods giving complementary measurements for various PEM fuel cell processes.
  • .
    Abstract (6075) PDF (2635)   Knowledge map   Save
  • Orginal Article
    Ming FU,Wenguo WENG,Hongyong YUAN
    Journal of Tsinghua University(Science and Technology). 2014, 54(6): 719-723.
    Abstract (6020) PDF (2546) HTML (3)   Knowledge map   Save

    Bench scale tests are often used to measure the thermal insulation of protective clothing. The thermal insulation is one of the main parameters controlling the clothing thermal comfort and protective performance. A bench-scale system is developed to measure the heat transfer within a multi-layer fabric using a cone calorimeter. A model is then developed to calculate the thermal insulation in hot environments. This paper presents measurements of the thermal insulation of the multi-layer fabric for low intensity thermal radiation of 1-10 kW/m2. The tests show that the total thermal insulation decreases with increasing heat flux as the local resistances of the outer shell layer, the moisture layer and the thermal barrier layer decrease, while that of the inner layer increases at first and then decreases. The influence of the thermal radiation reflectivity can be neglected for low intensity thermal radiation less than 10 kW/m2.

  • Orginal Article
    Mingxiang YANG, Yunzhong JIANG, Yu TIAN, Hao WANG
    Journal of Tsinghua University(Science and Technology). 2014, 54(1): 133-136.
    Abstract (5953) PDF (2520) HTML (6)   Knowledge map   Save

    The idea of “Smart Water Resource” arises under the background of “Smart City”. Combining with the current actual problems in the development of water industry and the national strategic deployment, this paper expounds the urgency and necessity of the Smart Water Resource construction, and analyzes the feasibility of the construction of the Smart Water Resource from the existing basis, new technology and social cognition. Finally, this paper suggests that the top-level design should be implemented and the evaluation system should be improved, which provides references for the future construction of Smart Water.

  • Orginal Article
    Wenbo CHU,Yugong LUO,Jian LUO,Keqiang LI
    Journal of Tsinghua University(Science and Technology). 2014, 54(6): 724-728.
    Abstract (5717) PDF (2438) HTML (1)   Knowledge map   Save

    This paper presents a vehicle mass estimation method based on high-frequency information extraction to improve existing mass estimation methods that are susceptible to the influence of road slope with poor real-time performance. Accurate estimates are needed to accurately predict the driving force provided by the electrical drive systems. A high-pass filter is used to extract high-frequency longitudinal driving force and acceleration information. Then, a recursive least squares algorithm estimates the vehicle mass. Then, the road slope is estimated based on a combined kinematic and dynamic model. This method solves the problem that road slope estimates require an accurate vehicle dynamic model and are susceptible to acceleration sensor bias. The algorithm combines the dynamic method in a recursive least squares algorithm with a factor to neglect some previous information to estimate the road slope and a kinematic method that uses the relationship between the longitudinal vehicle acceleration and the acceleration sensor to calculate the road slope. Experimental tests show that this method is robust and can accurately estimate the vehicle mass and road slope in real-time.

  • FENG Guangshuo, ZHOU Ming
    Journal of Tsinghua University(Science and Technology). 2016, 56(10): 1114-1121. https://doi.org/10.16511/j.cnki.qhdxxb.2016.22.048
    Abstract (5696) PDF (2412)   Knowledge map   Save CSCD(14)
    The objective of this study is to assess spark ignition (Otto cycle) and compression ignition (Diesel cycle) engines as heavy fuel aircraft piston engines. Spark ignition aircraft engines give a higher power to weight ratio, but have higher specific fuel consumption (SFC), knocking, poor starting, higher electro magnetic interference (EMI), lower reliability and narrower power regions. Spark ignition systems use direct injection, pneumatic atomizers and high energy ignition systems. Compression ignition aircraft engines have lower SFC, improved range, lower EMI, higher reliability and wider power regions, but have worse power to weight ratios and more vibration. Compression ignition engines have advanced electronic controls, fuel injection systems and variable high-pressure ratio superchargers. The assessments show that both approaches are feasible with some technical challenges, with compression ignition engines as the more promising approach.
  • INFORMATIONENGINEERING
    GUO Yongde, GAO Jinhuan, MA Hongbing
    Journal of Tsinghua University(Science and Technology). 2016, 56(10): 1122-1130. https://doi.org/10.16511/j.cnki.qhdxxb.2016.22.049
    Abstract (5687) PDF (2406)   Knowledge map   Save CSCD(12)
    Nighttime light imagery removes most natural disturbances, so nighttime images can be used to reflect human activity, especially for research on spatialization of socio-economic development. This study utilized night-light data collected by the Suomi-NPP satellite in a spatial correlation model with GDP data to analyze both the spatial distribution and factors influencing development. The model extracted the lighting information and calculated night-light indexes to select the best night-light index for each Chinese mainland province. Tests show that the normalized total radiance index relates well to the GDP, so this index is related to the GDP using linear and nonlinear spatial models to develop a model to predict the GDP. The fitting results for the linear, power law and logistic models are all greater than 0.8. The power law model predicts the GDP of each mainland provincial-level region in 2014 with an average relative error of only 26.0%.
  • AUTO MATION
    HE Zhichao, YANG Geng, LU Languang, Wu Haisang
    Journal of Tsinghua University(Science and Technology). 2015, 55(5): 532-537.
    Abstract (5515) PDF (2343)   Knowledge map   Save
    The direct current internal resistance (DCIR) is the sum of a battery's ionic and electronic resistances. The DCIR test indicates the battery's power characteristics and reflects the batteries' aging and uniformity characteristics. Thus, it is important for battery modeling and applications. This paper describes a DCIR test method based on the battery's constant current external characteristics. This method normalizes the battery's state of charge (SOC) changes for different constant current conditions. Then, the DCIR for different operating currents and SOC are obtained using constant current charge/discharge curves. This method is easier to implement than existing methods and has good accuracy.
  • PHYSICS AND ENGINEERING PHYSICS
    ZHANG Xianwen, ZHANG Liyan, DING Lichao, WEI Rongrong, WANG Jie, TANG Jintian
    Journal of Tsinghua University(Science and Technology). 2017, 57(7): 763-767. https://doi.org/10.16511/j.cnki.qhdxxb.2017.25.034
    Abstract (5416) PDF (2303)   Knowledge map   Save CSCD(8)
    A ballistocardiogram (BCG) unobtrusively records the weight variation caused by the reaction force of blood ejection and blood flow in the aorta. A BCG system was developed to record BCG signals and detect the heart beats in the BCG signal. Data from 24 healthy subjects was collected to investigate the system with simultaneous electrocardiograms (ECG) as a reference. The results show no significant difference between the heartbeats detected by the ECG and BCG system (p=0.95) with the BCG system having a sensitivity of 99.56% and an accuracy of 99.77%.
  • BIOLOGICAL SCIENCE AND ENGINEERING
    FENG Qianqian, WANG Wenjuan, LI Haidong, PAN Xun
    Journal of Tsinghua University(Science and Technology). https://doi.org/10.16511/j.cnki.qhdxxb.2017.26.034
    Accepted: 2017-06-22
    Abstract (5419) PDF (2293)   Knowledge map   Save
    The autofluorescence of nicotiana benthamiana chloroplasts was studied using a laser scanning confocal microscope. The chloroplast autofluorescence was obtained by spectral scanning with various excitation wavelengths. Then, chloroplast autofluorescence images were obtained using laser scanning confocal microscopy with various excitation wavelengths with intensity quantification. Finally, nicotiana benthamiana chloroplasts were illuminated with 488 nm light to test the autofluorescence stability. 488 nm light has the highest excitation efficiency for chloroplast autofluorescence, while 561 nm light has the lowest excitation efficiency. The autofluorescence chloroplast emission spectrum starts from 637 nm and peaks at 681 nm. The measurements show that the chloroplast autofluorescence is very stable and that laser scanning confocal microscopy can accurately measure chloroplasts autofluorescence. This work will facilitate improved plant tissue imaging chloroplast genetic engineering research.
  • NUCLEAR ENERGY AND NEW ENERGY
    NIE Junfeng, TANG Zhenrui, ZHANG Haiquan, LI Hongke, WANG Xin
    Journal of Tsinghua University(Science and Technology). 2017, 57(7): 780-784. https://doi.org/10.16511/j.cnki.qhdxxb.2017.25.037
    Abstract (5347) PDF (2280)   Knowledge map   Save CSCD(2)
    Crystal plasticity theory is a fundamental theory that combines the crystal microscopic slip mechanism with macroscopic plastic deformation to predict meso-scale plastic deformation. The dislocation density has an important influence on the hardening behavior of metal crystals. This paper presents a constitutive model based on crystal plasticity theory and dislocation motion theory for the BCC crystal structure. The model is used to study the mechanical behavior of a BCC lattice. Using the UMAT subroutine in ABAQUS for numerical simulations of a uniaxial tensile tests of single crystal and polycrystal iron. The results show that the constitutive model effectively simulates the mechanical behavior of the uniaxial tensile test for single crystal and polycrystal iron.
  • COMPUTER SCIENCE AND TECHNOLOGY
    ZOU Quanchen, ZHANG Tao, WU Runpu, MA Jinxin, LI Meicong, CHEN Chen, HOU Changyu
    Journal of Tsinghua University(Science and Technology). 2018, 58(12): 1079-1094. https://doi.org/10.16511/j.cnki.qhdxxb.2018.21.025
    Abstract (5101) PDF (2160)   Knowledge map   Save CSCD(15)
    In recent years, the increasing size and complexity of software packages has led to vulnerability discovery techniques gradually becoming more automatic and intelligent. This paper reviews the search characteristics of both traditional vulnerability discovery techniques and learning-based intelligent vulnerability discovery techniques. The traditional techniques include static and dynamic vulnerability discovery techniques which involve model checking, binary comparisons, fuzzing, symbolic execution and vulnerability exploitability analyses. This paper analyzes the problems of each technique and the challenges for realizing full automation of vulnerability discovery. Then, this paper also reviews machine learning and deep learning techniques for vulnerability discovery that include binary function identification, function similarity detection, test input generation, and path constraint solutions. Some challenges are the security and robustness of machine learning algorithms, algorithm selection, dataset collection, and feature selection. Finally, future research should focus on improving the accuracy and efficiency of vulnerability discovery algorithms and improving the automation and intelligence.
  • ELECTRONIC ENGINEERING
    FANG Yong, LIU Daosheng, HUANG Cheng
    Journal of Tsinghua University(Science and Technology). 2017, 57(6): 620-624. https://doi.org/10.16511/j.cnki.qhdxxb.2017.26.029
    Abstract (4934) PDF (2114)   Knowledge map   Save CSCD(1)
    Since there are many malicious users on the Internet, popular online websites sometimes have millions of registered users. The system cannot easily distinguish between fake accounts and legitimate users. Fake accounts registered by a single malicious user often have similar profiles. This paper presents a new framework to find fake accounts in large numbers of users. The framework uses username string patterns to classify the original data and then calculates the similarity as measured by the Levenshtein distance between any two elements in one category. Hierarchical clustering with a proper threshold then finds groups of fake accounts hidden in the large amount of registration data. Tests demonstrate the effectiveness of this framework which algorithm relies less on data dimensions and features than other algorithms.
  • INFORMATION ENGINEERING
    ZHAO Zhuofeng, LU Shuai, HAN Yanbo
    Journal of Tsinghua University(Science and Technology). 2017, 57(2): 220-224. https://doi.org/10.16511/j.cnki.qhdxxb.2017.22.018
    Abstract (4816) PDF (2042)   Knowledge map   Save CSCD(7)
    Vehicle license plate recognition data provides a kind of traffic monitoring data that is a large spatial-temporal stream with fixed positions. Similar trajectory queries of such data face several problems. This paper presents a similar trajectory query method based on site companions with multistage task parallelization based on the MapReduce computing model. This method gives more efficient similar trajectory queries in a distributed computing environment for massive license plate recognition data. Tests show that this method can correctly query similar trajectories more efficiently than traditional stand-alone methods based on tests with almost ten million real vehicle license plate data points.
  • ELECTRONIC ENGINEERING
    SHI Chenbo, MIAO Quan, CHEN Qixin
    Journal of Tsinghua University(Science and Technology). 2016, 56(9): 930-936,941. https://doi.org/10.16511/j.cnki.qhdxxb.2016.21.061
    Abstract (4529) PDF (1916)   Knowledge map   Save CSCD(7)
    Many new internet systems and patterns are being developed that will positively influence the development and use of renewable energy. The rise of the third industrial revolution based on the energy internet will use cyber physical systems (CPS) combining information, communication and sensors integrated with energy systems in the electrical grid. The resulting cyber energy systems can optimize renewable energy use; however, the integration of CPS and energy systems is not well understood. This paper describes the structure and characteristics of the energy internet. The internet and energy systems are integrated for the production, transmission, storage and sharing of energy. The new application and recent trends in the energy internet are discussed along with challenges and suggestions on how to promote the system development.
  • Orginal Article
    Lungui ZHENG,Zheng YOU,Gaofei ZHANG,Shumin ZHAO
    Journal of Tsinghua University(Science and Technology). 2014, 54(6): 794-798.
    Abstract (4498) PDF (1907) HTML (7)   Knowledge map   Save

    When antennae are blocked by buildings or the signals are drowned by strong noise, global navigation satellite system (GNSS) receivers require both long coherent integration times and a large number of non-coherent integration operations to acquire the signal. However, navigation data bit transition during the long coherent integration results in integrator signal energy attenuation and non-coherent integration will have squared losses that limit the ability to acquire weak signals. A signal acquisition algorithm is given here based on non-coherent integration that estimates the optimal combination of navigation data bits with the maximum energy criterion to eliminate the length limit and improve the ability to acquire weak signals. Tests demonstrate that these techniques greatly increase the gain such that low signal to noise ratio (SNR) signals as low as 20 dBHz can be successfully acquired.

  • Orginal Article
    Fan MENG, Yonggui DONG
    Journal of Tsinghua University(Science and Technology). 2014, 54(2): 207-211.
    Abstract (4455) PDF (1901) HTML (7)   Knowledge map   Save

    When a square wave pulsed signal is adopted as excitation to obtain electrical impedance information for a conductive medium, the amplitude decay of the higher harmonics in square wave spectrum can have a significant effect. Adding an inductor or capacitor to a first-order resistant-capacitive-type sensing system creates a second-order resonant element that creates a freely damped oscillating signal at both the rising and falling edges of the excitation. The excitation amplitude near the resonant frequency is, therefore, enhanced. The equivalent resistance in the resonant element can be obtained by measuring the first positive peak voltage of the freely damped signal at the falling edge of the pulse. Tests indicate that resistance measurements for frequencies from 30 kHz to 1 MHz can be implemented only by a fixed frequency square wave excitation signal generated by a digital circuit. This method can be used for conductivity measurements of conductive media, such as electrolyte solutions, water and human skin, using both wired and wireless transmitters.

  • INFORMATION ENGINEERING
    Aisikaer Rouzi, YIN Shi, ZHANG Zhiyong, WANG Dong, Askar Hamdulla, ZHENG Fang
    Journal of Tsinghua University(Science and Technology). 2017, 57(2): 182-187. https://doi.org/10.16511/j.cnki.qhdxxb.2017.22.012
    Abstract (4469) PDF (1879)   Knowledge map   Save CSCD(4)
    Speech data plays a fundamental role in research on speech recognition. However, there are few open speech databases available for researchers in China, especially for minor languages such as Uyghur. This paper develops a Uyghur continuous speech database which is totally open and free. The database consists of 20 h of training speech and 1 h of test speech, as well as all the resources needed to construct a full Uyghur speech recognition system, including a phone set, lexicon, and text data. A recipe used to construct the baseline system is also described with results for two test sets involving clean speech and noisy speech. This paper provides a standard database for Uyghur speech recognition.
  • AUTOMOTIVE ENGINEERING
    DING Jieyun, DANG Ruina, WANG Jianqiang, LI Keqiang
    Journal of Tsinghua University(Science and Technology). 2015, 55(7): 769-774.
    Abstract (4356) PDF (1841)   Knowledge map   Save
    An accurate lane change intention recognition algorithm is developed to improve real-time performance. The algorithm analyzes the drivers' lane change decisions to develop a new symbol and a comprehensive decision index (CDI) based on fuzzy theory to assess the probability that the driver will change lanes. Then, the driver intention recognition algorithm is designed based on a hidden Markov model. Using the new symbol as well as representative lateral motion parameters as observed signals, and the driver's intention as the hidden state, a hidden Markov model is built and trained. The driver's intention is recognized by the HMM decoding method. Lane change data collected on a driving simulator are used to verify the overall algorithm performance. The results show that the algorithm with the CDI as one of the observation signals both guarantees the accuracy of the recognition results and improves the real-time performance.
  • ELECTRONIC ENGINEERING
    GUO Wu, MA Xiaokong
    Journal of Tsinghua University(Science and Technology). 2016, 56(11): 1190-1195. https://doi.org/10.16511/j.cnki.qhdxxb.2016.26.010
    Abstract (4261) PDF (1830)   Knowledge map   Save CSCD(2)
    A voice activity detection (VAD) algorithm was developed for robust voice detection in complex noise conditions. The energy, the most dominant component and the spectral entropy are used to form three dimensional features that have been demonstrated to strongly complement each of them in the presence of complex noise. The K-mean algorithm is used to adaptively select the feature and to calculate the utterance dependent thresholds, which are applied in the following speech detection process. Tests on the NIST SRE 2008 and 2012 corpus show that this algorithm gives better performance for different noise conditions and is more robust and efficient than conventional unsupervised and supervised methods.
  • Orginal Article
    Hongyu LIU,Peiyan JIANG
    Journal of Tsinghua University(Science and Technology). 2015, 55(1): 56-62.
    Abstract (4284) PDF (1821) HTML (8)   Knowledge map   Save

    This study investigates the influence of the price elasticity of the land supply in China (PELS), which represents the characteristics of the local government land supply behavior, using a theoretical model of the land and housing markets to analyze the situations when the land supply exceeds the demand and the demand exceeds the supply. This paper finds that the PELS greatly affects the land and housing market equilibrium, with a higher PELS bringing local governments more land leasing revenue. Monthly data for 35 major cities in China from January, 2006 to September, 2013 show that the fluctuation in housing price and trading volume and the financial dependence on the land leasing revenue of the local governments significantly affect the PELS. Local governments should make reasonable plans for the land supply according to development in these cities, and ensure an effective land supply for housing.

  • Orginal Article
    Run TAO,Zhichao HOU
    Journal of Tsinghua University(Science and Technology). 2014, 54(6): 744-749.
    Abstract (4286) PDF (1812) HTML (6)   Knowledge map   Save

    A device is developed to measure the friction torque of rolling bearings subjected to radial loads while running at steady speeds. The friction torques are measured at various speeds and radial loads for typical rolling bearings used in an engine front end accessory belt drive of a passenger vehicle. The least squares method is used to fit the friction torque and the rotational speed with the optimal function chosen according to the correlation coefficient and significance level. A set of empirical formulae are then derived for the friction torque in terms of both the rotational speed and the radial load. The expressions are more accurate than the Palmgren formula. The test method can be applied to measure the friction torque of rolling bearings used in other fields as a valuable method to characterize the torque characteristics.

  • PHYSICS AND ENGINEERING PHYSICS
    SU Boni, HUANG Hong, ZHANG Nan
    Journal of Tsinghua University(Science and Technology). 2015, 55(6): 684-690.
    Abstract (4265) PDF (1804)   Knowledge map   Save
    A quantitative risk assessment method was developed for urban waterlogging caused by rainstorms. The spatio-temporal distributions of water are simulated through a two-dimensional hydraulic model, with the losses caused by waterlogging then estimated from a vulnerability curve obtained using a domestic field investigation. An area in Xinluo District, Longyan, Fujian Province was chosen as the study area. The waterlogging risk was assessed by calculating the spatio-temporal water distribution and the disaster losses for various rainstorm scenarios. The results show that rainstorms with longer duration and longer return period cause more serious waterlogging and economic losses. The waterlogging points of different rainstorms are similar. The effects of drainage well capability on reducing the urban waterlogging risk were analyzed through economic loss estimated for various drain distributions. The results show that drainage wells effectively reduce the waterlogging risk, but the effect is limited during short and heavy rainstorms.
  • INFORMATION SECURITY
    PENG Yong, XIANG Chong, ZHANG Miao, CHEN Dongqing, GAO Haihui, XIE Feng, DAI Zhonghua
    Journal of Tsinghua University(Science and Technology). 2016, 56(1): 14-21. https://doi.org/10.16511/j.cnki.qhdxxb.2016.23.013
    Abstract (4244) PDF (1795)   Knowledge map   Save CSCD(5)
    Industrial control systems (ICSs) are cyber-physical systems (CPSs) which supervise and control physical processes in critical infrastructure industries such as electric power, water treatment, oil & natural gas exploration, transportation, and chemical industry. Based on the observation of ICS'stable and persistent communication data flow control patterns, a concept and a methodology of ICS scenario fingerprinting were proposed which analyze industrial control protocol interactive behavior to represent ICS system-level normal behavior characteristics. ICS scenario fingerprint can identify unique ICS installation, while being used as a more generalized method to establish ICS systems'behavior benchmark and further being used to identify ICS systems'abnormal behavior. Experiments were made to validate the proposed viewpoint, which use real equipment for ICS cyber domain and use simulation for ICS physical domain. Experimental results demonstrate that ICS scenario fingerprinting technique provides ICS security research with a promising method.
  • Orginal Article
    Yuan LIN, Shu LIANG, Shengjin WANG
    Journal of Tsinghua University(Science and Technology). 2014, 54(3): 334-340.
    Abstract (4157) PDF (1757) HTML (6)   Knowledge map   Save

    3-D scanned face registration is the foundation and the prerequisite for 3-D faces statistical learning. To solve the problem of registering these digital points accurately, a 3-D faces registration method was developed using a non-rigid iterative closet point (ICP) algorithm combined with elastic deformation. By using the sparse 3-D feature vertices, the method fits high-resolution 3-D template meshes (3-D template faces) to 3-D face range scans (3-D target faces) through deformations step by step. This deformation process can be formulated as an optimization problem for the affine transformation at each vertex of the template face. In addition, this algorithm accelerates the registration process without sacrificing the registration accuracy by introducing elastic deformation. On this basis, a head missing data completion method for 3-D faces, combining the elastic deformation and the proposed 3-D registration, was developed to enhance the reality and the completeness of the 3-D face models. The experimental results show that this method is more robust than a widely used face completion method.

  • COMPUTER SCIENCE AND TECHNOLOGY
    ZHANG Min, DING Biyuan, MA Weizhi, TAN Yunzhi, LIU Yiqun, MA Shaoping
    Journal of Tsinghua University(Science and Technology). 2017, 57(10): 1014-1021. https://doi.org/10.16511/j.cnki.qhdxxb.2017.25.039
    Abstract (3971) PDF (1695)   Knowledge map   Save CSCD(5)
    Collaborative filtering based on matrix factorization has been very successful, while cold-start and data sparseness problems have not been well resolved. Hence, many studies have attempted to include review information into rating predictions. This paper presents a hybrid model that introduces deep learning into recommendation system with collaborative filtering. The algorithm combines a stacked denoising auto encoder (SDAE) with a latent factor model (LFM) to make use of both review and rating information to improve the rating predictions. Evaluations on a large, commonly used Amazon dataset show that this approach significantly improves the rating prediction accuracy in comparison with traditional models, with up to 64.43% better predictions.
  • COMPUTER SCIENCE AND TECHNOLOGY
    YI Shengwei, ZHANG Chongbin, XIE Feng, XIONG Qi, XIANG Chong, LIANG Lulu
    Journal of Tsinghua University(Science and Technology). 2017, 57(1): 50-54. https://doi.org/10.16511/j.cnki.qhdxxb.2017.21.010
    Abstract (3948) PDF (1673)   Knowledge map   Save CSCD(9)
    Fuzzing tests are important for discovery of unknown vulnerabilities and risks. A security analysis method was developed for industrial control networks using the Peach fuzzing framework. The system uses the mutation strategy by fabricating abnormal network packets, sending these packets to the target and then executing tests. The tests monitor the status of the industrial control network protocols. The system then identifies exceptions in the industrial control network protocols. Modbus TCP, a widely used industrial control network protocol is analyzed as an example using a fuzzy Modbus TCP protocol. The results show that this method can effectively identify vulnerabilities in industrial control network protocols.
  • Orginal Article
    Biqing HUANG,Yi ZHANG,Xiaochun YI
    Journal of Tsinghua University(Science and Technology). 2014, 54(4): 522-529.
    Abstract (3904) PDF (1660) HTML (7)   Knowledge map   Save

    The rapid development of the offshore wind power industry requires excellent operation and maintenance systems for the offshore wind turbines. However, there is little offshore wind farm operating and management experience and the overall operating costs are high. These restrict healthy development of the wind power industry. Hence, this paper describes operating and maintenance systems for large equipment industries that can be used in the wind power industry. The special needs for the operation and maintenance of wind farms, especially offshore wind farms, are analyzed to identify upcoming problems and future development needs. The overall system framework is given with analyses of the business functions of the various subsystems and their relationships. Two key systems for the bill of materials (BOM) and maintenance workflows are discussed. Software systems are given for wind farm applications.

  • COMPUTER SCIENCE AND TECHNOLOGY
    ZHANG Yangsen, ZHENG Jia, HUANG Gaijuan, JIANG Yuru
    Journal of Tsinghua University(Science and Technology). 2018, 58(2): 122-130. https://doi.org/10.16511/j.cnki.qhdxxb.2018.22.015
    Abstract (3920) PDF (1656)   Knowledge map   Save CSCD(17)
    Microblog sentiment analysis is used to get a user's point of view. Most sentiment analysis methods based on deep learning models do not use emotion symbols. This study uses a double attention model for microblog sentiment analysis that first constructs a microblog emotion symbol knowledge base based on existing emotional semantic resources including emotion words, degree adverbs, negative words, microblog emoticons and common Internet slang. Then, bidirectional long short-term memory and a full connection network are used to encode the microblog text and the emotion symbols in the text. After that, an attention model is used to construct the semantic representations of the microblog text and emotion symbols which are combined to construct the final semantic expression of the microblog text. Finally, the emotion classification model is trained on these semantic representations. The combined attention model and emotion symbols enhance the ability to capture the emotions and improve the microblog sentiment classification. This model gives the best accuracy for many sentiment classification tasks on the Natural Language Processing and Chinese Computing (NLPCC) microblog sentiment analysis task datasets. Tests on the 2013 and 2014 NLPCC datasets give F1-scores for the macro and micro averages that are 1.39% and 1.26% higher than the known best model for the 2013 dataset and 2.02% and 2.21% higher for the 2014 dataset.
  • COMPUTER SCIENCE AND TECHNOLOGY
    HONG Zhixu, CHEN Hao, CHENG Liang
    Journal of Tsinghua University(Science and Technology). 2017, 57(3): 264-269. https://doi.org/10.16511/j.cnki.qhdxxb.2017.26.007
    Abstract (3882) PDF (1640)   Knowledge map   Save
    This paper presents distributed data integration and visualization methods for distributed data integration for complex social governance data sets which have weak data analysis. The enhanced dynamic data description and Web visualization capabilities facilitate service-oriented intelligence society governance decision analyse using the large data processing mode to access, extract, integrate, and mine scattered data in different network routing databases. This paper presents a prototype system with trial evaluations. The results show that this method enhances real-time data integration, security, accuracy and data analysis visualization, and is more streamlined than the traditional model.
  • CIVIL ENGINEERING
    ZHANG Zhihui, ZHANG Jianhan
    Journal of Tsinghua University(Science and Technology). 2016, 56(12): 1327-1332. https://doi.org/10.16511/j.cnki.qhdxxb.2016.25.046
    Abstract (3843) PDF (1628)   Knowledge map   Save
    The basic features of urban rail transit (URT) such as the shared operational responsibilities and the partial public ownership determine the public-private partnership (PPP) mode for developing URT projects. In this paper, the use of the PPP mode to develop URT projects is analyzed to study the relationship between the external benefits and government subsidies on the URT. Government regulations were used to develop a model for URT annual operating subsidies that was used to predict the government operating subsidies for Beijing subway line 4 in 2012. Then, the "with and without methods" were used to analyze Beijing subway line 4 in 2012 to determine the external benefits of this project. The results show that the government operating subsidies for Beijing subway line 4 in 2012 were significantly lower than the external benefits and that the government operating subsidies for this project are reasonable.
  • CIVIL ENGINEERING
    MA Ding, CHEN Wenying
    Journal of Tsinghua University(Science and Technology). 2017, 57(10): 1070-1075. https://doi.org/10.16511/j.cnki.qhdxxb.2017.25.047
    Abstract (3851) PDF (1627)   Knowledge map   Save CSCD(17)
    In this study, an integrated carbon emission peak path model system was built based on China TIMES model, and was used to analyze China's carbon emissions peak and peak path. The results show that China's carbon emissions will maintain rapid growth in the reference scenario between 2010-2050, and give enormous pressure on China's energy security and addressing climate change; in emission peak scenarios, through the development of non-fossil energy and the adoption of energy-conservation and emission-reduction measures, carbon-intensive sectors (power and industry sectors) can achieve early emissions peak and guarantee the overall carbon emissions peak between 10.0-10.8 billion t; power sector and energy-intensive sectors are the main carbon mitigation sectors, the contribution of carbon mitigation are 75% and 15%, respectively. In addition, adopting non-fossil energy and the energy-efficient technologies are main carbon mitigation measures, and the contribution of carbon mitigation are 65% and 15%, respectively.