15 August 2020, Volume 60 Issue 8

    

• Select all
  |

SPECIALSECTION: FRONTIER RESEARCH AND APPLICATIONS OF TRIBOLOGY
• Select
  Development and prospects of liquid superlubricity

  YI Shuang, GE Xiangyu, LI Jinjin

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 617-629. https://doi.org/10.16511/j.cnki.qhdxxb.2020.25.023

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  The rapid industrial development in China is requiring more energy consumption based on non-renewable energy resources. The energy consumption caused by unnecessary friction accounts for about 4.5% of the GDP in China. Many researchers in tribology are investigating superlubricity applications to reduce friction in machinery. Superlubricity improves the lubrication efficiency which reduces the energy consumption and considerably increases the energy utilization rate. This study reviews superlubricity characteristics and mechanisms for lubrication applications. The characteristics of recently developed liquids are reviewed with descriptions of the liquid superlubricity mechanism. Finally, the advantages and disadvantages of the current research on superlubricity are summarized.
  
• Select
  Polyelectrolyte lubrication coatings for biomedical applications

  WANG Xizi, HAN Ying, ZHAO Weiwei, ZHANG Hongyu

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 630-638. https://doi.org/10.16511/j.cnki.qhdxxb.2020.25.016

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Polyelectrolytes are charged polymers that are widely used in biomedical applications. This paper reviews various polyelectrolyte lubrication coatings. This review introduces the lubrication mechanisms and polyelectrolyte classifications which are based on the polyelectrolyte brush and hydration lubrication mechanisms. Then, the polyelectrolytes are divided into three categories. The review also describes how the polyelectrolytes adhere to the substrates through physical methods, coupled grafting and initiation grafting. The initiation grafting is described in detail as a reference for the design of polyelectrolyte lubrication coatings. Finally, the potential for polyelectrolyte lubrication coatings in the biomedical field is discussed with future prospects.
• Select
  Optimization of the tribological characteristics of lubricant materials with granulated ZrO₂

  JI Zhengjia, JIN Hongyun, LUO Wanyue, ZHOU Keke, HOU Shuen, XIE Guoxin

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 639-647. https://doi.org/10.16511/j.cnki.qhdxxb.2020.25.027

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  This study investigated the effects of granulated ZrO₂ on the tribological characteristics of lubrication composites. The composite materials were evaluated using constant speed friction and wear tests with the data analyzed using the preference selection index (PSI) method. The results showed that the optimum granulated ZrO₂ content in this composite was 10% with the friction factor of 0.49. Fade tests and recovery tests also showed the variations of the wear rate, friction coefficient, fade ratio, recovery ratio, stability coefficient, and the friction fluctuation and variability coefficient for various granulated ZrO₂ concentrations.
  
SPECIALSECTION: DATABASE
• Select
  Multi-neural network collaboration for Chinese military named entity recognition

  YIN Xuezhen, ZHAO Hui, ZHAO Junbao, YAO Wanwei, HUANG Zelin

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 648-655. https://doi.org/10.16511/j.cnki.qhdxxb.2020.25.004

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Web data contains a large amount of high-value military information which has become an important data source for open-source military intelligence. Military named entity recognition is a basic, key task for information extraction, question answering and knowledge graphs in the military domain. Military named entity recognition faces some unique challenges not seen in searches for named entities in other domains, such as military named entity boundaries being vague and difficult to define, lack of standardized military terms in Internet media, extensive use of abbreviations, and the lack of a public military-oriented corpus. This paper presents an entity labeling strategy that includes the effects of fuzzy entity boundaries and a military-oriented corpus called MilitaryCorpus based on microblog data constructed by combining domain expert knowledge. A multi-neural network collaboration approach is then developed based on a named entity recognition model. The character level features are learned in the BERT (bidirectional encoder representations from transformers)-based Chinese character embedding representation layer with the context features extracted in the BiLSTM (bi-directional long short-term memory) neural network layer to form the feature matrix. Finally, the optimal tag sequence is generated in the CRF (conditional random field) layer. Tests show that the recall rate and the F-score of the BERT-BiLSTM-CRF model are 28.48% and 18.65% higher than those of a CRF-based entity recognition model, 13.91% and 8.69% higher than those of a BiLSTM-CRF-based entity recognition model, and 7.08% and 5.15% higher than those of a CNN (convolutional neural networks)-BiLSTM-CRF-based model.
• Select
  TFP: Efficient algorithm for fastest path queries

  DU Ming, ZHENG Kaiwen, CHEN Ziyang, ZHOU Junfeng

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 656-663. https://doi.org/10.16511/j.cnki.qhdxxb.2020.22.007

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  The fastest path query returns the time span of the shortest path between two vertices in a temporal graph. Efficient responses to a fastest path query effectively improve system use and enhance user desire to continue using the system. However, existing fastest path query methods on temporal graphs have many redundant operations in their processing strategies, so they are not very efficient. Three heuristic rules are presented here to reduce the number of redundant computations with verification of their effectiveness. These three heuristic rules are then integrated into an efficient universal query algorithm to find the fastest path. This method reduces the query invocation reachability 5-8 fold on various data sets compared with the original method by significantly reducing redundant computations for higher query processing efficiencies.
• Select
  Alzheimer's disease classification method based on multi-modal medical images

  HAN Kun, PAN Haiwei, ZHANG Wei, BIAN Xiaofei, CHEN Chunling, HE Shuning

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 664-671,682. https://doi.org/10.16511/j.cnki.qhdxxb.2020.25.003

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Multi-modal medical image information has been widely used in computer-aided detection and computer-aided diagnosis in the medical community. Feature information from multi-modal medical images can be used to accurately classify and diagnosis Alzheimer's disease (AD) characteristics. This paper presents a convolutional neural network model for 3D convolution operations on magnetic resonance imaging (MRI) and positron emission computed tomography (PET) images of Alzheimer's subjects to extract the feature information for the various modalities. Then, models are used to fuse these modal information sets into a rich multi-modal feature information dataset. Finally, this dataset is classified and predicted using a fully connected neural network. Tests on the public data set of the AD neuroimaging initiative show that this model more accurately evaluates accuracy (ACC) and area under the curve (AUC) conditions.
• Select
  Online route planning algorithm in spatial crowdsourcing

  CUI Junyun, CHEN Di, YUAN Ye, MA Yuliang, WANG Guoren

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 672-682. https://doi.org/10.16511/j.cnki.qhdxxb.2020.26.009

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Worker route planning is an important spatial crowdsourcing research topic. Previous research on route planning has mainly focused on offline situation. However, online situation are more relevant to actual needs. Therefore, this study analyzes an online route planning problem extracted from a typical spatial crowdsourcing platform involving crowdsourcing logistics and shared bus, which is called the online route planning problem with fixed endpoints in spatial crowdsourcing. The first step is to analyze the route planning in the Euclidean space using a particle swarm optimization method. The algorithm plans the route by tracking the optimal solution online. A local particle swarm optimization method for online route planning is also used based on the k nearest neighbours. The route planning algorithm is then applied to a road network using weighted shortest path boundary indexing. The effectiveness and efficiency of this algorithm are verified through extensive experiments on real datasets with the local particle swarm optimization method for online route planning showing better performance than the compared method.
• Select
  Mining spatio-temporal co-location fuzzy congestion patterns from traffic datasets

  WANG Xiaoxu, WANG Lizhen, WANG Jialong

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 683-692. https://doi.org/10.16511/j.cnki.qhdxxb.2020.25.012

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Traffic congestion occurs when the total traffic volume on the road network exceeds the road capacity which disrupts normal traffic flow. The congestion patterns in traffic datasets need to be mined to effectively address urban traffic congestion problems. However, existing research work has failed to reasonably and accurately define "traffic congestion" and ignores the spatio-temporal attributes of the traffic flow data and the fuzziness of the traffic congestion concept. This paper presents the concept of spatio-temporal co-location fuzzy congestion patterns by introducing fuzzy set theory into the definition of traffic congestion to measure the degree of traffic congestion. The algorithm also adds the time attribute to the traditional spatial co-location pattern mining. Two algorithms are then presented for mining spatio-temporal co-location fuzzy congestion patterns. The methods are evaluated using real traffic datasets with the results showing that the methods provide better mining results than the existing methods.
SPECIALSECTION: PROCESS SYSTEMS ENGINEERING
• Select
  Fault detection based on orthogonal local slow features

  ZHANG Zhanbo, WANG Zhenlei, WANG Xin

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 693-700. https://doi.org/10.16511/j.cnki.qhdxxb.2020.25.026

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  A local time-space regularized slow feature extraction method was developed to improve data-driven fault detection in the chemical industry based on the process dynamics of closed-loop control systems and the local information contained in the data manifold. An objective function was defined based on the local time-space term to obtain a projection matrix and the pre-extraction feature, S. The span of S contains the static information, while the first derivative of the span of S contains the dynamic information. An independent component analysis was used to obtain statistics for S² and SPE for both spaces for real-time fault detection. A case study on the Tennessee Eastman process shows the validity of this method.
• Select
  Computer-aided design method of crystallization solvents for the recovery of high-purity MBT

  CHAI Shiyang, LIU Qilei, LIANG Xinyuan, ZHANG Song, GUO Yansuo, XU Chengqiu, ZHANG Lei, DU Jian, YUAN Zhihong

  Journal of Tsinghua University(Science and Technology). 2020, 60(8): 701-706. https://doi.org/10.16511/j.cnki.qhdxxb.2020.22.009

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  2-Mercapotobenzothiazole (MBT) is an important vulcanization accelerator that is widely used in the rubber industry. The solvent-based methods for the preparation of high-purity MBT need to use a suitable crystallization solvent. The traditional trial-and-error solvent selection method is time consuming and expensive. This study presents a computer-aided molecular design (CAMD) model for designing crystallization solvents. The CAMD problem is expressed as a mixed-integer non-linear programming (MINLP) model with objective functions, structural constraints, property constraints and process constraints. The objective functions are the product purity and yield. The constraints include the normal melting point, normal boiling point, flash point, solubility parameters and solid-liquid equilibrium. The activity coefficients are predicted by the conductor-like screening model based on segment activity coefficient (COSMO-SAC). The model is solved using the decomposition-based approach and 10 candidate solvents are obtained with 8 solvents having better performance than the current industrial level. Finally, the candidate solvents are experimentally verified with the results consistent with the simulation results, thus proving the validity of the model.