Most Read articles

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

In last 3 years

Please wait a minute...


For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails

Select

Massive personalized customization:New development of mass personalization XIAO Renbin Computer Integrated Manufacturing System    2023, 29 (12): 4215-4226.   DOI: 10.13196/j.cims.2023.0651 Abstract （4102）      PDF （1797KB）（11080）       Knowledge map    Save Based on the review on evolution of the production mode,a three-dimensional structure including customer,production and service dimensions was proposed to describe the production mode,and the basic connotation of each dimension was stated in a clear way.In view of the analysis of personalized requirements and personalized products,it was found that there existed asynchronous inconsistency between these two,and then the non-routine natural requirement overrun law was extracted.Furthermore,by introducing the concept of targeted service,the consistency between the two was promoted,and a new production mode,massive personalized customization was formed.Under the three-dimensional structure description of the production mode,the craft production and mass production had been degraded to a one-dimensional structure,while mass customization and mass personalization belonged to a two-dimensional structure.Two core components for implementing massive personalized customization including data-driven design for mass personalization and resilient manufacturing systems for massive personalized customization were proposed,and especially several key technologies contained by them were discussed.Targeted service was illustrated at both conceptual and technological levels,and five major production modes from craft production to massive personalized customization were analyzed from the perspective of value co-creation.The results showed that the massive personalized customization could effectively realize individual requirement through targeted service and had sufficient value co-creation capability.Examining the Haier Group′s “RenDanHeYi” mode,the analysis showed that such a mode was mass customization mode through targeted service to make it conduct the conversion of the interval value of the service dimension from small to large.Relying on targeted service,massive personalized customization could achieve overall optimization of the three dimensions of production mode,which was new development of mass personalization. Related Articles | Metrics

Select

Digital twin maturity model Computer Integrated Manufacturing System    2022, 28 (5): 1267-1281.   DOI: 10.13196/j.cims.2022.05.001 Abstract （1628）      PDF （4795KB）（1521）       Knowledge map    Save The research and application of digital twin have been entering a blowout period.In practice,there are three questions plaguing researchers,engineer and administrators.①How to tell if an application is a digital twin? ②How to tell whether an existing digital twin can meet the requirements? ③How to optimize a digital twin application to meet the requirements? Unfortunately,there is still a lack of a theoretical system to answer the above questions.To this end,the digital twin maturity model was put forward to help correctly understand and practice digital twin.Six levels of digital twin maturity were established.Then 19 digital twin maturity evaluation factors were proposed for operational digital twin maturity evaluation.The application process of digital twin maturity model was illustrated in detail with two application examples toward unit-level digital twin and system-level digital twin respectively. Related Articles | Metrics

Select

Evaluation index system for digital twin model    2021, 27 (8): 2171-2486.   DOI: 10.13196/j.cims.2021.08.001 Abstract （1177）      PDF （3882KB）（1082）       Knowledge map    Save As a critical enabling technology for realizing digital transformation,intelligence and servitization,as well as an effective method for the fusion of the digital economy and the real economy,the digital twin has received extensive attentions in various fields recently.To better support the implementation and promotion of digital twin-based applications,a systematic evaluation theory for digital twin model was required to assist decision-making process in different stages,such as modeling,Verification,Validation and Accreditation(VV&A),operation,management,reconfiguration,optimization,migration,reuse,circulation and delivery,which is still a research gap.This study extracts The evaluation criteria of the digital twin model was extracted by analyzing the specific requirements of its performance at each stage.Then a digital twin model evaluation index system was established,which could quantify the quality,performance,applicability,adaptability and value of the digital twin model,so as to assist correct and effective decision-making. Related Articles | Metrics

Select

Industrial big data-driven fault prognostics and health management Computer Integrated Manufacturing System    2022, 28 (5): 1314-1336.   DOI: 10.13196/j.cims.2022.05.005 Abstract （921）      PDF （9702KB）（908）       Knowledge map    Save With the development and application of  artificial intelligence technology,equipment has accumulated massive amount of industrial big data,which pushed the equipment Prognostics and Health Management (PHM) technology into the era of industrial big data.There had great economic and social value to extract useful information in industrial big data for PHM by combining with the function,structure and working characteristics of the equipment.The development and application of PHM technology were reviewed,and the industrial big data analysis methods were discussed.Two case studies of unity-scale wind turbines and hard disk drives in big data environments were presented to demonstrate the advantages of industrial big data-driven PHM,which could provide a reference for researchers in related fields. Related Articles | Metrics

Select

Data-driven industrial intelligence:Current status and future directions REN Lei, JIA Zidi, LAI Liyuanjun, ZHOU Longfei, ZHANG Lin, LI Bohu Computer Integrated Manufacturing System    2022, 28 (7): 1913-1939.   DOI: 10.13196/j.cims.2022.07.001 Abstract （813）      PDF （14264KB）（1128）       Knowledge map    Save The rapid development of a new generation of artificial intelligence is profoundly affecting a new round of global industrial revolution.As the core element in the digital economy era,data elements are releasing great value under the integration of intelligent manufacturing application requirements and the new generation of artificial intelligence.Data driven industrial intelligence,especially the research frontier of industrial intelligence represented by deep learning has become the focus of academic and industrial circles.The data-driven industrial intelligence was analyzed from various dimensions,especially the representative new theories and technologies based on deep learning,starting from the key links of data preprocessing,data modeling,data analysis and application in the whole life cycle of industrial data.At the same time,the typical applications of intelligent manufacturing were discussed.The challenges and future development direction of data-driven industrial intelligence research field were pointed out.This study would provide important theoretical and technical support for the development of the new cross research field of industrial intelligence based on the new generation of artificial intelligence. Related Articles | Metrics

Select

Overview of federated learning:Technology,applications and future LI Shaobo, YANG Lei, LI Chuanjiang, ZHANG Ansi, LUO Ruishi Computer Integrated Manufacturing System    2022, 28 (7): 2119-2138.   DOI: 10.13196/j.cims.2022.07.018 Abstract （804）      PDF （4840KB）（1027）       Knowledge map    Save Federated Learning (FL) is driven by multi-party data participation,and it maximizes the value of the data itself through data encryption interaction.In recent years,FL has attracted extensive attention from researchers from all walks of life and gradually moved from basic theoretical research to practical applications,which provides new technologies for further exploiting the value of data for enterprises.Based on the definition and classification of FL,a comprehensive analysis and summary of the research progress of related technologies at home and abroad was conducted,including privacy protection,communication efficiency,heterogeneity,and incentive mechanisms.The current application platforms and frameworks of FL were introduced,and the application frameworks of FL was proposed in the fields of intelligent manufacturing,medical treatment and education.Combined with the deficiencies of FL in some key open issues,its future development trends and directions were summarized for providing a reference for the theoretical research and applications of FL. Related Articles | Metrics

Select

Future equipment exploration:digital twin equipment Computer Integrated Manufacturing System    2022, 28 (1): 1-16.   DOI: 10.13196/j.cims.2022.01.001 Abstract （696）      PDF （7397KB）（892）       Knowledge map    Save Intelligent equipment plays an important role in promoting industrial upgrading and economic development.Many countries have been competing for developing intelligent equipment.However,further enhancing the intelligence of equipmenthits a bottleneck caused by the limitation of time,space,technology and cost.Inspired by the digital twin,it could be feasible to give equipment new capabilities and improve its performance by data,model,and service in the information world.Thus,the concept of digital twin equipment was first proposed from four aspects,including definition,composition,ideal capabilities and key technologies.Then,development stages of digital twin equipment were further discussed.Finally,the application of the proposed digital twin equipment was verified by logistics equipment in textile shop-floor and autoclave in composite processing shop-floor. Related Articles | Metrics

Select

Key technologies and applications for cloud CAD software Computer Integrated Manufacturing System    2022, 28 (4): 959-978.   DOI: 10.13196/j.cims.2022.04.001 Abstract （683）      PDF （6223KB）（404）       Knowledge map    Save The deep penetration and extensive applications of the advanced information technologies such as Internet and cloud computing in the manufacturing field perform new SaaS industrial software-cloud Computer Aided Design (CAD) software,which is characterized by “sharing wisdom and cloud collaboration”.Based on analyzing the concept,characteristics and application requirements of the cloud CAD software,the key technologies were expounded,which included Software as a Service (SaaS) cloud computing service framework,lightweight visualization of 3D model,3D geometric modeling and data exchange of CAD model.Through the integration of the various cloud CAD software currently in commercial or non-profit product,the application scenarios of the cloud CAD software were summarized from two aspects: online 3D modeling based on 3D printing cloud platform and cloud-based industrial product design.With the application and development requirements of the cloud CAD software,the future challenges and trends of cloud CAD software under the background of the smart manufacturing were discussed based on the combination of the cloud CAD software and 5G,digital twin. Related Articles | Metrics

Select

Digital twin-based smart gas system:Concepts,architecture and applications WANG Shanggang, CHENG Jiangfeng, GAO Shunli, YANG Shunkun, SHI Xiang, ZHANG Xiaoye, JIN Zhijun, CUI Yao, XU Ming, JIN Xiaohui, ZOU Xiaofu, TAO Fei Computer Integrated Manufacturing System    2022, 28 (8): 2302-2317.   DOI: 10.13196/j.cims.2022.08.003 Abstract （664）      PDF （11205KB）（476）       Knowledge map    Save As one of the important energy sources,gas plays an important role in production and life.The development trends in gas industry were investigated and analyzed,such as automation,precision and convenience.At the same time,the challenges such as incomplete data acquisition,incompatibility of information systems,inaccurate operations,and inconvenient user services were pointed out.Based on the investigation and analysis,a digital twin-based smart gas system was proposed,which tightly integrated with digital twin technology in key scenarios,key objects and key services of the gas supply side,the transmission and distribution side and the demand side.The architecture and common key technologies of this system were explained from the physical entity,virtual model,data center,information systems and application service.Furthermore,six types of service solutions in the system were discussed and analyzed,including pipe network design,integrity management,inspection,emergency response,transmission and distribution scheduling and user service.The application practice of the proposed system solution in some gas companies was explained in detail to provide a reference for the development and construction of smart gas. Related Articles | Metrics

Select

Temporal convolutional attention network for remaining useful life estimation LIU Li, PEI Xingzhi, LEI Xuemei Computer Integrated Manufacturing System    2022, 28 (8): 2375-2386.   DOI: 10.13196/j.cims.2022.08.009 Abstract （661）      PDF （4754KB）（383）       Knowledge map    Save Remaining useful life (RUL) prediction is of great significance for ensuring the safe operation of modern industrial equipment and reducing maintenance costs.At present,the existing RUL models based on recurrent neural networks are complex in structure and lack an effective mechanism to extract important degradation information from multi-sensor data.The new Temporal Convolutional Attention Network (TCAN) model was proposed for RUL estimation.A Temporal Convolutional Neural (TCN) network with a simple structure was used in  TCAN to extract the degradation features from the sensor data,and then the attention mechanism was used to extract the important degradation information.The learned high-level feature representation was flattened and fed into a fully connected layer to output the predicted RUL.Compared with other methods on the C-MAPSS dataset,the experimental results showed that the TCAN could more effectively improve the accuracy of remaining life prediction. Related Articles | Metrics

Select

Method of digital twin logic model oriented to production line simulation#br# Computer Integrated Manufacturing System    2022, 28 (2): 442-454.   DOI: 10.13196/j.cims.2022.02.010 Abstract （605）      PDF （8601KB）（1208）       Knowledge map    Save Production line simulation is a prerequisite to ensure the correctness,rationality and effectiveness of a plan of the production design,and the simulation of the correctness of the production line logic is the most critical.To realize the production line simulation technology based on the digital twin,a method of digital twin logic model construction oriented to the production line simulation was proposed.First of all,the composition architecture of the production line simulation system was proposed.Then,the construction method of the digital twin logical model was explained from the four dimensions of geometry,physics,production behavior,and simulation rules.The definition methods of geometric attributes and physical attributes were introduced in detail,and the production behavior was defined by finite state machine and three types of simulation rules,which include messaging rules,synchronous advance rules and mutual exclusion priority rules.The production line logic simulation prototype system was developed based on the method of model construction finally,and the effectiveness of the modeling method was verified through the simulation of the production line example. Related Articles | Metrics

Select

Named entity recognition based on deep learning Computer Integrated Manufacturing System    2022, 28 (6): 1603-1615.   DOI: 10.13196/j.cims.2022.06.001 Abstract （597）      PDF （2416KB）（395）       Knowledge map    Save To address the problems of knowledge extraction and representation difficulties caused by the coexistence of multimodal heterogeneous data in advanced manufacturing industry,the methods of Named Entity Recognition (NER) methods  become the research hot spot.The commonly used methods of NER in industry was summarized,which had introduced the traditional methods of NER firstly and then expounded the methods based on deep learning.The common methods were classified and analyzed from  three stages of distributed input representation,context encoder and tag decoder.The various methods of distributed input representation were compared,and their advantages and disadvantages were pointed out.The context encoder models from long-distance dependence capture,local context information,parallelism,information loss degree and transportability were also compared,and the characteristics of each model were given.Finally,the challenges to be addressed and research directions in the future were illustrated. Related Articles | Metrics

Select

Theory and key technologies of digital twin connection and interaction TAO Fei, MA Xin, QI Qinglin, LIU Weiran, ZHANG He, ZHANG Chenyuan Computer Integrated Manufacturing System    2023, 29 (1): 1-10.   DOI: 10.13196/j.cims.2023.01.001 Abstract （592）      PDF （2559KB）（1700）       Knowledge map    Save Digital twin,which strives to realize cyber-physical fusion,has made giant strides with its widely applications in various fields.However,a vital issue when utilizing digital twin is how to realize data connection and interaction.To tackle this problem,the connotation and criterion of digital twin connection and interaction were first introduced.Then,a theoretical system of digital twin connection and  interaction was proposed from five aspects,including perception,communication,physical-virtual mapping,data-model coupling and fusion.Furthermore,the key technologies from the above five aspects were discussed in detail.Combined with the research on physical entity,digital twin modeling,digital twin data and digital twin service,a theoretical system based on five-dimension digital twin model was enriched,expecting to provide references for the construction and application of digital twin. Related Articles | Metrics

Select

Intelligent operation and maintenance for advanced equipment based on digital twin:Challenges and future GAO Shigen, ZHOU Min, ZHENG Wei, ZHANG Linxuan, ZHANG Bin, SONG Haifeng, WU Xingtang, LI Ni, WANG Kunyu Computer Integrated Manufacturing System    2022, 28 (7): 1953-1965.   DOI: 10.13196/j.cims.2022.07.003 Abstract （585）      PDF （4039KB）（2358）       Knowledge map    Save The development of enabling technologies including big data,industrial Internet of things and artificial intelligence has promoted the deep integration of digital twins and high-end equipment operation and maintenance,which make the traditional regular-repair and failure-repair operation and maintenance mode upgrade to intelligent mode preventive-repair and state-repair,and has become a research hotspot in the field of intelligent operation and maintenance of high-end equipment.By fully using information such as mechanism models,real-time sensor data,historical data and expert knowledge and integrating modeling and simulation processes of multi-disciplinary,multi-variable,multi-level,multi-scale,multi-granularity and multi-probability,digital twin could accurately characterize data characteristics and perform efficient and accurate calculations,which achieved high-precision,high-reliability and high-credibility mapping and evolution of virtual and real space.It provided support for state assessment,fault warning and operation and maintenance decision-making of actual physical systems.The development status,key technologies and engineering applications of digital twin technology in high-end equipment intelligent operation and maintenance were reviewed,and the future challenges and difficulties were summarized. Related Articles | Metrics

Select

Research and application of manufacturing enterprise digital twin ecosystem LU Jianfeng, XIA Luyao, ZHANG Hao, XU Mengying Computer Integrated Manufacturing System    2022, 28 (8): 2273-2290.   DOI: 10.13196/j.cims.2022.08.001 Abstract （540）      PDF （11919KB）（1031）       Knowledge map    Save Many manufacturing enterprises have built digital twin factory for factory planning,simulation optimization and real-time monitoring.However,the single-domain and short-period digital twin system does not fully meet the interaction and cointegration of the physical and information world required by manufacturing companies to implement smart manufacturing.To solve this problem,the concept,population composition and characteristics of the manufacturing enterprise digital twin ecosystem were proposed from the perspective of the construction needs of manufacturing enterprise digital twin system.Combined with digital twin technology,the construction process and method of three population digital twin systems in manufacturing enterprises and the interactive configuration and dynamic evolution process among the three populations were studied.The construction and evolution of the proposed digital twin ecosystem of the manufacturing enterprise was verified by combining the intelligent upgrade case of a hydraulic cylinder factory.It could effectively improve the manufacturing flexibility and intelligence of the manufacturing enterprise and shorten the product development cycle to meet the individual needs of users for high-quality products. Related Articles | Metrics

Select

Concept,system structure and operating mode of industrial digital twin system    2021, 27 (12): 3373-3390.   DOI: 10.13196/j.cims.2021.12.001 Abstract （539）      PDF （8609KB）（840）       Knowledge map    Save The concept and technology of Digital Twin was initially formed in the manufacturing industry.With the advancement of modeling,simulation and control technology,and the development of a new generation of information technology,digital twins are rapidly used in more than 10 fields such as cities,agriculture,and construction.Digital Twin has been vigorously promoted in more than 50 directions,and this concept has now become a frontier and hot spot in these fields.Through comparing the critical issues,main application scenarios,key technologies,data sources,etc of different types of digital twin systems,the concept of industrial Digital Twin System (iDTS) was proposed with summarizing the typical characteristics of iDTS,including people-centered thinking,“human-machine-environment” mutual fusion,high fidelity of the system and complexity of twin models.An iDTS integrated these characteristics with the functional structure composed of physical layer,perception layer,twin layer,application layer and control layer.In addition,a four-stage iDTS maturity model was developed,and simulation-oriented and control-oriented iDTS were proposed.Four kinds of iDTS operating modes were analyzed,which were localized configuration mode,“cloud-end” mode,“cloud-edge-end” mode and distributed computing mode respectively,and the feasibility of the proposed iDTS operating modes were verified with typical cases for specific industrial application scenarios. Related Articles | Metrics

Select

Multi-dimensional and multi-scale modeling and edge-cloud collaborative configuration method for digital twin manufacturing cell ZHANG Chao, ZHOU Guanghui, XIAO Jiacheng, QIN Tianyu, ZHOU Yaguang Computer Integrated Manufacturing System    2023, 29 (2): 355-371.   DOI: 10.13196/j.cims.2023.02.001 Abstract （535）      PDF （15272KB）（596）       Knowledge map    Save Multi-dimensional,multi-scale and high-fidelity modeling of a discrete intelligent workshop and its software and hardware configuration and collaboration are still one of the bottleneck problems for the current research and practice of intelligent manufacturing.In response to this problem,a basic realization unit of the discrete intelligent workshop——Digital Twin Manufacturing Cell (DTMC) was taken as the object,and the constituent elements and key points of DTMC empowered by a new generation of information technologies were explored from the perspective of data-knowledge hybrid drive.The multi-dimensional and multi-scale intelligent space model of DTMC and its high-fidelity modeling method were proposed.From the perspective of edge-cloud collaboration,a software and hardware integrated configuration model of DTMC was constructed,and the smart contract-based edge-cloud collaborative operation and intelligent control mechanism of DTMC was established.A prototype system of DTMC was constructed,where an integral impeller was taken as an example to demonstrate the feasibility and effectiveness of the proposed approach. Related Articles | Metrics

Select

Scheduling optimization of multi-deep four-way shuttle warehousing system ZHAN Xiangnan, XU Liyu, LING Xufeng, CHEN Chen Computer Integrated Manufacturing System    2022, 28 (8): 2496-2507.   DOI: 10.13196/j.cims.2022.08.020 Abstract （507）      PDF （5783KB）（451）       Knowledge map    Save At present,the multi-deep four-way shuttle warehousing system has problems such as multiple four-way shuttle conflicts and deadlocks,which cause the blockage of inbound and outbound tasks,affecting the overall efficiency of the system.Hopcroft-Tarjan algorithm was used to formulate route orientation strategy for the multi-deep storage area.A scheduling optimization model with the goal of minimum operating time was established,and an Improved Hybrid Genetic Algorithm (IHGA)  was designed to solve the optimization model.The adjustment of coding and the improvement of mutation repair mechanism could effectively avoid the problem of illegal solutions in the iterative process.A multi-position neighborhood exchange method based on task sorting was proposed for increasing the diversity of the solution space effectively.The case study showed that the route orientation strategy could effectively avoid conflicts and deadlocks and improve system operation efficiency.Meanwhile,the IHGA had faster convergence speed and higher optimization efficiency,which could effectively shorten the time of inbound and outbound operations and improve the efficiency of inbound and outbound scheduling. Related Articles | Metrics

Select

QFD customer needs mining driven by product review data Computer Integrated Manufacturing System    2022, 28 (1): 184-196.   DOI: 10.13196/j.cims.2022.01.018 Abstract （493）      PDF （3836KB）（787）       Knowledge map    Save Aiming at the problem that the existing Quality Function Deployment (QFD) customer needs and their weights contained many subjective factors,which led to the poor objectivity of QFD analysis results,a mining method for QFD customer needs driven by product review data was proposed.A topic extraction based on attention latent Dirichlet allocation was proposed,and the customer needs collection was formed by Word2Vec similarity matching.The standardized customer needs expression was formed through the needs mapping model based on TRIZ.The improved proportional importance was cited and two thresholds were added to improve the single-value weight conversion rule of rough numbers,so as to obtain the final weight of customer needs.The normalized customer needs and their weights were input into the house of quality model to achieve data-driven QFD analysis.The feasibility and effectiveness of the proposed method was verified through examples. Related Articles | Metrics

Select

Multi-objective flow-shop scheduling optimization based on positive projection grey target model#br# Computer Integrated Manufacturing System    2022, 28 (4): 1087-1098.   DOI: 10.13196/j.cims.2022.04.012 Abstract （478）      PDF （2468KB）（227）       Knowledge map    Save To obtain high-quality solutions and good performance solution sets during the optimization of many-objective Permutation Flow-shop Scheduling Problems(PFSP),a positive projection grey target model with comprehensive objective weight was proposed based on grey target theory,which could overcome the shortcoming of obtaining information in the optimization process.A PFSP mathematical model with four-objective and a grey target model in the field of multi-objective optimization were defined,then target distance was calculated to judge the pros and cons of Pareto front and to extract the uncertainty information among the objective function values.To solve the different target distances of Pareto front in same section plane and to get more information in the solution space,a Positive Projection Grey Target(PPGT)model was proposed.Further,to acquire the information of the volatility and the correlation among the objective function values,a novel comprehensive objective weight method based on CRITIC method and entropy weight method was introduced into PPGT model.The modified PPGT model was integrated into the genetic algorithm to solve the multi-objective PFSP.The effectiveness of the proposed method was verified by three sets of experiments and four comparison algorithms. Related Articles | Metrics

Select

Multimodal intelligent interaction framework and realization for product conceptual design NIU Hongwei, HAO Jia, CAO Beining, LONG Hui, ZHANG Feifan, WANG Guoxin Computer Integrated Manufacturing System    2022, 28 (8): 2508-2521.   DOI: 10.13196/j.cims.2022.08.021 Abstract （471）      PDF （7591KB）（310）       Knowledge map    Save The traditional interaction mode of Computer Aided Design (CAD) based on “mouse+keyboard” is not conducive to the natural expression of design thinking.Facing the development demand of intelligent human-computer  interaction technology in product conceptual design,an intelligent interaction mode  was explored based on end-to-end model generation and manipulation,and a multimodal intelligent interaction framework for product conceptual design was constructed.According to the framework,key techniques of multimodal intelligent interaction  were studied including the multimodal signal synchronous acquisition,situation model construction and multimodal signal fusion on the level of feature and decision.On this basis,a multimodal intelligent interaction system based on brain-eye-hand was built and two typical application cases were provided to prove the feasibility of the framework and key technologies proposed in this paper.It provided ideas and technical means for the realization of a new generation of CAD intelligent interactive system. Related Articles | Metrics

Select

Workshop 3D visual monitoring system based on digital twin Computer Integrated Manufacturing System    2022, 28 (3): 758-768.   DOI: 10.13196/j.cims.2022.03.010 Abstract （469）      PDF （9680KB）（369）       Knowledge map    Save Aiming at the problems of low transparency,single mode,poor real-time performance,and lack of models in the current manufacturing workshop monitoring,a six-dimensional model of a three-dimensional visual monitoring system for workshop based on digital twin was proposed and the system development process was introduced.The model referred to the digital twin theory model and used the industrial internet of things platform as the system service platform.On this basis,key technologies such as digital data acquisition,virtual workshop construction and real-time data mapping in system development were elaborated.At the same time,according to the current situation of workshop data collection difficulties,a digital data collection method based on the industrial internet of things platform was proposed.A three-dimensional visual monitoring system for the seal production workshop was designed and developed,which verified the effectiveness of the proposed model.The proposed method provided a reference for the realization of the three-dimensional visual monitoring in the workshop. Related Articles | Metrics

Select

Unsupervised fault diagnosis method based on domain adaptive neural network and joint distributed adaptive ZHANG Zhao, LI Xinyu, GAO Liang Computer Integrated Manufacturing System    2022, 28 (8): 2365-2374.   DOI: 10.13196/j.cims.2022.08.008 Abstract （468）      PDF （2676KB）（345）       Knowledge map    Save Fault diagnosis is very important for the health management of mechanical equipment.At present,data-driven fault diagnosis methods have become a research hotspot in this field.However,the working status and conditions of mechanical equipment are constantly changing,which leads to different distributions of fault data and brings challenges to fault diagnosis.To solve this problem,an unsupervised fault diagnosis method was proposed based on domain adaptive neural network and joint distributed adaptive.The fault diagnosis data of different data distributions were preprocessed by the method of signal to image.Then,the domain adaptive neural network was used to generate features with similar data distribution,and finally the joint distribution adaptive method was used to process the generated features.The proposed method could effectively solve the problem of different data distribution caused by changes in working status and conditions.The generated model could more accurately diagnose the fault data sampled in another working state without a label.Using a classic case in Case Western Reserve University bearing data set,the method was tested and verified,and the experimental results proved the feasibility and effectiveness of the method. Related Articles | Metrics

Select

Federated learning based method for intelligent computing with privacy preserving in edge computing    2021, 27 (9): 2604-2610.   DOI: 10.13196/j.cims.2021.09.013 Abstract （463）      PDF （2487KB）（387）       Knowledge map    Save In federated learning,each terminal transmits the updated model parameters instead of the original data to the server,which becomes the key technology to guarantee data security in edge computing.On this basis,a Federated Learning Based Edge Computing (FLBEC) method was proposed to preserve the users’ privacy,while reducing the terminals’ expense for federated learning.A system framework for edge computing based on federated learning was designed and a mechanism for privacy preserving was proposed.The learning time and energy consumption for terminals were analyzed,and the study target to preserve the users’ privacy and reduce the learning time and energy consumption on the promise of guaranteeing accuracy was presented.The federated learning method was improved from the perspectives of participant selecting,local update and global aggregation.Comparative experiments were conducted to validate that there was a large amount of reduction on time and energy consumption for the majority of terminals in FLBEC by meeting the accuracy standards,which could abate the expense for federated learning and indicate the superiority of FLBEC. Related Articles | Metrics

Select

Overview on intelligent scheduling models and methods for industrial Internet-of-things LAI Liyuanjun, ZHANG Lin, REN Lei, WANG Ling Computer Integrated Manufacturing System    2022, 28 (7): 1966-1980.   DOI: 10.13196/j.cims.2022.07.004 Abstract （452）      PDF （3123KB）（775）       Knowledge map    Save The key of industrial Internet-of-Things (IoT) are the cloud-edge collaboration for full connection of industrial devices and full perception of industrial fields.The complex manufacturing processes are coupled with coordinated control of cloud resources,edge resources and industrial devices.The scheduling of computational tasks and manufacturing tasks becomes the key to determine the efficiency of the whole Industrial Internet-of-Things platform.Therefore,the studies on the scheduling of different tasks were reviewed in the environment of Industrial IoT in recent five years.Six kinds of modeling methods and Six kinds of scheduling algorithms for seven basic scheduling problems for Industrial Internet-of-Things were summarized.The main focuses of these studies,the key challenges,and possible solutions for the development of Industrial Internet-of-Things in intelligent manufacturing were discussed. Related Articles | Metrics

Select

Research and application of augmented reality in complex product assembly Computer Integrated Manufacturing System    2022, 28 (3): 649-662.   DOI: 10.13196/j.cims.2022.03.001 Abstract （447）      PDF （6271KB）（502）       Knowledge map    Save The rapid development of augmented reality has attracted much attention and research in the field of assembly.To study the application of augmented reality in complex product assembly,a complex product intelligent assembly system framework based on AR was proposed.By analyzing the relevant research and application of augmented reality technology,the key technologies related to the augmented reality assembly system were expounded,such as tracking registration,human-machine interaction,occlusion handling and content creation.The application of augmented reality technology in complex product assembly was mainly divided into three parts: assembly guidance,assembly training,assembly simulation and planning.The application pattern and technical characteristics of these three aspects were summarized.Based on the application and development requirements of complex product assembly technology,the development and challenges of augmented reality technology in assembly field were discussed. Related Articles | Metrics

Select

Matching method for distributed photovoltaic maintenance scheme based on knowledge graph    2021, 27 (7): 1860-1870.   DOI: 10.13196/j.cims.2021.07.002 Abstract （440）      PDF （4391KB）（445）       Knowledge map    Save To improve the efficiency of distributed photovoltaic maintenance industry,a matching method for maintenance scheme based on professional domain knowledge graph was designed.The maintenance scheme based on maintenance requirements and the distributed photovoltaic maintenance knowledge graph was matched accurately,which provided scheme support for maintenance personnel.A method of matching maintenance scheme with the knowledge graph as a bridge was presented.The matching process was improved according to the complex type of maintenance requirements.In view of incomplete requirements in practice,the method was enabled to match the scheme by inference with the knowledge graph.Experiments showed that the improved approaches helped to provide a higher degree of accuracy and better reliability. Related Articles | Metrics

Select

Review on security,privacy,and performance issues of blockchain    2021, 27 (7): 2078-2094.   DOI: 10.13196/j.cims.2021.07.022 Abstract （431）      PDF （2251KB）（314）       Knowledge map    Save As a cornerstone to build a low-cost trusted channel so as to achieve value interconnection,blockchain has been especially highlighted and intensive research has been conducted in recent years.Among other issues within blockchain,challenges on security,privacy,and performance have attracted most eyes.To pave the way to deal with the above challenges,a review of research particularly on blockchain security,privacy and performance issues was presented from the perspectives of technical principle,progress and mechanism.With understanding of blockchain hierarchy and operation principles,the security issues within the layers of peer-to-peer network,consensus mechanism and smart contract were analyzed,and the effectiveness of respective defence measures was evaluated.By analysing the threats to blockchain privacy,the various privacy-protection solutions were evaluated for different protection targets.By identifying the performance related factors,the state-of-the-art solutions on the on-chain and off-chain scaling techniques were discussed.To tackle the existing problems in security,privacy and performance,the possible solutions and directions of future research were proposed respectively. Related Articles | Metrics

Select

Surface defect detection algorithm of magnetic tile based on Mask R-CNN Computer Integrated Manufacturing System    2022, 28 (5): 1393-1400.   DOI: 10.13196/j.cims.2022.05.011 Abstract （426）      PDF （2809KB）（189）       Knowledge map    Save The magnetic tile image has the characteristics of uneven illumination,.complex surface texture and low contrast,but its defects are hard segmented with the traditional defect detection algorithms.For this reason,a defect detection algorithm based on Mask Region-based Convolutional Network (Mask R-CNN) was proposed.The image was preprocessed by contrast limited adaptive histogram equalization.Then Residual Network 50 (ResNet50) was used to construct the Feature Pyramid Network (FPN) to acquire image information and extract features.Then,the region of interest of the defect region were extracted by Region Proposal Network (RPN) to obtain the corresponding anchor frame,and the pixel class inside the region of interest was predicted by Fully Convolutional Network (FCN) to realize the defect.Through the fully connected layer of the network,the prediction the category and the corresponding anchor frames of each interest region was realized.The experimental results showed that the proposed algorithm had strong generalization ability,and strong robustness.It could accurately segment the surface of the magnetic tile image with complex texture,uneven illumination and low contrast. Related Articles | Metrics

Select

Digital twin-based product design process and design effort prediction method Computer Integrated Manufacturing System    2022, 28 (1): 17-30.   DOI: 10.13196/j.cims.2022.01.002 Abstract （422）      PDF （5902KB）（610）       Knowledge map    Save To mitigate the complexity during product design caused by inconformity between reality and ideal design process,a digital twin-based product design process and the design effort prediction method was proposed.A product design digital twin system in five dimensions was constructed.The product design digital twin in virtual space was specifically defined,which included functional digital twin,designer digital twin and design activity digital twin.From a knowledge perspective,the prediction approach for product design effort and the propagation analysis approach for function changes were proposed,which was able to provide lean service for the management of complexity during product design.A case study of system design of a cube satellite-based radio aurora explorer mission showed that the proposed method could measure,predict and manage the complexity during product design effectively. Related Articles | Metrics

Select

Visual analysis of digital twin development based on scientific knowledge graph Computer Integrated Manufacturing System    2022, 28 (6): 1673-1684.   DOI: 10.13196/j.cims.2022.06.007 Abstract （418）      PDF （3758KB）（362）       Knowledge map    Save With the introduction of technologies and development strategies such as industry 4.0 and intelligent manufacturing,the research and application exploration in the field of digital twin has gradually become a research hotspot in recent years.Through the quantitative and visual bibliometric analysis method,the research results of the digital twins field was combed in WoS database,and the external characteristics of the literature in the digital twins field was analyzed by using the scientific metrology theory algorithm and visual analysis software.The research hotspots and hot spots evolution trend was discussed in the field,and the future research direction was put forward,which provided reference for the field scholars and enterprise practitioners who had pay attention to the field of digital twin provide. Related Articles | Metrics

Select

Improved whale algorithm for solving engineering design optimization problems    2021, 27 (7): 1884-1897.   DOI: 10.13196/j.cims.2021.07.004 Abstract （416）      PDF （2656KB）（253）       Knowledge map    Save To better solve the engineering design optimization problems and improve the optimization performance and application ability of the whale optimization algorithm,the whale optimization algorithm based on piecewise random inertia weight and optimal feedback mechanism was proposed.For the random walk foraging strategy,a feedback mechanism based on the current global optimal solution was introduced to speed up the algorithm's convergence speed and enhance the stability of the solution.The piecewise random inertia weight was introduced into the shrinkage encirclement strategy and the spiral bubble net predation strategy,which improved the optimization accuracy and enhanced the ability of algorithm to jump out of the local extremum.The Cross-border processing was modified and improved to eliminate the potential loss of evolution results.Theoretical analysis proved that the improved algorithm had the same time complexity as the basic whale optimization algorithm.The experimental results of 6 representative comparison algorithms on 12 complex benchmark test functions and 3 engineering optimization design problems showed that the proposed algorithm had significantly better optimization performance,solution stability,applicability and effectiveness to different problems by comparing with 5 other comparison algorithms. Related Articles | Metrics

Select

Road to restart domestic CAD software    2021, 27 (11): 3057-3075.   DOI: 10.13196/j.cims.2021.11.002 Abstract （407）      PDF （5249KB）（335）       Knowledge map    Save The developing and application history of domestic CAD software was reviewed,and the advances in fundamental research of Computer Aided Design (CAD) and computer graphics was demonstrated in large software development and application in China.Taking Shanghai municipal CAD application engineering software product BYL CAD as an example,the theoretical foundation,technical foundation,algorithm foundation,system design foundation and customization foundation in China were sorted out.The possibility of restarting domestic CAD software was discussed,the restart strategies were proposed,and some restart plans were given. Related Articles | Metrics

Select

Multi-index quantitative evaluation model of gunner's operation posture based on human-machine simulation technology    2021, 27 (8): 2350-2361.   DOI: 10.13196/j.cims.2021.08.017 Abstract （406）      PDF （5851KB）（177）       Knowledge map    Save To solve the quantitative evaluation problem of gunner operating posture in the process of artillery aiming,loading,repairing and march-combat conversion,a comprehensive evaluation model was constructed from the top-down at the application layer,performance layer,logic layer,data layer and knowledge layer based on the multi-index dimensions,which was relied upon the human-machine simulation technology.In the right of the evaluation data analysis part,the gunner position assessment data analysis system named DASGOPE 1.0 was developed by combining with Grey Relation Analysis(GRA),Technique for Order Preference by Similarity to an Ideal Solution(TOPSIS),grey relation closeness and grey relational clustering.The operation evaluation of handwheel of a certain type of towed howitzer aiming mechanism was taken as an example to discussed the application of the proposed model.The digital model of gunner's body and design scheme for handwheel were constructed.The dynamic behavior process of gunners was simulated,and the operating posture was evaluated by eight indexes of wrists,elbow,trunk,hip,knee and ankle.The evaluation data of DASGOPE 1.0 was calculated and processed,and the research showed that the evaluation model was available.Aiming at the groups of artillery human-machine control equipment design scheme,the optimal or suboptimal scheme was quickly identified and extracted by evaluating the rationality of the corresponding behavior posture,which could provide guidance for the subsequent control equipment optimization. Related Articles | Metrics

Select

Survey on genetic algorithms for solving flexible job-shop scheduling problem Computer Integrated Manufacturing System    2022, 28 (2): 536-551.   DOI: 10.13196/j.cims.2022.02.018 Abstract （403）      PDF （5643KB）（297）       Knowledge map    Save Flexible Job-Shop Scheduling Problem (FJSP) is an important scheduling problem with extensive applications.As one of the most popular methods for solving FJSP,Genetic algorithms (GAs) have attracted significant attentions of a number of researchers.A survey of recent works on GAs for solving FJSP was given,especially five main chromosome representations and relevant crossover and mutation operators in GAs.Then seven evaluation criteria including encoding feasibility,mapping relation,memory space,decoding complexity,encoding completeness,the complexity of genetic operation and the diversity of genetic operation were proposed to evaluate the five chromosome representations. Related Articles | Metrics

Select

Evolution and State-of-the-art of intelligent manufacturing from HCPS perspective    2021, 27 (10): 2749-2761.   DOI: 10.13196/j.cims.2021.10.001 Abstract （392）      PDF （1873KB）（424）       Knowledge map    Save Human-Cyber-Physical System (HCPS) can provide the theoretical support to understand and develop Intelligent Manufacturing (IM) according to the concepts of human-centric development and integration of industrialization and information technology.To better understand the relationship between HCPS and IM as well as grasp IM development trends,HCPS was analyzed in detail including the definition,connotation and system elements,followed with discussions on IM evolution and characteristics of New-Generation Intelligent Manufacturing (NGIM).The state-of-the-art of NGIM was reviewed from HCPS perspective.Relevant suggestions for China's intelligent manufacturing development were proposed covering perspectives from human,cyber systems,physical systems and system integration,which was expected to provide a reference for understanding and adoption of intelligent manufacturing and HCPS. Related Articles | Metrics

Select

Visual real-time monitoring method for discrete manufacturing workshop based on digital twin    2021, 27 (6): 1605-1616.   DOI: 10.13196/j.cims.2021.06.007 Abstract （389）      PDF （1876KB）（387）       Knowledge map    Save Focusing on the problems of real-time monitoring,poor control ability and opaque management in discrete manufacturing workshop,an object-oriented method was proposed to realize a real-time monitoring method for discrete manufacturing workshops based on Digital Twin (DT).DT-based visual real-time monitoring method architecture for discrete manufacturing workshop was built,and its key implementation process was clarified.Four key technologies to realize real-time monitoring based on DT were described in detail,including data modeling and transmission method based on AutomationML and OPC UA,event-driven virtual and real mapping method,workshop logic modeling method based on complex event processing,information visualization and push.Taking an aerospace product machining workshop as an application case,the effectiveness of the monitoring method was verified by combining the actual production process and the developed prototype system. Related Articles | Metrics

Select

Deep reinforcement learning method for biped robot gait control    2021, 27 (8): 2341-2349.   DOI: 10.13196/j.cims.2021.08.016 Abstract （383）      PDF （3448KB）（194）       Knowledge map    Save Aiming at the stable control of gait during biped robot walking,a deep reinforcement learning method with improved Deep Q-Network (DQN) was proposed.By combining DQN algorithm with a deterministic strategy gradient,an improved DQN learning network was proposed to replace the critic network of actor-critic network with a clipped Double-Q network.A link model of biped robot was established,and the proposed network was used for biped robots gait control training as agents in a conventional flat road environment.MATLAB simulation results showed that compared with DQN and Deep Deterministic Policy Gradient (DDPG) algorithms,the proposed algorithm had a better training speed and its average reward curve had a good smoothness.Under the CPU training conditions,the agent training could be completed after about 20 hours of deep reinforcement learning.The biped robot could achieve stable and fast walking (average speed about 0.5m/s) under the conditions of small torque and long distance (about 5 meters). Related Articles | Metrics

Select

Digital twin high-fidelity modeling method for spinning forming of aerospace thin-walled parts Computer Integrated Manufacturing System    2022, 28 (5): 1282-1292.   DOI: 10.13196/j.cims.2022.05.002 Abstract （381）      PDF （6301KB）（1466）       Knowledge map    Save To solve the difficulties in the integration and interaction of information space and physical space caused by factors such as multi-discipline,multi-physical quantity,multi-scale and dynamic time-varying in the manufacturing process,a digital twin high-fidelity modeling method was proposed by taking the spinning forming process of aerospace thin-walled parts as the object.The spinning digital twin high-fidelity model was described and defined from three levels of geometry,mechanism and data.Based on the constructed digital twin model,the knowledge acquisition process of spinning processing was studied.The data mapping and comparison process of the spinning process was studied.In addition,to evaluate the validity of the built model,a fidelity evaluation method of the digital twin model was proposed.A case verification showed that the proposed digital twin high-fidelity modeling method was effective and the constructed model could be evaluated. Related Articles | Metrics

Select

Digital twin-driven intelligent assembly method for high precision products Computer Integrated Manufacturing System    2022, 28 (6): 1704-1716.   DOI: 10.13196/j.cims.2022.06.010 Abstract （380）      PDF （8244KB）（332）       Knowledge map    Save To solve the problems of low assembly efficiency and poor consistency of assembly quality caused by disconnection between virtual simulation analysis and physical assembly in assembly process of high precision products,an intelligent assembly method of high precision product driven by digital twin was proposed.The digital twins of high-precision products including all assembly elements were constructed.Aiming at the problems of poor readability of assembly process documents and weak knowledge correlation,an assembly process expression method based on knowledge map was proposed.At the same time,the dynamic adjustment of assembly process was realized by using the inferential property of knowledge map.Aiming at the problem of product quality control,a three-layer structure quality control strategy of operation-state-quality was proposed.Taking the assembly of a certain type of automobile engine block unit as an example,the practicability of the proposed method was verified. Related Articles | Metrics