- Manufacturing Process and Optimization
- Design Education and Practice
- Product Development and Customization
- Industrial Technology and Control Systems
- Probabilistic and Robust Engineering Design
- Technology Assessment and Management
- Semantic Web and Ontologies
- Digital Transformation in Industry
- Advanced Multi-Objective Optimization Algorithms
- Systems Engineering Methodologies and Applications
- Reliability and Maintenance Optimization
- Natural Language Processing Techniques
- IoT and Edge/Fog Computing
- BIM and Construction Integration
- Topic Modeling
- Service-Oriented Architecture and Web Services
- Data Quality and Management
- Collaboration in agile enterprises
- Simulation Techniques and Applications
- Fatigue and fracture mechanics
- Fault Detection and Control Systems
- Multimodal Machine Learning Applications
- Intellectual Property and Patents
- Industrial Vision Systems and Defect Detection
- Business Process Modeling and Analysis
Beihang University
2016-2025
Panzhihua University
2024
University of Chinese Academy of Sciences
2021-2023
Institute of Computing Technology
2021-2023
Chinese Academy of Sciences
2021-2023
China Resources (China)
2023
Foton Motors (China)
2021
University of Michigan
2018
University of Connecticut
2018
The Ohio State University
2018
Online reviews serve as significant channels for users to express their preferences, constituting an essential data source enterprises identify product requirements. However, with the widespread adoption of smartphones, act capturing spontaneous photographs has become a habitual practice majority, resulting in increasing prevalence supplementary visual expressions within online reviews. Therefore, important research question emerges: How can requirements be effectively extracted from...
In view of the low efficiency production line modeling method and poor quality model, this paper proposes concept digital twin by analyzing simulation & twin, analyzes composition line. Then real-time is proposed. Finally, effectiveness proposed verified taking a product assembly as an example.
Since innovation in complex product design hinges on thorough engineering knowledge application, high-quality patent recommendations foster design. However, many recommendation studies perform analysis without comprehensive exploration and proper organisation of knowledge, causing a superficial understanding patents returning arbitrary results. To mitigate this issue, deep learning-based approach for representation learning is proposed. First, four-dimensional model defined to formalise the...
Multidisciplinary design optimisation (MDO) involves multiple disciplines, coupled relationships and processes, which is implemented by different specialists dispersed geographically on heterogeneous platforms with analysis tools. The product data integration sharing among the participants hampers development applications of MDO in enterprises seriously. Therefore, a multi-hierarchical integrated model (MH-iPDM) supporting web environment services-based multidisciplinary (Web-MDO) framework...
In order to improve the resource utilization efficiency of heterogeneous multi-robots, minimize execution time multi-type tasks, effectively maintain load balancing robot resources, solve problem multiple resources and difficult find a near-optimal solution for multi-robot collaborative planning, task allocation strategy combining improved particle swarm optimization greedy (IPSO-G) algorithm is proposed. The divided into two steps: First, used search combination tasks robots; after that,...
The risk of failure for aerospace missions can be reduced if architects model the product in a systematic way and make decisions physical implementation based on stakeholder needs. Satellite communication system architecting should take one its dominating elements, satellite, into account consistent modeling during whole lifecycle. Model-based engineering (MBSE) serves as useful tool activities connections with manufacturing. In this paper, satellite is investigated preliminary design stage...
The knowledge distillation-based approaches have recently yielded state-of-the-art (SOTA) results for cross-lingual NER tasks in zero-shot scenarios. These typically employ a teacher network trained with the labelled source (rich-resource) language to infer pseudo-soft labels unlabelled target (zero-shot) language, and force student approximate these pseudo achieve transfer. However, previous works rarely discussed issue of pseudo-label noise caused by source-target gap, which can mislead...
A response surface method based on the all sample point interpolation approach (ASPIA) is proposed to improve efficiency of reliability computation. ASPIA obtains new points through linear interpolation. These are occasionally extremely dense, thus easily generating an ill-conditioned problem for approximation functions. mobile most probable failure strategy used solve this problem. The advantage proven by two numerical examples. With ASPIA, approximated process can rapidly actual limit...
Automated guided vehicle (AGV) is an important transportation equipment, which widely used in warehouses and factories. In the scenarios of multi-AGVs application, efficient AGVs task assignment strategy beneficial for costs, balance workload increasing distribution efficiency. The traditional method usually depends on a powerful scheduling system, solves problem regular way. this paper, we present decentralized framework multi-task allocation with attention (MTAA) deep reinforcement...
For named entity recognition (NER) in zero-resource languages, utilizing knowledge distillation methods to transfer language-independent from the rich-resource source languages is an effective means. Typically, these approaches adopt a teacher-student architecture, where teacher network trained language, and student seeks learn expected perform well target language. Despite impressive performance achieved by methods, we argue that they have two limitations. Firstly, fails effectively shared...
Abstract Design rationale (DR) explains why an artifact is designed the way it is. An explicit representation of DR helpful to designers, allowing them understand, improve, and reuse previous designs. The argumentation-based mainstream approach representation. It has a semiformal graphical format depict structure arguments for solving design problem. This paper argues that because not just problem-solving process but also cognitive activity continuously iterative evolving, conventional some...
Model-based approach with reusable mechanisms can serve as an effective way for complex system architecture design.Stakeholder needs should be satisfied while product and design need to consistent user requirements in all stages during the whole lifecycle.In this paper, satellite network example of is modeled a reusable, reconfigurable efficient manner using modeling language (SysML) together pattern viewpoints simulation constructs.Based upon abstract syntax described metamodels set...
This paper studies application architecture of aircraft large components assembly based on Internet Things (IO) for the characteristics with low degree automation and difficulty to inherit trace equipment information in process assembly. By studying process, this presents a method classifying obtaining IOT solution integration management. Finally, development characteristic industry, puts forward overall model framework provide foundation automatic joining.
A large number of publicly available documents, including patent documents and journal articles, can provide designers with creative stimuli, which could facilitate product innovation collaborative design. As an important tacit knowledge, the acquisition, sharing, reuse design rationale (DR) is great value to designers, help better understand intentions ideas, support automation, promote However, due fragmentation DR in documentation, this hinders designer acquisition reuse. If fragments be...
The demand for functionality, performance, safety, and reliability in complex product domains is continuously increasing, leading to growing complexity. Traditional document-based systems engineering approaches face numerous challenges terms of communication efficiency, traceability, maintenance when dealing with system design. Consequently, Model-Based Systems Engineering (MBSE) has been widely adopted. To improve the efficiency modelling, an effective method needed automate process by...