Biography
I gained my bachelor degree in mechanical engineering in 2010 and my first M.Sc. degree in mechanical engineering with specialization in logistics, in Tongji-University in 2013. I worked as junior project manager in pilot hall at SAIC Volkswagen between 2013 and 2015. Since 2015, I have been studying technology management in University of Stuttgart and gained my second M.Sc. degree in 2019. From 2019/11 to 2022/04, I worked as academic staff in the IAT of University of Stuttgart at Digital Engineering team of Fraunhofer IAO.
Research
Research interests are Modeling and Simulation, Hybrid Twin/Digital Twin, Numerical Solvers, Model-based Systems Engineering (MBSE), AI for Science, etc.
- ModelSolver (2022)
ModelSolver is a personal project due to personal interests. It can model multi-domain system from bond graph and solves system of ODEs derived from the graph. The main goal is to explore the meta-model approach for system-level simulation towards MBSE approach.
Both modeling, bond graph processing and numerical solvers (e.g. RK-Fehlberg order4/order5) are developed with Python as playground for the purpose of realizing Proof-of-Concept.
Further development: C++ or Python/C++ mixed; Investigate Modelica components; UI and widgets for modeling; transformation from UI-components in bond graph; semantic check; parsing to state matrix;
- ODEqSolver (2022)
ODEqSolver is a personal project due to personal interests. It solves system of ODEs described in a list of equation strings in Python. The main goal is to gain a insight in the backbone of simulation technology. Deepdived are explicit and implicit methods, rootfinding and linear system. Crude demos are realized in Python and C++.
Further development: find general expression for ivp and bvp; application in established models, methods of determining stiffness (matrix rank or?) PDE solver, via compiler
- QU4LITY Project Airbus Pilot (2020-2022)
Qu4lity (EU H2020 825030) is the biggest European project dedicated to Autonomous Qu4lity (AQ) and Zero Defect Manufacturing (ZDM) in the Industry 4.0, co-funded by the Horizon 2020.
This is a public funded project. The pilot is based on a case from Airbus. Research concept is a semantic integration in industrial system design towards MBSE approach. The aim of the reasearch is to integrate different engineering activities, processes, methods, tools in system design so that people from different domains/departments can colaborate efficiently and cross-disciplinary decision making are better supported (pic. below trade dashboard).
The DES tool is developed with Python based on an open source discrete event simulation engine SimPy. DES models are object-oriented designed. Models are instantiated according to design concept models in meta-model, then processed and simulated, aiming at a automatic generation of design concepts trade-off to support decision (pic. below data flow overview).
Video Link: Youtube - Industrial Co-Design at Airbus - Results from EU project QU4LITY
- QU4LITY Project Continental Pilot (2021)
This is a public funded project. The pilot is based on a case from Continental. A catalogue that records the failure types from SMT manufacturing line has different languages for world-wide factories. Misplacement or errors could occur by confusion of unexperienced people or misunderstanding due to cross-lingual semantic difference. A small PoC of semantic search aims to verify the feasibility of using NLP technology to tackle the situation.
The processing pipeline is developed with Python and open source tools: language detection by SpaCy, translation and similarity representation by by MarianMT and RoBERTa supported by transformers from HuggingFace. Results were enlightening.
- AI for Document Analysis (2021)
This is a industry project. There are experiments for document classification using different open source tools and methods to deal with txt formats, including Doc2Vec, FastText, TextCNN, BERT, RoBERTa, etc. Besides, OpenCV is used to deal with the image of standard format document to determine the document class. BM25 search algorithm is used to find relevant information in documents.
Pubulications
Zheng, X., Hu, X., Arista, R., Lu, J., Sorvari, J., Lentes, J., & Ubis, F. (2022). "A semantic-driven tradespace framework to accelerate aircraft manufacturing system design." Preprint
Hu, X., Arista, R., Lentes, J., Sorvari, J., Lu, Jinzhi, Zheng, X., Sorvari, J., Ubis, F. & Kiritsis, D. (2022). "Ontology-centric industrial requirements validation for aircraft assembly system design." IFAC-PapersOnLine in press
Hu, X., Lu, J., Zheng, X., Arista, R., Sorvari, J., Lentes, J., Ubis, F. & Kiritsis, D. (2022). "Semantic Modeling Supporting Discrete Event Simulation for Aircraft Assembly Process." International Symposium on Knowledge and Systems Sciences (pp. 83-98). Springer, Singapore.
Hu, X., Arista, R., Zheng, X., Lentes, J., Sorvari, J., Lu, J., Ubis, F. & Kiritsis, D. (2022). "Ontology-based system to support industrial system design for aircraft assembly." IFAC-PapersOnLine, 55(2), 175-180.
Zheng, X., Lu, J., Arista, R., Hu, X., Lentes, J., Ubis, F., Sorvari, J. & Kiritsis, D. (2020). Development of an application ontology for knowledge management to support aircraft assembly system design. CEUR Workshop Proceeding ISSN, 1613, 0073.
Fang, D. & Hu, X. (2014). Architecture of a Knowledge-Based Education System for Logistics. In Frontier and Future Development of Information Technology in Medicine and Education (pp. 2683-2692). Springer, Dordrecht.
Miscellaneous
Master thesis: Development of a method for company-specific advancement of digital engineering based on AI-functionalities at the example of a voice-controlled 3D-CAD application
Master student thesis: Study of the influences of Industry 4.0 on product development