Peijun Zhang

Full transcript

Hello, we represent the Beyond the Blueprint project. I'm Evan. I'm Harry. It's a pleasure to have Mr.

Zhang from Germany this time to have an interview. So we're going to ask him several questions about his engineering path. So to start with, what inspired you to become an engineer specifically in scientific machine learning and other fields? Okay, thank you for your question.

I have always been interested in understanding how physics systems work and how mathematic models can describe them. And during my research, I realized that the traditional methods and the machine learning framework can be effectively combined. And I would also like to give special thanks to my doctor supervisor, and he inspired me to research in this space. Yeah.

My other question is, why did you choose to study in Germany as a career path? And I think this is because the German has a very exciting reputation in engineering, especially in engineering mechanics and the industry technology. And I also want to experience a quite different culture, academic and working environment. Studying in Germany gave me the opportunity to develop both technology and personality.

At the same time, I was able to experience a quite different culture, lifestyle and cuisine. Yeah, I think that's very interesting because, you know, I was studying in America, but I'm from China, so the kind of cultural, like, integration and stuff, it's very, like, interesting. So, like, my last question for this topic is that, is there any childhood experiences that shaped your goal to become an engineer? Actually, there is not a single event that made me want to become an engineer.

Actually, when I was a child, I wanted to become a physicist, astrophysicist, and study the stars and the universe. And I think this early curiosity about the nature and the physical phenomena later developed into my interest in engineering and mechanics. From your research in University of Siegen to your experience at the Constructions Gruppe Bauenage to your studying at the University Hanover, what choices have you made on your path and what do you feel about them? During my studies at the Leibniz University Hanover, I choose to focus on solid mechanics and numeric methods, which is an important foundation of civil engineering.

And after completing my master's degree, I decided to enter industry. I wanted to improve my practical engineering skills and understand how real engineering projects such as bridges are designed and constructed. And I also wanted to improve my communication and teamwork skills. I enjoyed working with reliable colleges and solving practical problems together.

And after that, I decided to strengthen my theoretical and research background. So I came to the University of Siegen to pursue a PhD. Research is very different from industry work. It is highly interesting and also very challenging.

And that is what makes it exciting for me. Oh, OK. Thank you. That's really cool.

So why did you make your choice of studying abroad in Germany specifically? And do you think it is a good choice? Yes, I think it was a very good choice. It was some challenging because of the languages and also cultural differences and also academic pressure.

But it made me much more independent. When you live aboard, you have to take care of many things by yourself, such as cooking moving to a new apartment and also arranging a lot of contracts These experiences taught me how to manage my life independently and solve practical problems on my own Oh, that's very cool. Fully understandable. Yeah, as a boarding student from Texas, getting into a new environment and getting used to all the stuff, I feel very compassionate about it.

So talking about challenges, what are some very serious challenges you faced both during work or during the period of studying for your engineering PhD? You mean a technical day or? some challenges, like more like serious ones. Yes.

And one major challenge, I think the major challenge is that the real engineering problem are much more complicated than the idealized models described in the textbooks. Real projects may involve limited data on certain material properties and complicated geometries and also many interacting physical effects. The engineer must simplify the problem appropriately without losing the most important physical features. And also I can share a case that I faced in my career.

So during my work, we have to reassess the very old bridge I like, which was built 70 years ago. And we want to modeling them upon the technical drawing. But in fact, the technical drawing was, a lot of them are missing. So we have to deal with such problem.

Yes, so this is also challenging in the real life or in the real project. So talking about everyday life, so my life as a student is pretty full of learning and stuff. So what does a typical day of an engineer look like during work? During my PhD, a topic they include like reading papers or responsibility research papers and developing mathematical models, writing code and running the simulations and then analyzing the results.

I also discuss the research problems with colleges and also attend the conference meetings and prepare the publications and also supervise students and also teaching the classes of class. And as an engineering company, the work is usually more project-oriented. Engineers need to understand the client's problem, build a suitable model, evaluate the results, and provide more practical recommendations. Yes, so the work as a PhD and in an engineering company is quite different.

Yeah, it sounds like it takes a lot of effort each day to finish your work every day. Yes. So in your field of work, what skills and tools do you usually use? Yes, one of the most important skills, I think, is analytical thinking.

This means having the theoretical knowledge needed to derive mathematical models and understand the physical problem. And also programming is also essential because it allows you to turn an idea or mathematical model into a working computational method. Another important skill is, I think, the ability to analyze and interpret the results. And also producing numerical methods or producing numerical results is not enough.

An engineer must also understand whether the results are meaningful or reliable. The main tools I use for such programs or calculations are Python, MATLAB, and also the LavaCruise. Oh, okay. That's very interesting.

I saw that you have a field of work on crack analysis. So what skills and methodology did you use in this field of 2D crack analysis? Yes my work mainly focuses on the computational crack analysis Although this is not an experimental graph detection I use the fractured mechanics mathematical modeling and the finite element methods boundary element methods So for me it more like a computational simulation. And also I use the programming, the scientific machine learning.

And I also validate my results by comparing them with reference solutions. Okay, so you use mostly machine learning, right? Yes, it's a framework. So do you know of any sensors that could be used in crack analysis, like sensors that could detect materials?

To be honest, the sensors are not my main area of expertise. My research focuses more on numerical simulation and computational fracture mechanics, but I have heard a lot of crack detection methods, such as the ultrasonic testing. And I think it's a very important field for the structural health monitoring. Oh, have you heard of any current inspection technology of crack analysis?

Yes, I heard of it, but I have no direct application or direct knowledge of this. And I think it's also a very efficient crack detecting technology, right? Great. So you mainly use coding tools and other AI-related tools to assist your work.

Do I use the coding tools and also the AI assistant coding tools? Yes, the answer is yes. But at the beginning of my research, I mainly wrote code by myself because it helped me to develop a strong programming skills and also most important, a programming logic. But nowadays, the AI assistant program, especially the large languages model, can help us to more efficiently programming.

So I use the AI assistant program. But first off, you have to learn programming yourself. Yes. Okay, so talking about AI, do you think the future of engineering will be replaced by artificial intelligence?

Actually, I don't think engineers will be completely replaced by AI. AI will automate some repetitive tasks such as coding, data processing, other preparation, and the report generation. However, engineering still requires physical understanding and professional responsibility judgments judgment, creativity, and communications. In my opinion, engineers who know how to use AI effectively may replace engineers who do not use it.

So in your field of work, do you use AI-related tools or other tools like this artificial intelligence to assist your work? Yeah, I think also it is a global trend. In the beginning of my research, I mainly wrote my or writing wrote code by myself because it or this helped me develop a very strong programming foundation and a very good coding logic now ai related tools especially the large languages models like ai agents can significantly accelerate the coding process they are very useful for generating code or debugging organizing ideas and also improving efficiency yeah so i use them okay so now ai related tools are definitely very interesting and they are also very boost your in boost your efficiency and work a lot and i also wanted to ask you like in your field of work do you get in touch with a lot of like tools and sensors related to like detecting specific laws in like metal and other materials research field is not my research but i i heard a lot of methods that can detecting the floors or cracks in materials like ultrasonic methods which my college did and it can use ultrasonic tools to detect the flaws in a material But yes, and I think this is a very good topic in structural health monitoring. But yes, I don't have any concerns about it.

Have you ever heard of the eddy current sensor then? It's like electromagnetic sensors. I heard of them, but I don't know how it works. But I think it's very interesting.

It's a very good comparison, I feel like. So what are your thoughts on what you will try to do in the future? So what will you do in the future? Oh, it's a good question.

And I think in the future, I would like to continue working at this intersection of engineering simulation and artificial intelligence. My goal or my purpose is to develop a reliable AI-assisted method or a framework for computational mechanics and industry simulation. And also, I want to connect academic research with real engineering applications so that the advanced numeric methods and machine learning technologies or techniques can be used to solve practical industry problems. So if you wanted to give some advice for beginners.

So like what's some special traits and things you've noticed among the students you taught that were like especially made them successful? Yeah, that's a good question. I think the successful students usually prepared themselves before a discussion. And I mean, yes, they prepared them very good, very well.

And during their preparation, they had already thought deeply about the problem and also identified the main difficulty and clarified what they want to discuss. And as a result, the discussion was focused and also very efficient rather than spending too much time trying to define new problem. Yes, preparation is definitely very important. Yes.

So like if you meet a new student that who is trying to get in your field of work, what is some advice you would recommend to them? And I would recommend building a very strong foundations in mathematics, mechanics and also programming. and do not learn or do not only learn how to use engineering software. Try to understand the equations and physical assumptions and the numeric methods behind it.

And also always question your results. Oh, that's very good advice. So I think that's all the questions I have to ask. I wanted to ask you today.

So thank you, Mr. Zhang.