Master Thesis: »Deep learning for defect detection in battery cells«
At the Fraunhofer IPT in Aachen, we work with more than 530 employees every day to make the production of the future more digital, more flexible, and more sustainable.
In the department Production Quality we apply digital technologies to optimize production processes by using artificial intelligence to make production more sustainable.
One focus of our work is on optimizing the production processes for lithium-ion battery cells and fuel cells.
Within the scope of your thesis, you will investigate deep learning-based defect detection approaches. To this end, we use roll-to-roll processes for the efficient coating of electrodes.
However, defects can occur during this process step, causing high reject rates. To solve this problem, we are developing a modern deep learning-based defect detection system.
A central step for the application of this system is the reduction of the annotation effort by process experts. Transfer learning approaches for deep learning models are a promising way to reduce this effort.
Therefore, within the scope of this master's thesis, different deep-learning approaches will be implemented and evaluated.
What you will do
- Selection of suitable deep-learning approaches in the field of transfer
- Training of deep learning models for defect detection and analysis of the results
- Implementation of suitable deep learning models in the defect detection system
What you bring to the table
- You are studying mechanical engineering, computer science, CES or a comparable subject
- First experience with PyTorch and Deep Learning is favorable
- A high degree of initiative, independence, and problem solving skills
What you can expect
- Ideal conditions for practical experience alongside your studies
- GPU server for data science applications and for working efficiently with large models
- Flexible working to combine study and job in the best possible way
We value and promote the diversity of our employees' skills and therefore welcome all applications - regardless of age, gender, nationality, ethnic and social origin, religion, ideology, disability, sexual orientation and identity.
Severely disabled persons are given preference in the event of equal suitability.
With its focus on developing key technologies that are vital for the future and enabling the commercial utilization of this work by business and industry, Fraunhofer plays a central role in the innovation process.
As a pioneer and catalyst for groundbreaking developments and scientific excellence, Fraunhofer helps shape society now and in the future.
Interested? Apply online now. We look forward to getting to know you!
For any further information on this position please contact :
Alexander Kreppein, M.Sc.
Research fellow in the department production quality
Phone : +49 241 8904-289
Fraunhofer Institute for Production Technology IPT
Requisition Number : 74085 Application Deadline :
Bachelor-/ Master Thesis »Synthetic Data for Deep learning Models in production«
To increase the performance of deep learning models in these applications, customized pipelines for image augmentation must be designed that enrich existing datasets with synthetic images. With its focus on developing key technologies that are vital for the future and enabling the commercial utiliza...
Master Thesis - Efficient Markov Chain Monte Carlo Techniques for Studying Large-scale Metabolic Models
In this project, you will bring MCMC methods into the setting of metabolic flux inference and, with inspiration from existing algorithms, develop tailored MCMC algorithms. The main focus of the project can develop either more in the mathematical theory of MCMC, the implementation of code for the Jül...
Bachelor-/ Master Thesis: »Modeling of heating processes for mono-material sandwich panels«
In the group »Laser System and Heating Technology«, we design and construct various prototype machines for laser-based and laser-assisted processes. Within the scope of your thesis, you will develop the laser-assisted heating processes for recyclable mono-material sandwich panels (MMSP). This increa...
Student for Bachelor or Master Thesis / Working Student / Internship Software Engineering (m/f/d)
Student of one of the following disciplines (or similar): Computer Science, Technical Computer Science, Information Technology, Electrical Engineering, Communication Systems, Data Science & Artificial Intelligence, Business Information Technology, Media Engineering, Microelectronic Systems. Improvin...
Bachelor-/ Master Thesis: »Developing of a CAx-Software for rotating Laser Powder Bed Processes«
With its focus on developing key technologies that are vital for the future and enabling the commercial utilization of this work by business and industry, Fraunhofer plays a central role in the innovation process. As part of your thesis, you will investigate the current state of the art of software ...
Master Thesis - User Behavior in Buildings based on Automated Model Generation
Master studies in the field of of mechanical engineering, industrial engineering, process engineering, informatics and mathematics or similar. Based on a toolchain, which automatically generates building Modelica models based on an intuitive language for room descriptions developed at the institute,...
Internship / Master Thesis – Quantum Optimal Control Algorithms for Electrochemical NMR Applications
At IEK-9, we are developing exciting interdisciplinary methods to investigate the key mechanisms in batteries and electrolysis using pulsed NMR. By understanding the spin states that have to be experimentally prepared, we are tailoring magnetic pulses to push the limits of NMR in operando experiment...
Master Thesis or Internship - Graph Partitioning in Power System Studies
Develope new or tailor existing QUBO (Quadratic Unconstrianed Binary Optimization) formulations for graph partitioning objecitves to the use case found in step. Interdisciplinary work combining physics, mathematics, computer science, and engineering. An interesting and socially relevant topic for yo...
Master Thesis - User Behavior in Buildings based on Automated Model Generation
Master studies in the field of of mechanical engineering, industrial engineering, process engineering, informatics and mathematics or similar. Based on a toolchain, which automatically generates building Modelica models based on an intuitive language for room descriptions developed at the institute,...
Master Thesis or Internship - Graph Partitioning in Power System Studies
Develope new or tailor existing QUBO (Quadratic Unconstrianed Binary Optimization) formulations for graph partitioning objecitves to the use case found in step. Interdisciplinary work combining physics, mathematics, computer science, and engineering. An interesting and socially relevant topic for yo...