How is machine learning used at NOVUM to precisely determine the state of charge of batteries? And what does this have to do with the efficiency and service life of batteries? Our machine learning engineer provides exciting insights into his work in the field of battery monitoring and explains how innovative algorithms are used to optimize sustainability and resource usage.
In which area are you currently working at NOVUM?
I work as a Machine Learning Engineer at NOVUM. My tasks are very varied — for example, my current focus is on software development, especially in the area of Battery monitoring.
What exactly is machine learning and what does it have to do with the field of monitoring?
Machine learning is an important part of our battery diagnostics and management systems at NOVUM. With the help of this and artificial intelligence (AI), we analyze the operating data of the batteries — e.g. current and voltage — and create a digital twin from it. This enables us to determine the state of charge, the current capacity and the expected service life very precisely.
Specifically, my field involves calculating and monitoring various metrics in order to estimate the state of charge of batteries as accurately as possible. To do this, I use methods such as voltage measurement and model-based algorithms. Because I work directly with our customers’ data, I can incorporate individual factors and optimize the calculations so that we get the most accurate and reliable results possible for a battery’s state of charge.
Can you give us an overview of what the battery charge level actually is and what added value it gives us?
The state of charge of a battery — also known as the State of Charge (SoC) shows how much energy is still available in relation to the maximum capacity. This sounds simple at first, but in practice it is quite complex, because in addition to the voltage, the current flow, temperature and other influencing factors must also be taken into account. However, a precise SoC analysis is essential in order to:
- maximize the efficiency of the battery,
- extend their service life,
- avoid potential security risks and
- enable predictive maintenance.
Continuous monitoring and intelligent algorithms can then be used to optimize charging cycles, prevent deep discharge or overcharging and detect signs of degradation at an early stage. In this way, SoC analysis makes a significant contribution to the reliability and sustainability of battery systems.
What role does the State of Health (SoH) play in this context?
While the SoC shows the current charge status, the SoH This provides information on long-term performance, describes how “healthy” a battery still is and how much it has aged over time. This is particularly important for second-life projects because it shows whether a battery is suitable for further use or whether it should be recycled.
What other challenges would appeal to you at NOVUM in the future?
I would like to develop further in the field of machine learning and take on more responsibility — especially in model monitoring and improving existing algorithms. I really enjoy working with innovative approaches and developing new solutions that make our processes even more efficient.
What experience did you have before joining NOVUM, and how did you come to NOVUM?
Before I started at NOVUM, I was a working student at the University of Reutlingen. There I developed machine learning models that analyzed complex data sets — often in interdisciplinary teams, including with psychologists and roboticists. This collaboration was very exciting and further strengthened my interest in machine learning and data-driven solutions.
A friend who worked here as a student trainee told me about NOVUM over a beer after work. He said that the projects are exciting, the work culture is open and the customers from the energy industry are totally diverse. I was particularly impressed by the opportunity to work in the field of machine learning and to constantly face new challenges and drive innovation. It sounded like exactly the right environment for me. So I applied.
What I particularly appreciate now at NOVUM is the clear focus on sustainability. That motivates me in my work. I think it’s great that we can not only develop innovative technologies with our work, but also make a real contribution to the circular economy.
Why don’t you tell us a bit about your career — how did you get to where you are today?
I’m originally from India and moved to Germany for my Master’s in Computational Modeling. That was a really good decision! I was really enthusiastic about the course because it was much more focused on software development than my bachelor’s degree in mechanical engineering. I found mechanical engineering exciting, but I’m glad I went in the direction of software development because it offers many more opportunities in a professional context, but also to be creative.
After graduating from TU Dresden, I first moved to Reutlingen, where I found my first job. However, Dresden still felt like home, so I moved back here. I just feel completely at home here!
And how do you spend your free time in Dresden?
In my free time, I enjoy being outdoors — especially in summer. I love walking along the Elbe, cycling and bouldering. Dresden is a great place for this, thanks to the Elbe cycle path and various climbing gyms. I also often go hiking in Saxon Switzerland or to the gym with friends from my university days.
Another hobby of mine is dance classes, especially salsa and bachata. It’s the perfect balance to work for me. What I particularly appreciate about Dresden is the cultural diversity and the international contacts through which I have met some of my closest friends.
