Announcing Our New Course: Battery Modeling with Physics-based and Machine Learning Techniques

This course is designed for engineering and R&D teams working at the intersection of battery systems, machine learning, and simulation.

4 min read

June 6th, 2025

Last updated: June 11th, 2025

Announcing Our New Course: Battery Modeling with Physics-based and Machine Learning Techniques

As the global demand for electrification and energy storage accelerates, the limitations of traditional battery modeling workflows have become painfully clear.

R&D teams are bogged down by slow simulations and labor-intensive experimentation.

Machine learning engineers are eager to contribute to energy and climate tech but often lack the domain knowledge to make meaningful impact. Meanwhile, original equipment manufacturers (OEMs) and grid-scale analysts face unpredictable battery behavior under real-world conditions, and the tools to model these systems just aren't keeping up.

It’s time to rethink how we simulate, predict, and optimize battery performance.

That’s why we’re excited to announce Battery Modeling with Physics-Based and Machine Learning Techniques course, a 6-week expert-led course built to equip engineers, scientists, and technical founders with the skills to build scalable, accurate, and reproducible battery models.

Why THIS Course?

This is not just another online course.

It’s a hands-on, industry-informed program that bridges the gap between physical modeling and data-driven insights. Designed with industry requirements in mind, the course blends the rigor of electrochemical simulation with the flexibility of machine learning.

Using tools like PyBaMM, Python, and TensorFlow, you’ll learn to model batteries across multiple scales, from microscale to pseudo-2D, and integrate thermal, aging, and impedance behaviors into your simulations. Then, you'll take it further: exporting synthetic data, training ML models to predict degradation, and building your own hybrid modeling pipeline in a capstone project.

Who Is THIS Course For?

This course was built for those pushing the boundaries of energy technology.
If you are:

A battery R&D engineer trying to accelerate your workflow and reduce reliance on time-consuming lab experiments,
A machine learning engineer transitioning from big tech or fintech into energy and materials modeling,
An automotive or BMS specialist working on battery state estimation, lifecycle prediction, or thermal management,
A grid-scale energy analyst trying to make sense of dynamic storage behavior under fluctuating demand,
A PhD student, postdoc, or industrial researcher seeking reproducible, publication-quality modeling pipelines, or,
A startup founder or consultant aiming to prototype high-fidelity models that actually work in the real world

then this course is designed for you.

Check out the syllabus here!

More About The Course

Throughout the six weeks, you’ll dive into physics-based approaches like equivalent circuit modeling and pseudo-2D simulations, while also getting practical experience with ML frameworks tailored for battery systems.

You’ll compare modeling strategies, fine-tune thermal parameters, and use physics-informed neural networks (PINNs) to fuse domain knowledge with data. And because we know how important applied work is in this space, your final capstone will not only demonstrate your learning but serve as a showcase-worthy project for your portfolio, pitch deck, or product roadmap.

Filip Maletic, PhD

The course is led by Dr. Filip Maletic, a battery modeling expert with a Ph.D. in battery algorithms and control. His work spans advanced electrochemical modeling, Kalman filter-based state estimation, and data-driven aging models. Until recently at AVL List GmbH, one of the world’s largest automotive consultancies where he developed custom multiphysics models ranging from 3D microscale to full cell-level frameworks, bridging complex electrochemical behavior with real-world predictive performance.

This hands-on course is more than an educational experience.

It's an upgrade to your technical workflow, your modeling mindset, and your professional toolkit. Whether you're building next-gen BMS, modeling battery packs for renewable integration, or breaking into the energy AI space, you'll leave this course ready to deploy smarter, faster, more accurate simulations.

Battery modeling doesn’t need to be slow, siloed, or outdated anymore. With the right tools and instruction, it can be fast, integrated, and impactful.

Applications are open now

The next cohort begins on June 16 and space is limited. 👉 Explore the curriculum and apply here

For questions, contact Catherine at [email protected].

And if you know someone working in batteries, energy, or applied ML, pass this along! The battery industry needs more hybrid thinkers.

—

Neovarsity is a Berlin-based deep tech skills platform. We build industry-driven, cohort-based programs in collaboration with world-class experts to prepare talent and teams to solve problems in areas with real-world impact.

Stay tuned for more updates and insights. Follow us on LinkedIn and join the conversation using #FutureThroughDeepTech.

Looking to upskill for high-impact roles in battery modelling?

This course teaches you how to model batteries using physics-based simulations and machine learning. You will gain:

Fluency in Python-based battery simulation and ML integration
Tools to accelerate your workflow and modeling accuracy
Portfolio-ready outputs, including simulation plots and ML performance reports

Frequently Asked Questions (FAQs)

This course is designed for engineers, researchers, and graduate students working in battery technology, energy systems, or related fields. A basic understanding of electrochemistry and Python programming is helpful but not strictly required.

While the course doesn't require formal prerequisites, we do assume you're comfortable working in Python and using Jupyter Notebooks. You don’t need to be a machine learning expert, but if you've had some exposure to the ideas, that’s definitely a plus.

You’ll be working with PyBaMM for battery simulations, Python for coding, and TensorFlow when we shift into machine learning. Everything runs in Jupyter Notebooks, and we’ll walk through the setup together during the first week. We’ll guide you through installing the necessary packages and help you get your modeling environment up and running.

The course runs for six weeks and is delivered in a blended format. That means you'll get structured lessons alongside hands-on labs and coding exercises. Each week, we build on the last, from fundamentals all the way to advanced modeling techniques. We wrap it all up with a capstone project where you’ll bring everything together into a real-world hybrid modeling workflow.

You’ll get hands-on experience with a wide range of models. We’ll start with physics-based approaches like microscale simulations, thermal models, and pseudo-2D structures. Then we’ll dive into equivalent circuit models and finish by exploring how machine learning such as neural networks and Gaussian processes can help predict degradation. In the capstone, you’ll even combine these approaches into a hybrid model.

The capstone is where you get to put everything into practice. You’ll simulate battery behavior to generate synthetic data, then train a machine learning model to predict degradation. There’s a peer review step so you can give and receive feedback, and you’ll submit a final notebook, along with a short report. If you’d like, you can even include a video walkthrough of your project, totally optional, but encouraged!

Every week includes a mix of theory and hands-on work. You’ll simulate battery behavior, modify model parameters, analyze thermal and performance outputs, compare modeling methods, and train machine learning models. Assignments usually involve submitting notebooks, visualizations, and short written reflections on what you learned.

By the end of this course, you'll be able to model battery behavior across multiple scales, apply machine learning techniques to degradation analysis, and build hybrid workflows that integrate physics-based and data-driven methods. These are precisely the kinds of in-demand skills that are valuable whether you're aiming to contribute to electric vehicle development, energy storage innovation, or academic research.

Battery Technology Battery Modeling Energy Storage Machine Learning