Jakob Hærvig


Research group leader for Multiphase Flows and Heat Transfer

Hi, I am an Associate Professor at Aalborg University. My work focuses on fluid dynamics and currently more specifically around heat storage, the dynamics or agglomerating particulate flows and heat exchangers.

If you are interested in my teaching or research, feel free to browse below or reach out via email.


The following highlights some of my teaching activities. They range from traditional semester courses to specialised PhD courses to simple guides and workshops.

Course: Heat transfer

I teach a full course covering many aspects of heat transfer for engineering students.

Guide: Python

Python is gaining popularity for general purpose programming due to its ease of use and broad functionality. Quite a few approaches exist when setting up Python. I’ve created a set of interactive slides that explains how I personally prefer to install and work with Python.

PhD Course: Open-source Computational Fluid Dynamics Course

OpenFOAM is the most widely used open-source toolkit for CFD. The fact that OpenFOAM is open-source makes it highly versatile allowing you to modify any aspect of the code to his/her needs and set up fully automated workflows. While large scale simulations with commercial CFD software can be extremely expensive due to license fees, OpenFOAM® can be run on massively parallel HPC’s at no additional cost.

Workshop: OpenFOAM Days

OpenFOAM Days is an initiative I’ve crated to promote the use of open-source software for computational fluid dynamics simulations for engineering students. OpenFOAM has a notorously steep learning curve and the goal of this workshop is to students smoothen it out.


The following lists some of my recent research.


OpenTerrace is a pure Python framework for thermal energy storage packed bed simulations. It is built from the ground up to be flexible and extendable on modern Python 3.x with speed in mind. It utilises Nvidia CUDA cores to harness the power of modern GPUs and has automatic fallback to CPU cores.