Milk being stirred into a cup of coffee, water draining from a bath tub or smoke rising from a flame. We are all familiar with turbulence, but we still lack the precise description of what is happening in physical terms when it occurs, especially when it is fully developed and behaves far from “perfectly”. It is one of the main unresolved problems of classical physics, that researchers at DTU have now taken one step closer to solving.
“To date, we have been unable to describe how turbulence behaves when it is pushed out of equilibrium —for instance when speed increases in space or time. Non-equilibrium turbulence is one of the most prevalent types of turbulence found in both nature and industry,” says Associate Professor Clara Velte.
“These are simultaneously among the most problematic kinds and also the type of turbulence that we know the least about. I’ve been determined throughout my career to try and obtain this knowledge, and I’m thrilled that we have now been able to establish the theoretical framework and experimental facilities that will help us to finally make this knowledge breakthrough.”
The next step will be to gather the necessary data from practical turbulence experiments that will underpin and complete the theory. This will take place in a brand new, advanced turbulence laboratory that Clara Velte secured research funding to develop, and which has just opened its doors at DTU. Together with the theoretical framework, the group will be able to use the data they gather to formulate semi-empirical solutions to the governing equations for turbulent currents.
Companies poised
The data collected from the experimental work in the laboratory will be coupled to the theoretical framework. This will make it possible to develop improved analytical and numerical models for turbulence of varying degrees of non-equilibrium. Physics and the governing equations will be the basis for these models, and will typically be used to predict its dynamics under different conditions. Since the occurrence of objects such as droplets and particles can also have a significant impact on turbulence, the natural next step for the group has been to develop theory to allow them to better understand the dynamic between particles or droplets with surrounding turbulence.
“Many companies are interested in getting to the bottom of this. We are already collaborating with businesses that develop engines where fuel is injected at high speed, which means there is a keen interest in knowing more about turbulence in that context. Our other partnerships include companies producing spray dryers for use in food production, such as the drying of liquids like milk into powder,” says Clara Velte.
This new knowledge about turbulence will also be relevant to a range of other, different industries. Pharmaceutical businesses may want to get to grips with the turbulence found in the mixture of air and droplets in nasal sprays, while wind turbine manufacturers may be keen to gain a better understanding of the turbulence created in the complex rotating geometries that wind turbines comprise.
Obvious application for machine learning
Clara Velte and her research group are looking forward to using the new laboratory to carry out measurements and gather data that will increase our understanding of turbulence. They are also on the look out for colleagues with relevant expertise in machine learning.
“We have the perfect data. It’s high quality and uniquely well-calibrated against the degree of non-equilibrium in turbulence, which is the key parameter under investigation. That means that using artificial intelligence is an obvious approach to increase our knowledge and the opportunities for mathematical modelling of turbulence,” says Clara Velte.
The new turbulence laboratory was inaugurated in late September 2022 and was made possible thanks to funding awarded to Clara Velte by the European Research Council (ERC) and Poul Due Jensens Fond.
