Convection

The goal of room and passenger cabin air conditioning is to provide occupants with comfortably tempered interiors with efficient energy use. To characterize the indoor climate, flow velocity and temperature of indoor air are of great importance. Air flows in enclosed spaces (offices, lecture halls, concert halls, vehicle and aircraft cabins) are predominantly characterized by mixed convection, which is realized by complex air conditioning systems. The material and energy transport is characterized by temperature difference and supply air velocity as boundary conditions. Mixed convection (in short mixed convection) can therefore be understood as a spatial and temporal superposition of natural and forced convection. Although usually highly turbulent, the velocity and temperature field is arranged in large scale coherent structures, the so-called circulation currents (LSC = large scale circulation). These significantly characterize the mass and energy transport, which in turn is decisive for the evaluation of indoor air quality. It could be shown in experiments that these LSC show jump-like transitions with sharp boundaries. The structures of these large-scale circulation flows are strongly dependent on Rayleigh number (Ra), Reynolds number (Re), and Archimedes number (Ar). Previous experimental studies demonstrate sharp discontinuous structural transitions when the ratios are varied, which prevents predictions of indoor air quality. Established models have so far taken this aspect into account only to a limited extent or not at all. In the research project, these structural transitions are to be investigated experimentally in a wide Ra and Re range. The geometry of the already successfully used transparent model room is based on that of a passenger cabin. The structures will be investigated with stereo particle image velocimetry (SPIV), with thermistor sensors and laser-induced fluorescence (LIF). The characteristic variation in the range of real room airflows using the model room will be achieved in the SCALEX facility (Scaled Convective Airflow Laboratory EXperiment) while maintaining Ra, Re and Ar with air and sulfur hexafluorite with working pressures between 1 bar and 10 bar. This experimental facility, which is unique in the world, can be used to investigate structural transitions in terms of their multistability over a wide range of key figures with high spatial and temporal resolution. Furthermore, it is to be examined whether the multistability can be suppressed in a targeted manner by imposing disturbances in the supply air system. The aim of the research project is a comprehensive experimental description of the structure formation in room air flows as a basis for prediction algorithms and numerical simulations. In addition, a contribution to an active control of the structure formation shall be made.

Funding: DFG

Partners:

RWTH Aachen, Prof. Dirk Müller, Chair of Building and Indoor Climate Technology

TU Ilmenau, Prof. Jörg Schumacher, Institute for Thermo- and Fluid Dynamics

Processors:      

Dr. rer. nat. Christian Resagk (Project Manager)
Martin Herzberg

The application of machine learning (ML) techniques in the analysis of experimental measurements and numerical simulations of turbulence opens unique possibilities to analyse complex and comprehensive data sets by new physical criteria and thus to gather a deeper understanding of the fundamental transport processes in such flows. Our project aims at new effective modeling strategies of turbulent superstructures in extended turbulent convection flows – gradually evolving large-scale patterns – by means of machine learning techniques. We want to accelerate the analysis in optical flow measurements, develop low-dimensional reduced models that can predict the coarse-scale dynamics, and extend the mathematical foundations of ML applications to obtain a more efficient prediction of these processes.

Funding:

Carl Zeiss Foundation

Partner:

Group Fluid Mechanics

Group Optimization Based Control

Group Software Engineering for Safety-Critical Systems

Researcher:      

• Prof. Dr.-Ing. Christian Cierpka (project leader)