Short term water heat storage an experimental and numerical investigation of phenomena that affect the degree of thermal stratification

Abstract: Thermal stratification was studied by measurements of the temperature and velocity fields in water heat storage's and by numerical simulations. The work includes experimental investigations in three model heat storage's of different size and comparisons to numerical simulations for two of the storages. Thermal stratification is essential for most applications of water heat storage's. Different ways of measuring the degree of stratification were presented. Exergy efficiency was considered as best to use if the spatial resolution of the available temperature field is good. The thermal stratification is strongly dependent of the flow pattern in the storage. Mixing of hot and cold water near the inlet during charging can have a severe impact on the stratification. Much of the work was therefore focused on studies of this effect. Velocity measurements were first performed with Laser Doppler Anemometry techniques and later with the use of a Particle Image Velocimetry technique based on double or multiple exposure photographs of particles in seeded water. These measurements focused on studies of the boundary layer at the wall. The results show that boundary layer velocities increase with the available height up to a limit, determined by the temperature difference, after which the dependence was very weak. They also show an exchange of water between the boundary layer and the core of the storage. A video based Particle Image Velocimetry technique was finally developed which opened the possibility to measure the in plane velocities in a cross section of a small model storage during complete experiments. FLOW3D, a commercial software for flow simulations, was used to simulate measured velocity and temperature fields. Good agreement with the experimental results was found. This justified the use of FLOW3D to simulate the effect on stratification of parameters that could not be studied in the experiments. The simulations for the small storage show that differences in the exergy efficiency obtained in the initial part of charging cycles, due to different charging conditions, tended to be evened out by heat diffusion during the continuation of the cycle. This was found to be true at least as long the Richardson number was kept high enough to avoid forced mixing. An alternative choice of the length and velocity scales to use in the Richardson number was suggested. The width of the inlet slot was used as the characteristic length and the velocity of the inlet jet when it approaches the storage wall, as the characteristic velocity for the mixing. With this definition for the Richardson number the simulations indicate that unstable charging can be expected for Richardson numbers below 0.5-0.05. Further work is needed for verification of these findings for larger storage's. Still it is believed that the results of this work can be valuable for design optimization of short term water heat storage's.