Numerical Investigation of Thermal Stratification and Numerical Diffusion in a Large-scale Water Pits Heat Storage

Large scale water pit heat storage (PTES) is a key component to increase the efficiency of solar thermal utilization. Based on the 1535-1301 model, this paper presents a two-dimensional simulation of the annual thermal performance of PTES using TRNSYS. This model encrypts the soil grid according to the water pit layers to improve the accuracy of heat loss calculation. The accuracy of the model is verified by comparing it with measured data from a solar plant in Dronninglund, Denmark. The results show that the average difference between the model and the measured data is approximately 2.5%. In addition, the temperature prediction error and numerical diffusion of each layer become smaller as the nodes number increases. The stratification coefficient averages about 200 for summer charging and decrease to 100 for winter discharging, Mix number is 0.7 at charge and 0.35 at minimum, indicating that stratification is more pronounced in summer and mixing is higher in winter. The parametric coupling study found that the nodes number and the time step are more accurate when they are negatively correlated, i.e., fewer nodes are paired with longer time steps. Adiabatic admixture decreases the Mix number prediction accuracy but increases the stratification coefficient one.