Optimization of the Temperature Fields Using the ABC Algorithm by Selecting the Kappa Parameter in Heat Conduction

The goal of the study was to examine the distribution of temperature fields at each geometry point for heat conduction, using a type IV boundary condition at the point where the casting meets the casting mold. The focus was on comparing the temperature fields for reference calculations without disturbance and with 1% disturbance of reference temperatures. The artificial bee colony algorithm was used to optimize the kappa parameter value at the interface between the mold and the casting, which was a key element. Simulations were conducted for two different population sizes: 10 and 20 individuals. Each of these populations was tested with different numbers of iterations (4 and 8), and each iteration was run five times to ensure the reliability of the results. The ABC algorithm averaged the values of the calculated kappa parameter after five runs. We performed recalculations for 0% and 1% disturbances, using the obtained results to compare the actual temperature fields with the reference fields. We used the artificial bee colony algorithm in both cases to select the optimal value of the kappa parameter, thereby minimizing the differences between the actual and reference temperature field distributions. The analyzed system accurately represented the temperature distribution by partitioning the geometry of the mold and the casting into 1056 finite elements. The results allowed for a detailed analysis of the effect of disturbances on temperature distributions, as well as the artificial bee colony algorithm's effectiveness in optimizing the kappa parameter. Increasing the number of iterations and the number of individuals led to more accurate results, although this required more computational effort. The study showed that the ABC algorithm is an effective tool for optimizing thermal conductivity problems, especially in the context of varying boundary conditions. The study's final conclusions emphasize the importance of precise parameter selection in optimization algorithms, as well as the need to consider potential disturbances in simulation processes in order to obtain more reliable and accurate results on temperature distribution.