Font Size: 

The presented paper focuses on an investigation of the influence of the position of control points on the accuracy of heat transfer coefficient selection. In the presented study, the main focus is on the determination of the heat transfer coefficient in a layer separating two bodies. Such a situation occurs, for example, in solidification problems where a cast is held inside a mold and heat transfer takes place from cast to mold.
The authors use swarm intelligence algorithms for the task of determining the heat transfer coefficient. In the presented study, two different swarm intelligence algorithms are employed: Artificial Bee Colony (ABC) and Ant Colony Optimization (ACO). The numerical model is based on the authors own implementation of the transient heat transfer solver that uses Finite Element Method to solve the appropriate differential equation.
The study presents the results of heat transfer coefficient selection in the computational domain of one quarter of the square casting inside a square mold. Both swarm intelligence algorithms run for sets of 10, 15, and 20 individuals with 2 and 6 iterations. The study also takes into account possible inaccuracies in the reference temperatures in the form of 1%, 2%, and 5% noise. For both algorithms, three different sets of control points were used: one with points directly on the contact boundaries and two sets with increasing distance from the boundaries.
The results obtained in this work show that the location of control points has an impact on the quality of the results obtained in coefficient selection.