In internal combustion engines designing, the cooling system is very important. As a matter of fact, having a suitable cooling system will let the manufacturers use more powerful engines. Radiator is the main part in cooling system which has been studied for many decades.

In the present study finding the best geometry and dimensions of a radiator of a specific engine is the main objective. The design methods used are separated into two parts, thermal solution, and optimization (based on a genetic algorithm). Then, in order to find the best possible geometry to reduce the pressure drop and increase the heat transfer, genetic algorithm is used to find the best geometry based on seven different variables. The results show that optimization will cause 15% increase in the cooling rate as well as a 4 and 10.7% decrease in pressure drop than those of thermal solution which had been used for validation.

radiator, optimal design, algorithm genetic, heat exchanger.