Genetic Algorithm Optimization of Sand Casting Process Parameters on The Fatigue Strength of Recycled Petrol Generator Pistons

Authors : Blessıng Ngozı Goodluck ALIEMEKE, M. OLADEİNDE

Pages : 45-51

View : 13 | Download : 6

Publication Date : 2020-03-27

Article Type : Research Paper

Abstract :- Engine components are necessary for smooth running of generators, vehicles and various kinds of locomotives. One of the most important parts of an automobile engine is the piston. The piston of an engine converts thermal energy generated in the combustion chambers into mechanical energy necessary for power transmission. Aluminum alloys reputed for lightness in weight and good conductivity are mostly used in the production of pistons. The Design of Experiment employed in the production of the piston is the Taguchi Orthogonal array method. It provided a platform on which the various experimental conditions were observed. Sandcasting was used in the production of pistons. Mathematical model was developed using Multiple Linear Regression technique and ANOVA test was carried out to ascertain the developed model adequacy. The model was adjudged adequate with an adjusted R2 of 99.85%. Signal-to-noise ratio technique was used to determine the optimal level of the process parameters which are 7500C, 49.998Hz and 30.01 seconds for pouring temperature, vibration frequency and vibration time respectively. Furthermore, the developed model was used as objective function in the Genetic algorithm tool box and the result obtained was in consonance with that obtained in Signal-to-noise ratio technique. In confirmation of test carried out it was noticed that the actual experimental values were similar to that predicted by the model
Keywords : Genetic algorithm, , sandcasting, design of experiment, Taguchi