Comparision of soft computing methods in some engineering applications

Since the soft computing methods are successful in solving problems that are complex, difficult and including uncertain information processes and require relationship between the nonlinear parameters, they are used for many engineering problems. In this thesis, a comprehensive study is made about the soft computing methods. Relationships between the input and output parameter of electro erosion process which is commonly used in mechanical engineering were investigated as an example of engineering problem. Initially in EDM machine, some experimental studies were carried out by taking the discharge current (I), pulse on-time (Ton) and pulse off-time (Toff) as input parameters and surface roughness (Ra), electrode wear ratio (EWR) and material removal rate (MRR) as the output parameters. Using the experimental results and data set, ANFIS, ANN and GEP models are developed and then their performances were compared and best models for EDM process were proposed. According to the results of estimating the parameters of all models in the comparison in terms of statistical performance is sufficient, but observed that ANFIS model is slightly better than the other models.