Selection of Bending Parameters for Minimal Spring-Back Using an ANFIS Model and Simulated Annealing Algorithm

Abstract

In this research, a simulated annealing algorithm was used to minimize the spring-back in V-die bending process. First, an adaptive neuro-fuzzy inference system (ANFIS) model was developed using the data generated based on experimental observations. The output parameter of the ANFIS model is spring-back and the input parameters are sheet thickness, sheet orientation, and punch tip radius. The performance of the ANFIS model in training and testing sets is compared with those observations. The results indicated that the ANFIS model can be applied successfully for prediction of spring-back. Then, the ANFIS model was used as a function in simulated annealing algorithm to minimize the spring-back. The results showed that the proposed model has an acceptable performance to optimize the bending process.