Optimization of Roll Forming Process with Evolutionary Algorithm for Green Product

Knowledge-Based Neural Network model is known as one of the most useful methods which can predict every single variability to create the process parameters for the data on Roll Forming process. To get the best quality of product and process parameters in roll forming, the Knowledge-Based Neural Network has to be trained with high reliability. To obtain the target aimed, this paper proposes a new novel of the optimal algorithm for training in the Knowledge-Based Neural Network model with the integration between Genetic Algorithm and Hill Climbing Algorithm. Initially, a global optimization method is carried out to find the global optimum area by using Genetic Algorithm, and then the Hill climbing Algorithm will effectively detect the positions of that local optimal region with high accuracy in the training of the Knowledge-Based Neural Network model. Additionally, to obtain the trained data set of the Knowledge-Based Neural Network model, the Finite Element Analysis result of the high fidelity Finite Element Model is used From the results of simulation, we can find out that the efficiency of the proposed method is higher than the conventional methods in optimization of the roll forming process.