Modelling of the clearance effects in the blanking process of CuZn30 sheet metal using neural network − a comparative study

Abstract

Clearance effects on the product quality and blanking force in sheet metal blanking process are first investigated experimentally, and then modelled through neural network (NN) approach. Using eleven clearance values ranging from 8% to 18% with sampling of 1%, blanking process is applied to sheet material CuZn30 with a thickness of 1mm. During the experiments, blanking force, smooth sheared/fractured rate and burr height for the resulting products are measured for each clearance value, and as such, 11 data samples have been prepared. Six of them are taken as training data to train the network while with the remaining for testing purposes. Several estimation results are illustrated which verify that the presented NN can estimate the nonlinear relationship of blanking force, smooth sheared/fractured rate and burr height with the clearance with a maximum error of about 1%. These results are also compared to those offered by a recent study benefitting from fuzzy logic whose design is a challenge requiring proper and sufficient expert knowledge for tuning of numbers and shapes of membership functions, linguistic control rules. We conclude that better estimation accuracy and design simplicity are important advantages of our proposal.

Optimization of a Multi-Type PMSM Based on Pyramid Neural Network

“…Recently, many meta-modeling methods based on big data [19] and machine learning have attracted wide attention. Neural networks (NNs) based on fuzzy logic have been used to model the influence of clearance on product quality and the blanking force in blanking process [20].Artificial neural networks (ANNs) have been used to model the nonlinear relationship between the power harvested with a piezoelectric pendulum and its resistive load, magnetic excitation frequency, etc.…”