Applications of Cascade Correlation Neural Networks for Manufacturing Process Monitoring

Abstract

It is much more important for manufacturing products to accurately and quickly recognizing/monitoring quality problems in a complex manufacturing process. Back Propagation Neural Network (BPNN) is receiving increased attention in the process monitoring because of their universal function approximate. In this study, Cascade Correlation Neural Network and Back Propagation Neural Network simultaneously have been trained to monitoring faulty quality categories of the products being produced in manufacturing process. Two examples were used here for analysis. Promising results were received according to accuracy using both Neural Network models but it was concluded that the Neural Network Model based on Cascade Correlation algorithm performed better in comparison with the Neural Network Model based on Back Propagation algorithm.