A new flatness pattern recognition model based on CA-CMAC network

Abstract

In the traditional flatness pattern recognition neural network, the topologic configurations need to be rebuilt with a changing width of cold strip. Furthermore, the large learn assignment, slow convergence, and local minimal in the network are observed. Moreover, going by the structure of the traditional neural network, according to experience, it has been proved that the model is time-consuming and complex. Thus, a new approach of flatness pattern recognition is proposed, based on the credit-assignment cerebellar model articulation controller (CA-CMAC) neural network. The difference in fuzzy distances between samples and the basic patterns is introduced as the input of the CA-CMAC network. Simultaneously, a credit-assignment learning algorithm is imposed. The inverse of activated times of each memory cell is taken as the credibility, and the error correction is proportional to the credibility. The new approach with advantages, such as, fast learning speed, good generalization, and easy implementation, is efficient and intelligent. The simulation results show that the speed and accuracy of the flatness pattern recognition model are obviously improved.