Adaptive Neural Network Fuzzy Inference System for HFC Processes

The paper presents the design and implementation of a fuzzy inference system (FIS) trained with adaptive neural networks for the generation of specification references in high frequency current (HFC) hardening processes. The specification references are then further used for the control of the process in obtaining the desired outcomes in terms of material hardening and resistance. The FIS is trained using data obtained from experimentation on an industrial HFC device. The trained FIS is then compared to a manually tuned FIS, resulting from expert and operator designs. The results led to the development of intelligent control interfaces in real time through the ANFIS method