Neuro-fuzzy modeling of hot extrusion process

Agile manufacturing systems require models that can predict in real time the effect of various process parameters of a production process. Hot extrusion is a commonly used production process in forging industry. The relationship between the process variables (input), viz., die-velocity, temperature of billet, die angle and co-efficient of friction of a given material and extrusion force (output) required to extrude a shaft is very complex and is amenable to neuro-fuzzy approach. In this paper, a soft computing approach, i.e., neuro-fuzzy technique is used in modeling hot extrusion process to predict the punch force required to extrude a transmission shaft from ck-45 steel billet. The neuro-fuzzy model has been created out of training data obtained from the finite element (FE) simulation and correlates well with the FE results. This work has considerable implications in selection and control of process variables in real time and ability to achieve energy and material savings, quality improvement and development of homogeneous properties throughout the component and is a step towards agile manufacturing. The close agreement of the values of the final extrusion force obtained by the NF model and the FE simulation clearly indicates that the model can be used for predicting the extrusion force in the range of parameters under consideration in real-time without having to perform any extensive and costly computations. This technique opens new avenues of parameter estimation, optimization and on-line control of complex agile manufacturing systems.