NEURO-FUZZY GMDH AND ITS APPLICATION TO MODELING GRINDING CHARACTERISTICS

Mathematical models, in which many input variables are involved, requi re a range of input and output data, since the number of parameters in creases with the input variables. GMDH (Group Method of Data Handling) has been used for the identification of a mathematical model that has many input variables but limited data needs by using a hierarchical s tructure. This paper proposes a neuro-fuzzy GMDH model, adopting Gauss ian radial basis functions (GRBF) as partial descriptions of GMDH. GRB F is reinterpreted as both a simplified fuzzy reasoning model and as a three-layered neural network. As an example of applying the algorithm , the wheel wear equation is identified, using data from experiments o f abrasive cut-off. In the model, characteristics of work materials, g rinding fluids, factors of wheels, wheel velocity and table feed are u sed as input variables, and the grinding ratio is the resulting output . The validity of the model is confirmed within the predicted accuracy by using additional data.