artificial neural network-based predictive model for output characteristics in drilling of quartz cyanate ester polymeric composite

Apart from the widely used polymeric fibers, quartz fiber is the one which possess various characteristics. quartz polymeric fiber in combination with cyanate ester resin produces high-performance composite which has excellent properties and used primarily in military applications. the present investigation aims at developing a model to predict the output characteristics of hole in the drilling of quartz composite laminate. output parameters considered are thrust force, torque, exit delamination factor, hole diameter, cylindricity and surface roughness. vacuum assisted resin transfer moulding (vartm) process was adopted for the manufacturing of the laminate. full factorial design of experiments was considered for the selected input parameters and experiments were carried out. further model was developed to predict the output parameters employing back propagation neural network (bpnn) method and found that the optimal network architecture is 3-45-15-10-6 with mean squared error (mse) of 0.0105 . experimental results were analyzed and studied the influence of input parameters in this drilling process. the testing data show a good match with the output parameters predicted from the model and maximum error obtained is 7.58%. further, the model developed was validated with new batch of experiments and the values obtained are satisfactory with maximum error of 7.17%.