OBRABOTKAMETALLOV TECHNOLOGY Vol. 24 No. 3 2022 Further, the validation experiments were performed using the process parameters different than that are used for developing the models. A comparative of the predicted results with the experimental-based mathematical model and artifi cial neural network (ANN) is shown in Table 3. The model accuracy is assessed by obtaining % error between the predicted and experimental values of wear rate for different process parameters. The % error is obtained using Eq. (5). Predicted value –Expt value 100 Average error = . Expt value (5) Table 3 presents data on the specifi c wear rate predicted by the developed models. Predicted results are seen in good agreement with the experimental values with average error of 10.16 % for experimental-based model and 3.57 % for ANN model. It is apparent that the results predicted by the ANN model are having a better agreement with the experimental results as compared to experimental-based model. Conclusions This study attempted modelling sliding wear characteristics of PTFE composite reinforced with carbon fi ber (35 % by weight) against SS304 stainless steel. Experiments were carried out on the pin-on-disk at different normal loads, interface temperature, and sliding velocities. An experimental-based mathematical a b c d Fig. 5. Neural network (a) Training; (b) Validation; (c) Test; (d) All data set
RkJQdWJsaXNoZXIy MTk0ODM1