Heat flux on-line estimation in a locomotive brake disc using artificial neural networks

In this study, an inverse algorithm based on the Artificial Neural Networks and the Sequential Function Specification method was successfully applied for estimation of the heat flux absorbed by the locomotive brake disc. The three dimensional direct problem involving turbulent, unsteady and conjugate heat transfer boundary condition is numerically solved for known values of 47 different heat fluxes, and temperature histories of 18 different locations inside the brake disc were obtained. The braking process is experimentally simulated and the experimental data are used to verify the simulation results. Then 39 simulated heat fluxes are utilized to train the ANN and 8 remaining are used to test it. Results showed the ability of the ANN in accurate heat flux estimation. Furthermore, the consequence of changing the number and the locations of temperature sensors on the accuracy of the estimated results has been considered. Finally, the effect of noise on the exact temperature on the heat flux estimation has been investigated.