Heat Flux Estimation in Nonlinear Materials Using Kalman Filter-Enhanced Neural Network

摘要

This paper presents an efficient technique for analyzing the surface heat flux of a space shuttle upon reentry using annonlinear inverse heat conduction technique based upon a Kalman filter-enhanced Bayesian backpropagationnneural network. The continuous-time analog Hopfield neural network is used to solve various forward problems tonobtain training data for the Kalman filter-enhanced Bayesian backpropagation neural network. The calibratednKalman filter-enhanced Bayesian backpropagation neural network is then used to inversely compute the boundarynconditions fromgiven sets of temperature data obtained using the continuous-time analog Hopfield neural network.nThe results show that the proposed method can predict the unknown parameters of the current inverse problemsnwith an accuracy of 0.001%. The performance of the Kalman filter-enhanced Bayesian backpropagation neuralnnetwork scheme is shown to be better than that of a Bayesian backpropagation neural network or a stand-alonenbackpropagation scheme calibrated using a Levenberg–Marquardt backpropagation algorithm.