Multistep Ahead Prediction of Specific Enthalpy of Steam Using Nonlinear Autoregressive Neural Network with External Input (NARXNET)
摘要
This study presents nonlinear autoregressive neural network with external input (NARXNET) for multistep ahead prediction of specific enthalpy of steam. The artificial neural network model has been trained using experimental data collected for specific enthalpy of liquid and specific enthalpy of vapour. The Levenberg–Marquardt algorithm has been considered for the training of NARXNET model. The performance of these models was compared in terms of root mean square error (RMSE), mean absolute error (MAE) and bivariate correlation coefficient (BCC) for a maximum sample size of 30 multistep ahead predictions. The results highlight excellent prediction performance of NARXNET model giving marginal error responses and excellent correlation responses between outputs and targets.