A Methodology for Aero-engine Dynamic Characteristic Parameter Identification based on SSA-NARX
A dynamic characteristic parameter identification method based on the sparrow search algo-rithm(SSA)and nonlinear autoregressive exogenous(NARX)neural network was proposed for the aero-engine dynamic characteristics modeling.SSA was used to optimize the weights and biases of the NARX network iteratively to enhance its accuracy and generalization capability;the optimized NARX network was used for identifying dynamic characteristic parameters;simulation tests were conducted using flight test dataset.The results show that SSA-NARX method is better than the NARX and PSO-NARX methods obviously.By the SSA-NARX method,the maximum relative error absolute values δmax between the output parameters N1,N2 and exhaust gas temperature(EGT)and actual values are reduced to 3.81％,1.24％and 3.47％respectively;the relative errors between the dynamic characteristic indicators Ti and T,and ac-tual values are less than 5％;through ten cross tests,the mean values of root mean square errors(RMSEm)of validation results corresponding to parameters N,,N2 and EGT are 0.29,0.18 and 1.50 respectively.The accuracy,real-time performance and robustness of the model meet the requirements for simulation.
aero-enginedata drivensparrow search algorithm(SSA)nonlinear autoregressive exoge-nous(NARX)neural networkdynamic model identifiication
万方数据
维普
