To build a complex nonlinear prediction model among subjective and objective evaluation of passenger cabin seat comfort and improve the prediction accuracy of the model.In this paper,penalty parameter C,the channel control parameter ε and the kernel function parameter σ in support vector regression(SVR)are taken as the optimization objectives,the particle swarm optimization(PSO)algorithm is used to find the global optimal parameters,and PSO-SVR human-civil aircraft seat comfort evaluation and prediction model is built,and the prediction results are compared and analyzed.The analysis results show that:compared with the back propagation neural network model,the support vector regression model has good robustness;compared with the SVR model,the PSO-SVR model has higher prediction accuracy and smaller error fluctuation,the mean square error of the prediction results is reduced by 85.95%,and the determination coefficient is increased by 15.42%.So,the particle swarm optimization can effectively improve the prediction accuracy and generalization ability of the support vector regression model.
维普
万方数据