Prediction model of hot strip crown based on industrial data and hybrid the PCA-SDWPSO-ELM approach

The accurate prediction of strip crown is the precondition shape preset model in hot rolling. In this study, a new hybrid forecasting proposed combination with extreme learning machine (ELM) and industrial data. production data 1780 mm rolling are collected by on-site acquisition system to form dataset. Principal component analysis (PCA) used reduce dimension input for modeling samples. To improve accuracy ELM, an improved PSO based on S-curve decreasing inertia weight (SDWPSO) optimize initial weights biases ELM. Finally, optimal ELM simple dataset establish named PCA-SDWPSO-ELM. comprehensive performance PCA-SDWPSO-ELM evaluated MAE, MAPE RMSE. superiority also proved comparing results those other three comparison models. research shows that method can solve problem nonlinear strong coupling traditional engineering. It suitable parameter optimization iron steel manufacturing industry, especially process control