Prediction Model of Needle Valve Body Extrusion Grinding Process Based on GA-ELM

Abstract

There is a complicated non-linear relationship between the process parameters of the needle valve body extrusion grinding and the process effect, it is difficult to establish a complete and accurate process model. Therefore, by introducing genetic algorithm to optimize and improve the ELM extreme learning machine, a complete set of needle valve body process effect prediction model is established. The historical experimental data obtained by the self-developed needle valve body squeezing and grinding equipment were used as sample data for ELM algorithm model and GA-ELM algorithm model training. The results show that the optimized model can significantly improve the prediction accuracy of the data in the sample. The actual processing experiment of the needle valve body is carried out through the process parameter data in and out of the sample, the predicted value is compared with the actual value, and the performance of the accuracy of the predicted model is tested. The comparison results show that the actual data prediction error is kept within ±4%, which basically meets the actual prediction requirements, and thus provides a theoretical basis and reference value for subsequent process optimization.