A Hybrid Model Using Genetic Algorithm and Neural Network for Optimizing Technology Parameters in Skew Rolling Seamless Pipes

Abstract

In the production of seamless pipes, there are many factors influencing the quality of the seamless tubes such as diameter of original pipe, feed angle, temperature, minimum roll gap and so on. Unfortunately, the selection of the parameters still relies on manual operation and the experience of engineers. In this paper, a novel hybrid model through integration of genetic algorithm(GA) and neural network is proposed to optimize the technology parameters. Firstly, the neural network model is developed between technology parameters and final pipes dimensions. Then, genetic algorithm is used to optimize the neural network structure. At last, accuracy of the parameters is verified by experiment. The research results indicate that the proposed approach can effectively help engineers determine optimal technology parameters and achieve competitive advantages of product quality and costs