Research On Structural Optimization Of Rolling Shear Based On Surrogate Model

As a key equipment of the medium plate finishing line, rolling shear is widely used by iron and steel production enterprise because of its high efficiency and good section quality. Compared with traditional inclined blade shear, rolling shear regards pure rolling shear as its goal, which can effectively improve the quality of steel plate shear. Both mechanical rolling shear and hydraulic rolling shear mainly consist of linkage, and linkage is one of the most important parts of the rolling shear. The upper blade realizes rolling cut under the control of linkage. Meanwhile connecting rod is the main force component, and in the working cycle, the force is transmitted to the upper cutter through it.Based on the importance of the linkage, the structural optimization about shearing mechanism of rolling shear was carried out in this paper. The concrete research content is as follows:1) To solve the low accuracy and efficiency problems in structural optimization, a new method based on surrogate model of response surface methodology was proposed.The connecting rod of hydraulic rolling shear was taken as object, the proper design variables were chosen by Latin hypercube experimental design. Polynomial response surface model was built which taken the mass of connecting rod as the objective function. The best parameter combination was achieved by using Genetic Algorithm(GA) to optimize the model, achieving the purpose of lightweight design.2) Based on the mechanical reliability theory and the principle of Monte Carlo method, the reliability analysis for the optimized rod of rolling shear was conducted. Through parametric modeling, size parameters, material properties, and external load were taken as random input variate. The reliability of the rod and the main factors that affect the reliability were got, which were calculated by the Monte Carlo method. The result shows that the rod has a high reliability, and the reliability of the connecting rod and the sensitivity of output variables were obtained, which provides reference for the further design or optimization of the rod.3) To solve the problems of the uneven overlap quantity and the horizontal displacement of the upper blade during the cutting process, a new method based on surrogate model of radial basis function was proposed. Through uniform design, rational experimental samples were selected and the surrogate model of rolling shear based on radial basis function was established, then the model was optimized by using particle swarm optimization(PSO), realizing the purpose of optimization.