Research on deformation uniformity control of thin-walled conical aeroengine forgings based on GA-SVR

The performance and service life of aeroengine forgings, vital components of aeroengines, are profoundly impacted by their forming quality, which is substantially influenced by deformation uniformity. This study aims to control the deformation uniformity of thin-walled conical aeroengine forgings through preform design. Initially, the DEFORM DOE (design of experiments) module and MATLAB are utilized to simulate various scenarios, automatically acquiring samples with preform die geometry parameters as variables. Subsequently, a deformation uniformity prediction model is constructed by integrating genetic algorithm (GA) and support vector regression (SVR). The fitness function of the genetic algorithm is defined as the mean squared error (MSE) between the predicted values of the SVR model and the finite element simulation values, iteratively optimizing the SVR model. The results reveal that the established prediction model, based on limited sample data, achieves high accuracy and efficiency, significantly diminishing the frequency of trial-and-error simulations while enhancing the efficiency of preform design for thin-walled conical forgings. Finite element simulation outcomes and macro/microstructure test results following trial production further validate the deformation uniformity prediction model’s exceptional accuracy.