Establishment method and device of grain size prediction model and prediction method

Abstract

The invention discloses an establishment method and device of a grain size prediction model and a prediction method, and relates to the technical field of alloy materials. The establishment method of the grain size prediction model comprises the following steps: selecting process parameters of an alloy material as characteristic values of the grain size; predicting a grain size model by adopting a machine learning algorithm; and evaluating different models through a cross validation mode to screen out the optimal model. According to the method, the grain size can be accurately predicted according to the characteristic value of the grain size, so that prediction of the grain size of aluminum and aluminum alloy is realized, a guiding direction is provided for development of commercial software for grain size prediction and establishment of a database, and the mechanical property of an aluminum alloy material in industrial production is improved.

Description

technical field [0001] The invention relates to the technical field of alloy materials, in particular to a method, device and prediction method for establishing a grain size prediction model. Background technique [0002] With the continuous expansion of the application fields of alloys, people's requirements for its performance are also getting higher and higher. Grain refinement can increase casting speed, reduce cracks, cold shock and deflection, and improve the mechanical properties of aluminum, magnesium and other alloys. Therefore, grain refinement has become the most common and economical method in the industry. [0003] Grain size can well characterize the result of grain refinement, so the grain size prediction model has been paid more and more attention by researchers. At present, various specific prediction models have been developed. These models are generally obtained through physical or mathematical derivation under certain conditions based on certain assumpti...

AI Technical Summary

Problems solved by technology

At present, various specific prediction models have been developed. These models are generally obtained through physical or mathematical derivation under certain conditions based on certain assumptions, and some are obtained based on regression of empirical data, lacking grain refinement results. Quantitative description of the relationship with refiner and process parameters leads to the lack of systematic theoretical guidance for experimental research, which brings great challenges to the realization of intelligent and precise chemical industrial production

Images