Predicting method of aluminum alloy die pressed casting grain size based on GA-ELM algorithm

Abstract

The invention discloses a predicting method of an aluminum alloy die pressed casting grain size based on a GA-ELM algorithm. The predicting method comprises the first step of using the new algorithm GA-ELM which combines a genetic algorithm GA and an extreme learning machine ELM; the second step of using the GA-ELM algorithm to predict the grain size with high efficiency, high precision and low cost; directly inputting a technological parameter which needs to be predicted into a trained GA-ELM model through a user, and obtaining a corresponding predicted value of the grain size after calculation. According to the predicting method of the aluminum alloy die pressed casting grain size, the GA-ELM model is the combination of the genetic algorithm and the extreme learning machine, the GA is adopted to conduct optimization on a weight matrix of an input layer and a threshold value matrix of a hidden layer of the ELM, the influence of randomness of the weight matrix of the input layer and the threshold value matrix of the hidden layer on the predicting accuracy of the ELM is avoided, the predicting accuracy and the predicting efficiency are improved, and the cost is sharply reduced; in addition, a model base of the predicting method of the aluminum alloy die pressed casting grain size can be enriched.

Description

technical field [0001] The invention relates to the field of aluminum alloy die-casting forming control, in particular to a method for predicting the grain size of aluminum alloy die-casting parts based on the GA-ELM algorithm. Background technique [0002] Aluminum alloy die castings are widely used in automobile, aircraft and other equipment manufacturing industries, and the quality of die castings will directly determine the quality of equipment products. While avoiding the defects of aluminum alloy die castings, the mechanical properties of the die castings themselves are also worthy of attention. The higher the mechanical properties of the die castings, the longer the service life and the higher the reliability of the die castings, which in turn affects the life of the overall assembly product and reliability. [0003] The most essential factor determining the mechanical properties of a material is the microstructure such as the morphology, size, orientation and distri...

AI Technical Summary

Problems solved by technology

Simulation software has high simulation accuracy, but it takes a long time and the efficiency is low; the empirical method is that the engineer directly estimates the grain size based on previous engineering experience, predicts the time period, and has high efficiency, but this method requires the engineer to be experienced, even if In this way, the prediction accuracy is not high; the BP neural network is easy to fall into the local optimal solution, and the network structure is not easy to determine; the parameters of the support vector machine are difficult to determine; the original extreme learning machine has few parameter settings, fast learning speed, and good generalization performance advantages, but the algorithm randomly generates the weight matrix between the input layer and the hidden layer and the hidden layer offset. The initial weight and threshold parameters have a great influence on the result, and it is not easy to obtain the optimal ELM model for prediction. The experimental method is Carry out actual production verification on each set of process sample data, this method is not only extremely inefficient, but also expensive

Images