Application of Improved Genetic Algorithm in Optimization System of Mathematical Modeling Competition

With the high development of high-tech and productivity, people's life has brought earth-shaking changes. Improving the core literacy of mathematical modeling of Chinese students has become an important training goal at present, and it is also an important object for my country to implement the core literacy of mathematics. In order to solve the key problem of cultivating the core literacy of mathematics among high school students in our country, this paper makes a related study on how to shape the core literacy concept of mathematical modeling in students. The teaching of core literacy points out the direction, and puts forward suggestions for the formation of the application of the optimization system for mathematical modeling competitions. In the optimization process of mathematical modeling competitions, teachers should guide students to connect with the reality of life, analyze problems, build simple models and solve problems. To develop students' applied mathematics ability. In this paper, the optimized BP neural network algorithm based on the improved genetic algorithm is chosen as the modeling and forecasting method. Because BP neural network has strong nonlinear mapping ability, it can effectively model nonlinear systems. The final results of the study show that the absolute error of the GA-BP algorithm is also smaller than the other two algorithms, which is 0.689, and the ASSGA algorithm and the BP algorithm are 1.563 and 0.782, respectively. It can be seen that the GA-BP model has three advantages in RMSE, MAE, and MAPE. The indicators show strong superiority.