Research on sports load prediction method based on Aco-Bp algorithm

Abstract

Artificial Neural Network (ANN) belongs to a new type of information processing system. With the gradual expansion of the application scope of ANN, it is not only involved in environmental prediction, industrial production and other aspects, but also able to complete the work of disease diagnosis and data classification and processing. In this environment, BP neural network model as part of the neural network model, the influence in various industries is also becoming more and more obvious. Due to the slow convergence speed and insufficient data processing efficiency during the application of BP neural network, there is uncertainty in the analysis of part of the data, and the scope of application is limited. Ant colony optimization (ACO) algorithm, as a bionic algorithm, has shown obvious results in dealing with complex problems. In view of this, this paper collects and organizes the relevant data of sports athletes’ physiological indicators as research samples, and uses the ACO-BP algorithm to predict the sports mobilization load. Firstly, the weights of the neural network are optimized by ACO to avoid the deviation of the prediction results caused by the limitation of data processing of BP neural network. Finally, the ACO-BP algorithm, the standard BP algorithm and the improved BP algorithm with momentum term are compared. The results show that the mean square error values obtained by the ACO-BP algorithm are smaller in N = 100, 500 and 1000, which also indicates that the algorithm in this paper can predict the sports load more accurately.