Sports Training Model Based on GA Optimized Neural Network

With the continuous improvement of neural network and the combination of other new algorithms, its application field is more and more extensive. In this paper, a dynamic correction method of static load model parameters based on genetic algorithm is used to identify the constant parameters of power function of moving load model by genetic algorithm, and to modify the parameters of load model in real time according to load changes. The global convergence of the optimal preservation strategy is utilized and the diversity of the population is maintained. At the same time, the weight of the neural network is adjusted by the improved genetic algorithm to form a new improved genetic algorithm to optimize the BP neural network. Comparing with GA-BP network with selection operator as fitness proportional selection operator, the results show that the improved algorithm shortens the convergence time and reduces the operation error. Finally, using the literature data method to construct the physical characteristics and qualities that excellent middle distance running should possess, as the basis of training practice, combined with physical fitness, skills and combat energy training, a training model of middle and long distance running is established.