Sports Training Method Based on Multiobjective Optimization


  • Xianwu Wu , Jide Liu*, Dongming Zhao , Youqiang Liu


At present, most of the sports training methods adopt the recombination operator that is
designed for single objective optimization. Through the validation or experimental,
several typical single objective sports training methods are analyzed and proven that
they are not applicable for some multiobjective optimization problems. The
multiobjective optimization sports training method based on the mixture Kalman model
(Multiobjective evolutionary algorithm -based on decomposition and mixture Kalman
models, MOPE for short) is proposed. This algorithm firstly applies an improved
mixture Kalman model to carry out sampling to the group sports training and generate
new individuals, and then makes use the greedy strategy to update the group. In view of
the multiobjective optimization problem that is of complicated Pareto Front, the test
results show that for the majority of the given athletes, this training method can achieve
good effect.