Application And Research Of Ant Colony Network For Forecast Of Sport Load

Artificial neural network (ANN) is a nonlinear dynamic system.It is a new kind of intelligent information process system based on the development of the biological neural network. ANN has been applied broadly in many fields for its unique form of disposing of information, such as industry produce monitoring and controlling, stock analysis,classification. Back-propagation network, which has gotten broadly application,is representative network model in these different network models.But it has weaknesses such as inefficient, slow convergent speed and easy getting into local minimum,which restrict the neural network's application in all fields.Ant Colony Optimization (ACO) is a novel evolutionary algorithm,which has superiority of solving complicated combinatorial optimization problems especially the discrete optimization problems.ACO is a heuristics global optimization,it can be used for training neural network in order to avoid the defects of BP neural network.The research of the relation between physical index and biochemical index with sports training load is very important in the field of kinematics.Sports training load analysis is a typical uncertain and nonlinear problem because of the complex relationship among the physical index and biochemical index.The tennis of China is in development stage. In order to enhance the competitive strength of China's tennis and narrow the gap with the high-level athletes from the other country,the national tennis team must process scientific training.According to the characteristic of tennis,the paper provides a new method.In this paper the data of athletes'physical index and biochemical index are used as the training samples.And a new neural network training algorithm,ACO-BP algorithm, is proposed to establish ant colony network model of getting corresponding relationship between physical index and biochemical index with sports training load. ACO-BP scheme adopts ACO to search the optimal combinations of weights in the solution space,then uses BP algorithm to obtain the accurate optimal solutions quickly. The method has been proved to be valid. The model for forecast of sport load is applied to the information platform of national tennis team to verify the validity and practicality of the model, and it plays a desired effect.