Author(s)

Jack Mo, Hao He, Eric Zhang

Corresponding Author

Jack Mo

ABSTRACT

To make the most scientific arrangement of the physical strength of the cyclists in the race, our team has developed a model that determines the relationship between the rider's position on the course and the power the rider applies. In the paper, we describe in detail the process of our model from the initial idea to the final use, including three times of modification and optimization. At first, we started from the ideal environment and obtained the prototype of the model by simulating the rider's constant movement on the flat and straight track with no wind resistance, and added the influence of the rider's sprint based on uniform speed to obtain the initial power curve. We built a function model of time T on power and change distance using many physical and dynamic principles by constructing the function relationship between power and time and change distance. Then, we obtained the optimal solution through a genetic optimization algorithm. Next, we develop progressively optimized models based on the rudimentary model to make our model more closely match the power output of the actual race. In Model optimization 1, We add the influence of athletes' physical exertion on riding by introducing a punish coefficient which logistic model and standardization ideas guide, based on data. This coefficient will have a constraint on the athletes' real-time power. In Model optimization 2, We pay attention to the velocity component of wind along the direction of the athlete's speed. We add the influence of it by force analysis. This model clearly shows that when the velocity component is the same as the direction of the athletes' speed, the minimum time will be smaller, and the ideal power will be smaller. If not, it will have the opposite effect. In Model optimization 3, we add the influence of bicycle steering on velocity by a limit for maximum speed. There is a relationship between the velocity of the athlete and the tilt angle. So, we explore it by differential equation modeling. In a word, our model considers the differences in areas of expertise and physical fitness among cyclists of different types and genders. We define the power profile at the model preparation stage. After the sensitivity analysis test, our model has strong robustness and accurate simulation results.

KEYWORDS

Power profile, Genetic optimization algorithm, Differential equation modeling, Minimum race time, Most ideal power output