The Study Of Characteristics Of Energy Consumption And Establishment Of Predictive Equation For Power Cycling Exercise Of 20-30 Years People

To study the energy expenditure characteristics of power cycling in 20-30 years old adults under different intensity,and establish the predictive equation of their cycle energy expenditure.Methods:Selected 39 adults aged 20-30 with exercise habits.Subjects wore 3B-R2 gas metabolism analyzer and Polor heart rate belt,and completed cycling for 6min with 65rmp at three intensifies(30W,75W,125W),respectively.EE at three intensifies was calculated according to oxygen intake and carbon dioxide exhalation.The linear regression model was established by SPSS linear method with the data of 34 people,and the accuracy of the model was verified with the data of 5 people(3 males and 2 females),and the neural network model was established by using Matlab.BlandAltman scatter diagram was used to analyze the accuracy of model prediction.Mae,MSE and data prediction ability were used to compare the accuracy of different models in energy expenditure prediction.Results:1.With the increase of exercise intensity,the energy consumption in the EPOC stage and total energy consumption of both men and women increased significantly,but there was no significant difference between men and women2.With the increase of exercise intensity,the EPOC time increased significantly for men and women,but the proportion of energy consumption in the EPOC stage to the total energy consumption remained basically unchanged,and the proportion of fat energy supply in the EPOC stage was about 20%for women3.Regression model of energy expenditure:Male:Y=1.3-0.024*a-0.009*b+0.008*c+0.096*d,r=0.85 Female:Y=-1.709-0.005*a-0.011*b+0.060*c+0.063*d,r=0.85(Y for energy expenditure(kcal/min),a for body fat percentage,b for height(cm),c for weight(kg),d for ΔHR)BP neural network model:The input index of the model is%HR,body weight,body fat percentage,the output result is EE,using 4-6-1-1,maximum error=1e-7,learning rate=0.01,momentum=0.9,the results of EE prediction model are r=0.94,total mean square error=0.48,R2=0.86.BP neural network prediction model in R2,residual,MAE,MSE are higher than linear regression equation.Conclusions:1,In power cycling,the total energy consumption in the same period of time at the same exercise intensity does not have gender difference.In the case of power cycling at the intensity of 30W and 75 W,the energy consumption of EPOC stage increases with the increase of intensity for the same exercise time,but the proportion of total energy consumption does not change much.2.In the establishment of energy consumption prediction model of power bicycle,the BP neural network established by inputting%HR,weight,percentage of body fat,height and EE has high prediction accuracy.