Abstract: The accelerometer is one of the most commonly used tools for measuring physical activity and its energy expenditure,and with many energy expenditure algorithms for the user to select.The Actigraph(GT3X model) accelerometer analysis software has five energy consumption algorithms suitable for adults.The purpose of this study is to explore its suitability for monitoring energy consumption of physical activity of people aged 19~29 in China.The twenty participants(24±5 yr) including 14 male and 6 female,wore the GT3X accelerometer during walked or run,when they completed five minutes exercise respectively at 4 km/h,5.6 km/h,6.4 km/h,7.2 km/h and 8 km/h on a treadmill.The predicted energy expenditure(EE) from accelerometer were compared to indirect calorimetry,the validity of energy expenditure algorithms investigated by Pearson correlation,repeated-measures analyses of variance(ANOVA),Bland-Altman plots and paired t test so on.The results showed that energy expenditure algorithms is significant correlation with indirect calorimetry,R is 0.66~0.97(P≤0.01),but most of them underestimate the value of energy expenditure.in addition to the Williams work-energy equation,other algorithms have a more stable system error based on paired t test.In conclusion,the Freedson VM3 Combination(2011) algorithm is the most effective on predicting energy consumption of physical activity for young adults in China.