力量训练之最佳功率负荷预测方法构建与评估  

Construction and Evaluation of Optimal Power Load Prediction Methods for Strength Training

作　　者：梁美富 李妍 牛雪松[3] LIANG Meifu;LI Yan;NIU Xuesong(China Institute of Sport Science,Beijing 100061,China;Sports Department of Harbin University of Technology,Harbin 150001,Heilongjiang,China;School of Social Sports,Shenyang Sport University,Shenyang 110102,Liaoning,China)

机构地区：[1]国家体育总局体育科学研究所,北京100061 [2]哈尔滨工业大学体育部,黑龙江哈尔滨150001 [3]沈阳体育学院社会体育学院,辽宁沈阳110102

出　　处：《沈阳体育学院学报》2020年第5期108-116,共9页Journal of Shenyang Sport University

基　　金：国家社会科学基金项目(19BTY100)。

摘　　要：目的:构建不同最佳功率负荷预测方法,并横向评估其预测精度,为竞技体育高水平力量训练负荷"剂量化"提供理论与实践支撑。方法:招募52名受试者(测试对象46人,预测对象6人)进行最大力量测试和最大输出功率测试,根据测试对象的数据构建平均速度预测法、非线性回归模型预测法和BP神经网络模型预测法,利用预测对象数据对3种预测方法的预测精度进行对比分析。结果:卧推抛预测精度比较,平均速度预测法相对误差均值为25%,非线性回归模型预测法为31%,BP神经网络模型预测法为9%;半蹲起预测精度比较,平均速度预测法相对误差均值为13%,非线性回归模型预测法为20%,BP神经网络模型预测法为9%。结论:BP神经网络预测法预测精度和智能化程度最高,但其实际操作和应用的难度亦相对较大;平均速度预测法运用的便捷性最优,但其对设备的要求相对较高;非线性回归模型预测法对设备要求最低,但其预测误差相对较大。Purpose:To provide theoretical and practical support for the "dose" of high-level strength training load in competitive sports by constructing different optimal power load prediction methods and comparing their accuracy.Methods:52 subjects(46 test subjects and 6 predicted subjects) were recruited for maximum strength test and maximum output power test.According to the data of the test subjects,the average speed prediction method,the nonlinear regression model prediction method and the BP neural network model prediction method were constructed,and the prediction accuracy of the three prediction methods were compared and analyzed by using the data of the predicted subjects.Results:In the comparison of the prediction accuracy of bench press throw,the mean relative error of the average speed prediction method is 25%,that of the nonlinear regression model prediction method is 31%,and that of the BP neural network model prediction method is 9%.In the comparison of the prediction accuracy of half squat,the mean relative error of the average speed prediction method is 13%,that of the non-linear regression model prediction method is 20%,and that of the BP neural network model prediction method is 9%.Conclusion:The BP neural network prediction method has the highest accuracy and intelligence degree,but its practical operation and application are relatively difficult;the average speed prediction method has the best convenience,but its requirements for equipment are relatively high;the nonlinear regression model prediction method has the lowest requirements for equipment,but its prediction error is relatively large.

关 键 词：体能训练 力量训练 最佳负荷 预测方法 功率 

分 类 号：G804.2[文化科学—运动人体科学]