【摘要】：對(duì)長(zhǎng)跑中運(yùn)動(dòng)員耐力進(jìn)行準(zhǔn)確預(yù)測(cè),可以對(duì)運(yùn)動(dòng)員的身體變化進(jìn)行合理分析,提出科學(xué)合理的訓(xùn)練指導(dǎo)意見。進(jìn)行耐力預(yù)測(cè)時(shí),需要考慮跑步中的加速度、天氣等自變量,在這些自變量影響下提取運(yùn)動(dòng)員的運(yùn)動(dòng)負(fù)荷特征,以此為基礎(chǔ),建立耐力預(yù)測(cè)模型,但是傳統(tǒng)方法很難有效表達(dá)這些因素產(chǎn)生的動(dòng)態(tài)影響,無法提取準(zhǔn)確的運(yùn)動(dòng)負(fù)荷特征,存在建模誤差大的問題。提出基于改進(jìn)人工魚群神經(jīng)網(wǎng)絡(luò)算法的長(zhǎng)跑中運(yùn)動(dòng)員耐力預(yù)測(cè)建模方法。先利用BP神經(jīng)網(wǎng)絡(luò)算法獲取運(yùn)動(dòng)員在長(zhǎng)跑中的踝關(guān)節(jié)加速度綜合指數(shù),利用計(jì)算的結(jié)果定義能耗預(yù)測(cè)中的自變量,提取長(zhǎng)跑中運(yùn)動(dòng)員的運(yùn)動(dòng)負(fù)荷特征,利用人工魚群算法對(duì)BP神經(jīng)網(wǎng)絡(luò)的初值權(quán)值和閾值進(jìn)行優(yōu)化,組建運(yùn)動(dòng)員耐力預(yù)測(cè)方程,利用支持向量機(jī)的默認(rèn)RBF核函數(shù)搜索最佳的懲罰因子,利用搜索的結(jié)果對(duì)預(yù)測(cè)方差進(jìn)行訓(xùn)練,獲取最優(yōu)的長(zhǎng)跑中運(yùn)動(dòng)員耐力預(yù)測(cè)模型。仿真結(jié)果表明,所提算法可以為長(zhǎng)跑中運(yùn)動(dòng)員耐力預(yù)測(cè)提供參考。
[Abstract]:The accurate prediction of athletes' endurance in long-distance running can reasonably analyze the changes of athletes' body and put forward scientific and reasonable training guidance. When carrying out endurance prediction, it is necessary to consider the acceleration, weather and other independent variables in running. Under the influence of these independent variables, the sports load characteristics of athletes should be extracted, and based on this, the endurance prediction model should be established. However, the traditional method is difficult to express the dynamic effect of these factors effectively, and can not extract the accurate motion load feature, which has the problem of large modeling error. In this paper, an improved artificial fish swarm neural network (Ann) algorithm is proposed to predict the endurance of athletes in long-distance running. Firstly, the BP neural network algorithm is used to obtain the ankle acceleration index of the athletes in the long-distance running, and the independent variables in the energy consumption prediction are defined by the calculated results, and the sports load characteristics of the athletes in the long-distance running are extracted. The initial value weight and threshold value of BP neural network are optimized by artificial fish swarm algorithm, and the endurance prediction equation of athletes is set up. The optimal penalty factor is searched by the default RBF kernel function of support vector machine. The prediction variance is trained by the search results, and the optimal endurance prediction model of long distance runners is obtained. The simulation results show that the proposed algorithm can be used as a reference for the endurance prediction of long-distance runners.
【作者單位】： 鄭州大學(xué)西亞斯國(guó)際學(xué)院;
【分類號(hào)】：G822.3

【相似文獻(xiàn)】

相關(guān)期刊論文 前7條

1 王炳彝;對(duì)運(yùn)動(dòng)成績(jī)預(yù)測(cè)模型選定過程的研究[J];體育與科學(xué);1993年02期

2 汪鴻,吳云飛;十項(xiàng)全能運(yùn)動(dòng)員成績(jī)多因素非均衡性預(yù)測(cè)模型的研究[J];山西師大體育學(xué)院學(xué)報(bào);2005年01期

3 師義民，，劉桂林，付剛，楊昭軍;體能預(yù)測(cè)模型[J];預(yù)測(cè);1994年05期

4 鄧美蘭;張珍;;奧運(yùn)會(huì)男子十項(xiàng)全能前三名成績(jī)的灰色預(yù)測(cè)模型的研究[J];數(shù)學(xué)的實(shí)踐與認(rèn)識(shí);2009年18期

5 張杰,葉東惠;體育預(yù)測(cè)成績(jī)的無差異性檢驗(yàn)分析研究[J];北京體育大學(xué)學(xué)報(bào);2004年07期

6 馬祥海;;我國(guó)優(yōu)秀十項(xiàng)全能運(yùn)動(dòng)員成績(jī)的灰色關(guān)聯(lián)分析及灰色GM(1,1)預(yù)測(cè)模型的建立[J];首都體育學(xué)院學(xué)報(bào);2012年06期

7 ;[J];;年期

相關(guān)重要報(bào)紙文章 前1條

1 南方日?qǐng)?bào)記者 彭琳　實(shí)習(xí)生 周鵬程;預(yù)測(cè)世界杯之外 大數(shù)據(jù)還能做什么[N];南方日?qǐng)?bào);2014年

相關(guān)碩士學(xué)位論文 前1條

1 馬婷婷;基于序列與多因素分析的奧運(yùn)會(huì)運(yùn)動(dòng)成績(jī)預(yù)測(cè)模型[D];南京信息工程大學(xué);2012年

本文編號(hào)：2459773