基于边际谱熵的肌肉疲劳实时评估方法研究

发布时间：2018-10-09 14:14

【摘要】：肌肉疲劳是一种复杂的生理现象。针对利用表面肌电信号实时评估肌肉疲劳,要求疲劳指标兼具快速、可靠、抗噪的问题,提出基于边际谱熵的肌肉疲劳实时评估方法。首先,利用不同数据长度的确定性周期信号和高斯白噪声分析了边际谱熵快速性与数据长度稳健性;其次,利用10名受试者握力持续静态收缩状态下从100%MVC下降到50%MVC时桡侧腕长伸肌的肌肉疲劳信号,分析了边际谱熵评估肌肉疲劳的可靠性与应用于不同个体的稳定性;最后,在某一受试者肌肉疲劳信号中加入高斯白噪声和心电噪声考察了边际谱熵的抗噪性。实验结果表明,边际谱熵与近似熵和中值频率相比计算快速,数据长度稳健性更优;线性拟合优度较佳(0.46±0.14),能可靠地评估肌肉疲劳;斜率变异系数较低(30.30%),对不同个体稳定性高;加入高斯白噪声和心电噪声后边际谱熵拟合优度变化率较低(分别为34.39%和3.78%),具有良好的抗噪性。因此边际谱熵兼具快速、能可靠评估肌肉疲劳以及抗噪等优点,为实时评估肌肉疲劳提供一种新方法。
[Abstract]:Muscle fatigue is a complex physiological phenomenon. Aiming at the problem of real-time evaluation of muscle fatigue by surface electromyography (EMG) signals, which requires both fast, reliable and anti-noise indexes, a real-time evaluation method of muscle fatigue based on marginal spectrum entropy is proposed. Firstly, using deterministic periodic signals with different data lengths and Gao Si white noise, the paper analyzes the marginal spectral entropy rapidity and data length robustness. Using the muscle fatigue signal of extensor Carpi radialis longus under the condition of continuous static contraction of grip force from 100%MVC to 50%MVC, the marginal spectrum entropy was analyzed to evaluate the reliability of muscle fatigue and the stability of muscle applied to different individuals. The noise resistance of marginal spectral entropy was investigated by adding Gao Si white noise and electrocardiogram noise to the muscle fatigue signal. The experimental results show that the marginal spectral entropy is faster than the approximate entropy and the median frequency, the data length is more robust, the linear fit is better (0.46 卤0.14), the slope coefficient of variation is lower (30.30%), and the slope coefficient of variation is higher for different individuals. When Gao Si white noise and ECG noise were added, the variation rate of marginal spectral entropy goodness of fit was lower (34.39% and 3.78%, respectively), and had good noise resistance. Therefore, the marginal spectral entropy has the advantages of fast, reliable evaluation of muscle fatigue and anti-noise, which provides a new method for real-time evaluation of muscle fatigue.
【作者单位】：合肥工业大学智能制造技术研究院;合肥工业大学机械工程学院;
【基金】：科技型中小企业技术创新基金(11C26213402042)项目资助
【分类号】：R318;TN911.7

【相似文献】