摘要
为探究联合收获机驾驶员颈部肌肉疲劳的演化过程,实现颈部肌肉疲劳的客观检测与度量,基于表面肌电信号对联合收获机驾驶员颈部肌肉疲劳的识别问题进行研究。采集了120 min收获驾驶中10名联合收获机驾驶员颈部左右中斜角肌的表面肌电信号,探究颈部中斜角肌表面肌电信号的iEMG、RMS、MPF、MF和样本熵随驾驶时间的变化规律,基于支持向量机构建了联合收获机驾驶员颈部疲劳状态识别模型。结果表明:iEMG和RMS在驾驶任务后显著上升(P<0.05),而MPF、MF和样本熵在驾驶任务后显著下降(P<0.05);通过联合时频分析方法研究发现,联合收获机驾驶员在120 min收获驾驶任务后颈部中斜角肌产生疲劳,右侧肌肉产生的疲劳程度更高;联合收获机驾驶员颈部疲劳状态识别模型的识别正确率为89.91%,识别模型性能良好,可有效识别联合收获机驾驶员的颈部肌肉疲劳状态。
To explore the evolution process of neck muscle fatigue for combine harvester driver and realize the objective detection and measurement of neck muscle fatigue,the identification of neck muscle fatigue for combine harvester driver was studied based on surface electromyography. The surface electromyographic signals of the left and right middle scalene muscle of 10 combine harvester drivers were collected in the process of 120 min harvest driving. The change rule of iEMG,RMS,MPF,MF and sample entropy of the surface electromyographic signals for middle scalene muscle was analyzed with driving time.Based on support vector machine,the neck fatigue state recognition model of combine harvester was established. The results demonstrate that iEMG and RMS increase significantly after harvest driving task(P<0.05),while MPF,MF and sample entropy decrease significantly after harvest driving task(P<0.05).Through the joint analysis of EMG spectrum and amplitude(JASA),it is found that the middle scalene muscle of combine harvester driver in the neck has fatigue after 120 min harvest driving task,and the fatigue degree of the right middle scalene muscle is higher.The average recognition accuracy rate of neck fatigue state recognition model is 91.75%,which shows that the model has high-precise recognition and can effectively identify the combine harvester driver’s neck fatigue status.
作者
祝荣欣
ZHU Rongxin(Guangxi Aviation Logistics Research Center,Guilin University of Aerospace Technology,Guilin 541004,China)
出处
《工业工程与管理》
CSSCI
北大核心
2020年第5期138-144,共7页
Industrial Engineering and Management
基金
国家自然科学基金资助项目(61762045)
广西高校中青年教师基础能力提升项目(2018KY0661)。