基于表面肌电信号的绝缘手套法带电作业人员上肢肌肉疲劳分析

绝缘手套法是配网带电作业的主要作业方式,为获取穿戴绝缘手套的带电作业人员上肢肌肉的疲劳特性及其诱发的肌肉骨骼系统疾患(work-related musculoskeletal disorders,WMSDs)的风险,开展了基于表面肌电信号(surface electromyogra-phy,sEMG)的绝缘手套法带电作业人员上肢肌肉疲劳评估研究.针对配网绝缘手套法的典型作业工况,搭建了绝缘手套法带电作业上肢的sEMG试验平台,采集受试者上肢(右手)各目标肌肉在是否穿戴绝缘手套下的sEMG信号;基于时域特征参数积分肌电值(integrated electromyography,IEMG)、均方根值(root mean square,RMS)及频域特征参数平均功率频率(mean power frequency,MPF)、中位频率(median frequency,MDF)对绝缘手套的上肢肌肉疲劳特征进行评估;基于支持向量机(support vector machine,SVM)构建了带电作业人员上肢肱桡肌疲劳状态识别模型.结果表明:穿戴绝缘手套作业时各目标肌肉更容易进入疲劳状态;穿戴绝缘手套作业时,作业人员上肢部位的肱桡肌、肱二头肌、肱三头肌、三角肌的疲劳程度依次递减,与仿真计算的分析结果一致;sEMG时域特征参数IEMG、RMS对作业人员上肢肌肉疲劳的表征效果要优于频域特征参数MPF和MDF;带电作业人员上肢肱桡肌疲劳状态识别模型总体平均准确率为86.56％,能有效识别上肢肱桡肌肌肉疲劳状态.

Analysis about Fatigue Characteristics of Upper Limb Muscles in Live Working by Insulating Gloves Method Based on Surface Electromyography

A study on the fatigue assessment of upper limb muscles of live working operators by insulated glove method based on sEMG was carried out. The live working operation test with insulated gloves was designed and carried out to collect the target muscles sEMG of the subject''s upper limb (right hand);The time domain characteristic parameter: IEMG, RMS, and frequency domain characteristic parameters: MPF, MDF were analyzed; Based on support vector machine(SVM), a fatigue identification model of brachioradialis muscle of upper limb was established. The results show that the target muscles are more likely to enter the fatigue state when wearing insulating gloves. When working with insulating gloves, the fatigue degree of brachioradialis, biceps brachialis, triceps brachialis and deltoid muscle of the upper limb of the operator decreases successively, which is consistent with the analysis result of simulation calculation. sEMG time-domain characteristic parameters IEMG and RMS have a better characterization effect on upper limb muscle fatigue than frequency-domain characteristic parameters MPF and MDF. The overall average accuracy of the fatigue state recognition model of brachioradialis of the upper limb is 86.56%, which can effectively identify the fatigue state of brachioradialis of the upper limb.