基于心率变异性的联合收割机驾驶员疲劳分析与评价

Analysis and evaluation of combine harvester driver fatigue based on heart rate variability

为探究联合收割机驾驶员的疲劳变化规律,应用RM6240C多通道生理信号采集系统,在约翰迪尔S660型联合收割机上进行了驾驶疲劳监测试验,采集了10名驾驶员120 min收获驾驶的心电数据。选取非线性动力学指标样本熵作为疲劳监测的特征参数,分析样本熵随驾驶时间的变化规律,确定驾驶疲劳发生时间段,对比不同作业环节的疲劳程度。结果表明:样本熵值随驾驶时间的增加呈下降趋势;样本熵值与主观驾驶疲劳程度的皮尔逊相关系数为-0.824,两者显著负相关;根据样本熵值判定,驾驶疲劳于50 min后开始出现,100 min后疲劳程度加深;转向行驶阶段比直线行驶阶段的驾驶疲劳程度高。基于样本熵的驾驶疲劳判定方法可客观的反映联合收割机驾驶员的体力和精神疲劳状况。

The study on combine harvester driver fatigue is important and necessary to reduce the accidents, improve the operation efficiency and protect the health of the driver. In order to explore the change rule of combine harvester driver fatigue, monitoring experiment of combine harvester driver fatigue was carried out with John Deere S660 at Gegiushan farm of Bei＇an Agricultural Reclamation Administration in Heilongjiang province from October 1, 2014 to October 7, 2014. The experiment was performed in sunny day during the forenoon to eliminate the influences of time and weather on the experiment. The crops harvested were soybean, and the conditions of test land were similar. The noise of cab was 95 d B（A）,of which temperature basically remain unchanged. The monitoring equipment was RM-6240 C multi-channel physiological signal acquisition processing system produced by Chengdu Instrument Factory with four channels and one interface of 12 lead ECG, which is suitable for multi-channel synchronous detection, records and analysis of human body physiological signal such as Electrocardiogram（ECG）, blood pressure, muscle tension. Before the test, skin preparation work was carried out such as removing dead skin, oil and grease. ECG signals were measured by three electrodes method; The electrodes were pasted on three places, for instance between the fourth rib on the left armpit front, below the right clavicle middle and the lower right of xiphoid process, which were connected with the positive（red）, the negative（green） and the reference（black） wire respectively. The sampling frequency of multi-channel physiological signal acquisition system was 1 k Hz,scanning speed 0.2 cm/s, sensitivity 1 m V. The ECG data of 10 male drivers sitting quietly in the cab were recorded for 5minutes before harvesting（marked as quiet segment）, at the same time subjective fatigue questionnaire were finished. Then the ECG data of drivers were recorded for 120 minutes when combine harvester running at the speed of 8 ~10 km/h.Subjective fatigue questionnaire were filled in every 20 minutes. The ECG data collected in driving were divided into 12 parts with 10 minutes per part. The ECG data both of quiet segment and 12 parts were denoised and detected for R waveform by the way of Wavelet Transform, and then the R-R interval value of each part was computed. Nonlinear dynamic index Samp En was selected as the characteristic parameter of fatigue testing which characterizes the complexity of heart rate variability. Firstly, the change curve of Samp En along with driving time and the scores of subjective fatigue degree at specified moment were achieved, and correlation analysis was researched between Samp En and scores of subjective fatigue degree. Secondly, driver fatigue occurred time was determined by the results of paired-samples T test of Samp En between quiet segment and other 12 parts. Finally, degrees of fatigue in straight section and that of turn section were compared by the results of paired-samples T test of Samp En between each section and quiet segment respectively. The results showed that the average values of Samp En significantly declined with the increase of the driving time. Pearson correlation coefficient between Samp En and subjective fatigue score was-0.824, which showed that their relationship was negatively significant. According to the results of paired-samples T test of Samp En between quiet segment and other 12 parts, the values of Samp En of the fifth part was significantly different from that of quiet segment（P0.05）, and the values of Samp En of tenth part was very significantly different from that of quiet segment（P 0.01）, which indicated that combine harvester driver fatigue began to appear after 50 minutes, and deeped after 100 minutes. The values of Samp En in turn section was significantly different from that of quiet segment（P0.05）, there was not significant difference between straight section and quiet segment（P0.05）, and the values of Samp En in turn section was smaller than that of straight section, which indicated that degree of fatigue of the former was higher than that of the latter. Compared with the subjective evaluation method of driver fatigue, determining diver fatigue method according to the change of the value of Samp En can more accurately reflect the beginning and deepening period of combine harvester driver fatigue, and objectively reflect the driver＇s physical and mental fatigue status.