为精准识别矿工疲劳程度，减少因疲劳引发的煤矿人因事故，提出了一种基于遗传算法(GA) 优化最小二乘支持向量机(LSSVM)的矿工疲劳程度识别模型。首先，通过疲劳诱发试验采集矿工心电数据，利用Friedman检验优选矿工疲劳程度的特征指标；然后，采用主成分分析法对选取的特征指标进行降维处理，建立表征矿工疲劳程度的特征集；在此基础上，利用遗传算法优化最小二乘支持向量机的关键参数，构建矿工疲劳程度识别模型。结果表明：选取的矿工疲劳程度特征指标能够有效反映矿工的疲劳程度；相较GA-SVM和LSSVM模型，融合GA-LSSVM模型可显著提高矿工疲劳程度的识别准确率(平均识别准确率为96.87%)。构建的矿工疲劳程度识别模型可较为高效地识别矿工的疲劳程度，对煤矿人因事故的防控具有一定的现实指导意义。

In order to accurately identify the fatigue degree of miners and reduce the accidents caused by miners' fatigue, a recognition model of mine fatigue degree based on genetic algorithm (GA) optimized by the least squares support vector machine (LSSVM) was proposed. First, the ECG data of miners were collected through fatigue induction experiments, and Friedman test was used to optimize the characteristic indicators of miners' fatigue degree. Then, the principal component analysis(PCA) was used to reduce the dimension of the selected feature indexes, and the feature parameter set representing the fatigue degree of the miner was established. On this basis, the key parameters of the least squares support vector machine were optimized by genetic algorithm, and the fatigue degree recognition model of miners was constructed. The results show that the selected characteristic indexes of miners' fatigue degree can effectively reflect the miners' fatigue degree. Compared with GA-SVM and LSSVM models, the fusion of GA-LSSVM model can significantly improve the recognition accuracy of miners' fatigue degree (the average recognition accuracy is 96.87%). The fatigue degree recognition model constructed can identify the fatigue degree of miners more efficiently, and has certain practical guiding significance for the prevention and control of coal mine accidents.