基于PSO-KELM的煤与瓦斯突出预测研究  

Research on coal and gas outburst prediction based on PSO-KELM

作　　者：王小生[1] 尹亚红[1] 涂军 张小健 杨晋 Wang Xiaosheng;Yin Yahong;Tu Jun;Zhang Xiaojian;Yang Jin(Jizhong Energy Co.,Ltd.,Xingtai 054000,China)

机构地区：[1]冀中能源股份有限公司,河北邢台054000

出　　处：《能源与环保》2025年第3期60-64,共5页CHINA ENERGY AND ENVIRONMENTAL PROTECTION

摘　　要：煤与瓦斯突出是煤矿开采过程中常见的一种地质灾害,为保障井下工作人员的生命安全和国民经济的稳定增长,融合智能优化算法和机器学习算法,以核极限学习机(KELM)作为基准预测模型,结合粒子群算法(PSO)优化KELM关键参数,规避了人为预设导致的性能缺陷,提高预测模型分类精度。结合现场实测数据,对指标预处理进行有效性分析,验证优化预处理后的模型精度较未处理有所提升;以预处理后的数据样本作为模型输入,对各基准预测模型进行对比,证明了KELM基准预测模型的稳定性和优越性;将PSO-KELM模型与其他常用模型进行30次预测对比实验。结果表明,PSO-KELM模型平均预测准确率达到86.33%,较其他模型具有更好的预测精度和更快的收敛速度,为煤与瓦斯突出预测工作提供了一种新的有效方法和理论支撑。Coal and gas outburst are a common geological disaster during coal mining.To ensure the safety of underground workers and the stable growth of the national economy,the intelligent optimization algorithms and machine learning algorithms were integrated,the kernel extreme learning machine(KELM)was used as the baseline prediction model,and the particle swarm optimization(PSO)algorithm was combined to optimize the key parameters of KELM,the performance defects caused by manual presetting were avoided,and the classification accuracy of the prediction model was improved.Combining field measured data,the effectiveness of the indicator preprocessing was analyzed first,verifying that the model accuracy after optimized preprocessing was improved compared to the unprocessed one;using the preprocessed data samples as model input,various baseline prediction models were compared,proving the stability and superiority of the KELM baseline prediction model;the PSO-KELM model was compared with other commonly used models in 30 prediction experiments,and the results show that the average prediction accuracy of the PSO-KELM model reaches 86.33%,which has better prediction accuracy and faster convergence speed than other models,providing a new effective method and theoretical support for the prediction of coal and gas outbursts.

关 键 词：PSO-KELM模型 煤与瓦斯突出灾害 风险预测粒子群算法 核极限学习机 

分 类 号：TD712[矿业工程—矿井通风与安全]