基于ISSA-ELM的煤与瓦斯突出危险等级预测  

Prediction of coal and gas outburst risk level based on ISSA-ELM

作　　者：邵良杉 毕圣昊[2] 王彦彬 赵硕嫱[2] SHAO Liangshan;BI Shenghao;WANG Yanbin;ZHAO Shuoqiang(Liaoning Institute of Science and Engineering,Jinzhou Liaoning 121010,China;School of Business Administration,Liaoning Technical University,Huludao Liaoning 125105,China)

机构地区：[1]辽宁理工学院,辽宁锦州121010 [2]辽宁工程技术大学工商管理学院,辽宁葫芦岛125105

出　　处：《中国安全生产科学技术》2023年第9期76-82,共7页Journal of Safety Science and Technology

基　　金：国家自然科学基金项目(71771111);辽宁省教育厅高校科研项目(LJKZ0359);葫芦岛市哲学社会科学研究课题(HLDSKY2023045)。

摘　　要：为提高煤与瓦斯突出危险等级预测的准确性,提出改进麻雀搜索算法(ISSA)优化极限学习机(ELM)的煤与瓦斯突出预测模型。首先,选用60组煤与瓦斯突出数据作为数据样本,采用主成分分析法(PCA)对其影响因素进行降维处理。然后,利用ISSA算法优化ELM算法的参数,建立ISSA-ELM模型。最后,选取样本后15组作为测试样本来验证模型的有效性,并与其他模型进行对比。研究结果表明:ISSA-ELM模型具有预测准确率更高、收敛速度更快和稳定性更佳等优点。研究结果可为煤与瓦斯突出危险等级准确判别提供参考。To improve the accuracy of coal and gas outburst risk level prediction,a coal and gas outburst prediction model with improved sparrow search algorithm(ISSA)optimized extreme learning machine(ELM)was proposed.Firstly,60 groups of coal and gas outburst data were selected as data samples,and the dimensionality reduction processing of their influencing factors was conducted by principal component analysis(PCA).Then,the parameters of ELM algorithm were optimized by ISSA algorithm,and the ISSA-ELM model was established.Finally,the last 15 groups of samples were selected as test samples to verify the validity of the model,and it was compared with other models.The results show that the ISSA-ELM model has the advantages of higher prediction accuracy,faster convergence,and better stability.The research results can provide a reference for the accurate determination of coal and gas outburst risk level.

关 键 词：矿山安全 煤与瓦斯突出预测 主成分分析法 改进麻雀搜索算法 极限学习机 

分 类 号：X936[环境科学与工程—安全科学]