基于SSA-SVM的煤矿瓦斯监测预警模型研究  

Research on Coal Mine Gas Monitoring and Early Warning Model Based on SSA-SVM

作　　者：张朝 冯锋 ZHANG Zhao;FENG Feng(School of Information Engineering,Ningxia University,Yinchuan 750021,China)

机构地区：[1]宁夏大学信息工程学院,宁夏银川750021

出　　处：《宁夏工程技术》2023年第4期359-365,共7页Ningxia Engineering Technology

基　　金：宁夏重点研发计划项目(2022BEG02016);宁夏自然科学基金重点项目(2021AAC02004)。

摘　　要：针对煤矿数据海量化、影响瓦斯浓度预警环境因素多等问题,提出了基于麻雀搜索算法(SSA)和支持向量机(SVM)的煤矿瓦斯监测预警模型。该模型首先利用物联网技术对生产环境数据进行收集和监测;其次利用SSA对支持向量机的惩罚因子和核参数进行寻优,建立SSA-SVM安全预警模型;最后训练了安全预警模型,并且对瓦斯安全等级进行预警分类。实验结果表明,SSA-SVM模型的预测准确率达到91.667%,能够为煤矿瓦斯安全预警工作提供有力支撑。Confronted with the substantial data volumes characteristic of coal mining operations and the complex array of environmental factors impinging upon methane pre-warning systems,a sophisticated gas monitoring and preemptive alerting model predicated on the Sparrow Search Algorithm(SSA)integrated with Support Vector Machine(SVM)has been advanced.Initially,Internet of Things frameworks facilitate the systematic aggregation and surveillance of production-related environmental datasets.Subsequently,the SSA is deployed to refine the penalty coefficient and kernel parameters intrinsic to the SVM,culminating in the formulation of a robust SSA-SVM safety prognosticative model.This model undergoes rigorous training and is instrumental in the categorization and early warning of methane hazard levels.Empirical analyses corroborate the model’s prognostic efficacy,evidencing a predictive accuracy of 91.667%,thus significantly bolstering the methane safety early warning mechanisms within coal mining infrastructures.

关 键 词：煤矿瓦斯 预警模型 麻雀搜索算法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]