Research on safety assessment of gas explosion hazard in heading face based on BP neural network

Introduction Gas explosion hazard is the most dangerous haz-ard and always the archenemy of safety production in coalmine, and it becomes the essential problem of safety production in coalmine at present. It is neces-sary to identify and assess gas explosion hazard be-cause gas accumulates frequently and it is apt to ex-plode in many places of coalmines especially in head-ing face. Statistically, gas explosion in heading face accounts for about 60%~70% of the total gas explo-sion accident[1]. …     

煤矿井下作业人员的工作压力个体因素与不安全行为的关系

以煤矿井下作业人员的个体因素的3个状态为外源变量,煤矿井下作业人员工作压力为中介变量,煤矿井下作业人员的不安全行为为内生变量,构建了"个体因素、工作压力与不安全行为的关系模型"并运用结构方程建模的方法进行了实证研究,实证结果表明个体因素、工作压力对不安全行为具有显著的影响作用。基于实证研究结果,从个体因素的3个状态和工作压力出发,提出了对减少煤矿井下作业人员不安全行为的策略和建议。     

基于三类危险源的掘进工作面瓦斯爆炸模糊综合安全评价

瓦斯事故多发是制约我国煤矿安全生产的最主要因素。依据三类危险源理论,对掘进工作面瓦斯爆炸影响因素进行了辨识、分析,在此基础上建立了基于三类危险源的掘进工作面瓦斯爆炸安全评价指标体系。结合河南某矿掘进工作面的实际情况,运用模糊综合评价法对建立的指标体系进行评价,得出其安全性等级为一般,验证了安全评价指标体系的合理性。     

基于险兆事件的煤与瓦斯突出灾害概率模型构建

任何一起煤与瓦斯突出灾害事故发生,都有征兆显现。为了防止煤与瓦斯突出灾害事故发生,以煤炭开采过程中的Near-miss为研究对象,阐述了煤与瓦斯突出Near-miss中的致突险兆因子及其作用机制,在现有灾害理论以及煤与瓦斯突出机理、预警技术研究的基础上,提出了煤与瓦斯突出灾害由4个要素组成,即致突险兆因子、承灾体、孕灾环境和相互作用形式,并进而提出了描述基于Near-miss的煤与瓦斯突出灾害的方法。     

基于三叉树定价模型的煤矿安全投资期权价值研究

针对传统的安全投资决策方法存在的缺陷,对传统的二叉树模型进行了拓展,以煤矿安全投资为例,在无风险利率为常数的情况下,构建了在煤矿资产自身价值有收益率和安全投资项目价值有期权价值的三叉树模型,并以实际数据进行了期权定价模拟,说明该方法的可靠性和模型的科学性,为决策者提供科学依据。     

也论矿井开采系统安全可靠性的评价

本文就矿山井采系统安全可靠性的评价给出了新的数学模型和计算方法，对发表在＜＜煤炭工程师＞＞１９９２年第１期上的矿井开系统安全可靠性评价初探一文进行了补充。     

The risk early-warning of gas hazard in coal mine based on Rough Set-neural network

This article proposed the risk early-warning model of gas hazard based on Rough Set and neural network. The attribute quantity was reduced by Rough Set, the main characteristic attributes were withdrawn, the complexity of neural network system and the computing time was reduced, as well. Because of fault-tolerant ability, parallel processing ability, anti-jamming ability and processing non-linear problem ability of neural network system, the methods of Rough Set and neural network were combined. The examples research indicate： applying Rough Set and BP neural network to the gas hazard risk early-warning coal mines in coal mine, the BPNN structure is greatly simplified, the network computation quantity is reduced and the convergence rate is speed up.     

煤自燃环境及其影响因素

在提出自然环境的基础上，探讨了影响浮煤自燃的特征因素，应用模糊群体决策法确定了各因素的重要度。为进一步深化自燃研究奠定了基础。