煤与瓦斯突出事故预警研究

发布时间：2018-04-20 06:45

本文选题：煤与瓦斯突出 + 预警　；参考：《辽宁工程技术大学》2015年博士论文

【摘要】：针对开采深度的不断增加,瓦斯含量、瓦斯压力与煤层地应力增大,以及环境、管理等诸多方面因素而导致重大瓦斯事故数量增加所造成的巨大人身伤亡及企业财产损失问题,本课题探索了煤与瓦斯突出事故预警方法,在研究了煤与瓦斯突出事故形成机理以及对煤与瓦斯突出因素风险评价基础上,建立了合理的预警评价指标体系,并构建了具有较高实用价值与预测精度的煤与瓦斯突出预警模型,研究了煤与瓦斯突出事故严重程度评估方法与事故分级标准,并制定了预警准则,提出了煤与瓦斯突出事故应急相应救援管理流程。论文主要的研究内容有以下几个方面：1.构造了事故树并采用关联分析对煤与瓦斯突出事故进行了评价,提出了基于混沌粒子群优化RBF的可拓综合评价法来进行煤与瓦斯突出可拓综合评价,再用基于混沌粒子群优化的RBF神经网络进行权重的确定。在此基础上建立了预警指标体系。2.构建了主成分分析-响应面法预警模型。首先采用主成分分析法构造出少数的影响特征指标以代替原来繁冗的多数指标,再运用响应面方法建立一种可判断多种影响因素共存时回采工作面绝对瓦斯涌出量动态预测预警模型。所建模型基本能够真实地刻画矿井煤与瓦斯瓦斯突出影响因素与突出风险实际水平之间的复杂非线性关系,因此,具有较高的应用价值。3.构建了合理遗忘选择训练样本的时序动态预警模型。为了提高时序瓦斯涌出量的预测精度,针对最小二乘支持向量机的训练样本选择问题,本研究中引入了遗忘因子概率,即对历史数据加上遗忘因子,提出一种基于合理遗忘训练样本的瓦斯涌出量时序动态预测预警模型。该模型既考虑了历史数据的影响,又突出了新数据的作用,预测仿真实验分析证明该模型具有较高的精度。4.对煤与瓦斯突出事故严重程度评估方法与事故分级标准进行了研究。本课题结合应急管理理论方法,对煤与瓦斯突出预警风险进行相应的应对对策分析,并制定了预警准则,提出煤与瓦斯突出事故应急相应救援管理流程。为煤与瓦斯突出预警的有效实施提供依据。
[Abstract]:In view of the increasing of mining depth, gas content, gas pressure and ground stress of coal seam, and environment, management and so on, the increase in the number of major gas accidents leads to huge personal casualties and enterprise property losses. Based on the study of the formation mechanism of coal and gas outburst and the risk evaluation of coal and gas outburst factors, a reasonable early warning evaluation index system is established. The early warning model of coal and gas outburst with high practical value and prediction precision is constructed. The evaluation method and classification standard of accident severity of coal and gas outburst are studied, and the early warning criteria are established. The corresponding rescue management flow of coal and gas outburst accident is put forward. The main contents of this paper are as follows: 1. The accident tree is constructed and the correlation analysis is used to evaluate the coal and gas outburst accidents. The extension comprehensive evaluation method based on chaos particle swarm optimization (RBF) is proposed to evaluate coal and gas outburst comprehensively. Then the RBF neural network based on chaotic particle swarm optimization is used to determine the weight. On this basis, the early warning index system. 2. 2. The principal component analysis (PCA)-response surface method (RSM) early warning model was constructed. First of all, the principal component analysis method is used to construct a few characteristic indicators to replace the original redundant ones. The response surface method is used to establish a dynamic prediction and early warning model of absolute gas emission in mining face when many influencing factors coexist. The model can describe the complex nonlinear relationship between the influence factors of coal and gas outburst and the actual level of outburst risk, so it has high application value. A time series dynamic early warning model for selecting training samples for forgetfulness is constructed. In order to improve the prediction accuracy of gas emission in time series, aiming at the problem of training sample selection of least squares support vector machine, the probability of forgetting factor is introduced in this study, that is, adding forgetting factor to historical data. A dynamic prediction and early warning model of gas emission time series based on reasonable forgetting training samples is proposed. The model not only considers the influence of historical data, but also highlights the role of new data. The simulation results show that the model has a high precision. The evaluation method and classification standard of coal and gas outburst accident severity are studied. Combining with the theory and method of emergency management, this paper analyzes the corresponding countermeasures of coal and gas outburst warning risk, and formulates the early warning criterion, and puts forward the corresponding rescue management flow of coal and gas outburst accident. To provide the basis for the effective implementation of coal and gas outburst warning.
【学位授予单位】：辽宁工程技术大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TD713

【相似文献】