煤矿瓦斯爆炸灾害风险预测研究

发布时间：2018-04-28 06:47

本文选题：瓦斯爆炸 + 核主成分分析　；参考：《辽宁工程技术大学》2017年硕士论文

【摘要】：煤矿的开采挖掘属于危险性极高的领域,其中瓦斯爆炸所带来的危害是最大的,严重影响了国民经济的增长和社会的稳定运行。煤矿发生瓦斯爆炸的几率越大,其未来出现灾害的损失也越大。如何快速和准确地对瓦斯爆炸作出预测,一直是一个需要解决的问题。基于此,本文对煤矿瓦斯爆炸预测系统做了详细研究,通过瓦斯监测系统与模式识别理论相结合,提出了KPCA-FOA-SVM预测模型,并且对模型进行了仿真。由于煤矿瓦斯爆炸监控系统具有非线性,通常所用的控制理论很难满足瓦斯爆炸风险预测系统的要求,因此本文建立果蝇算法对支持向量机寻优的网络,将此网络应用到预测系统中。鉴于引起煤矿瓦斯爆炸的因素较多,而每个因素的影响程度均不同,利用核主成分分析对致爆炸的指标做特征提取,精简特征维数。因为支持向量机预测模型的参数会影响模型的收敛性和泛化能力,因此本文采用果蝇算法全局寻优,搜索模型最佳参数组合,从而提高预测系统的识别精度,使瓦斯爆炸监控系统的可靠性得以提高。通过搭建瓦斯爆炸预测模型,建立实时分析各影响指标的监控系统,同时对FOA-SVM模型在瓦斯爆炸监控系统中做了仿真分析。仿真表明,文中采用的模型可以对煤矿瓦斯爆炸作出预警,且准确性和可靠性较好。论文的研究是基于国家自然科学基金项目(51274118)展开研究的,将模式识别、煤矿安全等多学科进行结合对煤矿瓦斯爆炸灾害展开研究,具有一定的理论研究价值与实际应用意义。
[Abstract]:The mining and mining of coal mine belongs to the extremely dangerous field, among which the gas explosion brings the greatest harm, which seriously affects the growth of the national economy and the stable operation of the society. The greater the probability of gas explosion in coal mine, the greater the loss of disaster in the future. How to predict gas explosion quickly and accurately has always been a problem to be solved. Based on this, this paper makes a detailed study on the coal mine gas explosion prediction system. Through the combination of gas monitoring system and pattern recognition theory, the KPCA-FOA-SVM prediction model is put forward, and the model is simulated. Because the monitoring system of coal mine gas explosion is nonlinear, it is difficult to satisfy the requirement of gas explosion risk prediction system by using the control theory, so this paper establishes the network of Drosophila algorithm to optimize support vector machine. The network is applied to the prediction system. In view of the fact that there are many factors that cause the gas explosion in coal mine, and the influence degree of each factor is different, we use the kernel principal component analysis (KPCA) to extract the feature of the explosive index and simplify the feature dimension. Because the parameters of the prediction model of support vector machine will affect the convergence and generalization ability of the model, this paper uses the Drosophila algorithm to search for the best parameter combination of the model, so as to improve the recognition accuracy of the prediction system. The reliability of the gas explosion monitoring system is improved. By building a gas explosion prediction model, a monitoring system for real-time analysis of each impact index is established. At the same time, the FOA-SVM model is simulated and analyzed in the gas explosion monitoring system. The simulation results show that the model adopted in this paper can make early warning of coal mine gas explosion, and the accuracy and reliability of the model are good. The research of this paper is based on the project of National Natural Science Foundation of China 51274118). It is of certain theoretical value and practical significance to study the gas explosion disaster in coal mine by combining multiple disciplines such as pattern recognition, coal mine safety and so on.
【学位授予单位】：辽宁工程技术大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD712.7;TP18

【参考文献】