基于RBF神经网络的瓦斯测值分析及预测应用研究

发布时间：2018-07-08 20:03

本文选题：矿井瓦斯 + RBF神经网络　；参考：《西安科技大学》2013年硕士论文

【摘要】：瓦斯灾害作为我国煤矿的主要灾害之一，长期困扰着煤矿的安全生产，本文通过对瓦斯预测方法、预警技术调研分析的基础上，以反应瓦斯涌出特征的矿井瓦斯实时监测数据、防突监测数据为研究对象，通过将瓦斯涌出显现出的瓦斯浓度变化特征与瓦斯实时监测数据与防突监测数据之间的关联特征，用神经网络方法来描述，进行瓦斯浓度预测预警与煤与瓦斯突出危险性预测预警的研究，主要研究工作如下：首先介绍了径向基函数神经网络（RBFNN）方法的基本理论及其预测方法，在此基础上，，分析了其应用于煤矿实际监测数据分析的可行性，以及应用于矿井瓦斯预测的基本原理。其次，针对矿井实际监测数据的特征，使用插值法对实测数据进行预处理，建立基于综采工作面瓦斯实时监测数据处理的综采工作面瓦斯浓度预测预警模型，实现了回采工作面瓦斯浓度的实时预测预警。再次，针对检/监测数据，提取瓦斯实时监测数据的统计特征参数，建立了掘进工作面煤与瓦斯突出危险性预测预警模型，基于检/监测数据融合分析，实现煤与瓦斯突出危险性预测预警。最后，将预测预警模型应用于实例矿井进行现场分析验证，分析结果表明：预测的误差较小，预测结果较准确，从而保证了预警分析的可靠性。本文研究的以瓦斯检/监测数据处理为手段的瓦斯预测预警技术，针对现场实测数据的应用分析，表现出了良好的适用性，可作为扩展煤矿安全监测监控系统功能的有效手段，具有一定的实际应用价值。
[Abstract]:Gas disaster, as one of the main disasters in coal mine in China, has been puzzling the safety production of coal mine for a long time. Based on the investigation and analysis of gas prediction method and early warning technology, the real-time monitoring data of gas emission in coal mine are used to reflect the characteristics of gas emission. The outburst prevention monitoring data is used as the research object. The relationship between the gas concentration variation characteristics and the real-time gas monitoring data and the anti-outburst monitoring data is described by the neural network method. The main research work is as follows: firstly, the basic theory and prediction method of radial basis function neural network (RBFNN) are introduced. The feasibility of its application in coal mine monitoring data analysis and the basic principle of mine gas prediction are analyzed. Secondly, according to the characteristics of actual mine monitoring data, the interpolation method is used to preprocess the measured data, and a prediction and warning model of gas concentration in fully mechanized coal mining face based on real-time monitoring data processing of fully mechanized mining face is established. The real-time prediction and early warning of gas concentration in mining face are realized. Thirdly, aiming at the inspection / monitoring data, the statistical characteristic parameters of the real-time gas monitoring data are extracted, and the prediction and early warning model of coal and gas outburst risk in the tunneling face is established, which is based on the fusion analysis of the inspection / monitoring data. Coal and gas outburst risk prediction and early warning. Finally, the prediction and early warning model is applied to the field analysis and verification of an example mine. The analysis results show that the prediction error is small and the prediction result is more accurate, thus ensuring the reliability of the early warning analysis. The gas prediction and early warning technology based on gas inspection / monitoring data processing is studied in this paper. According to the application and analysis of field measured data, it shows good applicability and can be used as an effective means to expand the function of coal mine safety monitoring and monitoring system. It has certain practical application value.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TD712;TP183

【参考文献】