基于小波分析与ARMA组合模型的矿压显现预报研究

发布时间：2018-06-15 23:14

本文选题：小波分析 + ARMA建模　；参考：《西安科技大学》2015年硕士论文

【摘要】：煤矿生产中,矿山压力周期性地作用于回采工作面,对煤矿安全生产造成一定影响,并有可能引发煤矿事故。为此,回采工作面来压显现及其强度预测研究对促进煤矿安全生产具有重要理论与现实意义。本文在分析国内外相关研究综述基础上,通过文献分析、理论分析与实地调研,综合运用矿压显现理论、统计分析、小波分析、时间序列分析、定性与定量分析相结合的研究方法对矿压显现预报进行系统研究,主要工作和结论如下:(1)分析了矿压显现预报研究现状,针对目前以矿压理论为基础的力学方法应用最为广泛、概率统计方法也在来压判据、动载系数等方面有所应用的现状,本文采用研究非线性问题的新兴智能技术方法,目前该方法在矿压显现预报方面应用较少。(2)以南梁20307工作面煤层为实例,根据数据统计结果,得出如下结论:该工作面周期来压判据为24.4MPa;直接顶来压强度为24.86MPa;直接顶来压步距为10.36m;工作面老顶初次来压强度为31.68MPa;老顶周期来压步距为22.5m,来压强度为37.1MPa。(3)在分析顶板来压影响因素基础上,构建了小波去噪与ARMA的组合模型以及Mallat算法与ARMA的组合模型,分析了这两种组合模型在矿压显现预报中的可行性。(4)将提出的两种组合模型运用于陕北某煤矿工作面,通过模型误差分析得出:在矿压显现预报中,组合模型要优于单个ARMA模型,两种组合模型皆能较为准确的预测支架阻力值与周期来压步距、持续推进距离。并且,Mallat算法与ARMA组合模型误差更低,更能满足安全生产需要。同时,在顶板压力及步距预测中,各影响因素提取相比去噪对预测结果的精度影响更大,在此类研究中,应充分考虑各影响因素的作用,次之是噪声信号的作用。本文的研究成果对工作面矿压显现规律以及周期来压预报研究具有一定的参考价值,并为矿压显现预报提供了一种可以借鉴的新方法。
[Abstract]:In the coal mine production, the pressure acting periodically on the mining face has a certain impact on the safety of coal mine production, and may lead to coal mine accidents. Therefore, the study of pressure appearance and strength prediction in mining face has important theoretical and practical significance for promoting coal mine safety production. In this paper, based on the analysis of domestic and foreign related research, through literature analysis, theoretical analysis and field investigation, comprehensive use of rock pressure theory, statistical analysis, wavelet analysis, time series analysis, The main work and conclusions are as follows: 1) the present situation of the prediction of rock pressure appearance is analyzed. Aiming at the most extensive application of the mechanics method based on the theory of mine pressure at present, the research method of rock pressure appearance prediction is combined with the qualitative and quantitative analysis method, and the main work and conclusion are as follows: (1) the present situation of mine pressure appearance prediction is analyzed. Probabilistic statistical methods are also applied in the fields of loading criterion, dynamic load coefficient, etc. In this paper, a new intelligent method is used to study nonlinear problems. At present, this method is seldom used in the prediction of mine pressure appearance, taking the coal seam of Nanliang 20307 face as an example, according to the statistical results of the data, The results are as follows: the criterion of periodic pressure is 24.4MPa, the intensity of direct top pressure is 24.86MPa, the distance of direct roof pressure is 10.36m, the initial pressure intensity of working face is 31.68MPa, the interval of periodic pressure is 22.5 m, and the intensity of coming pressure is 37.1MPa.f3) On the basis of influencing factors of plate pressure, The combined model of wavelet denoising and ARMA and the combination model of Mallat algorithm and ARMA are constructed, and the feasibility of these two combined models in prediction of mine pressure appearance is analyzed. Through model error analysis, it is concluded that the combined model is superior to the single ARMA model in the prediction of rock pressure behavior, and both of the two combined models can accurately predict the support resistance value, the periodic pressure step distance and the continuous propulsion distance. Moreover, the combined model of Mallat algorithm and ARMA has lower error and can meet the need of safety production. At the same time, in the prediction of roof pressure and step distance, each factor extraction has more influence on the accuracy of prediction results than de-noising. In this kind of research, the effect of each influencing factor should be fully considered, followed by the effect of noise signal. The research results of this paper have some reference value for the study of the regularity of mine pressure appearance and the prediction of periodic pressure, and provide a new method for prediction of mine pressure appearance.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD326

【相似文献】