高频GNSS变形序列的谱分析方法

发布时间：2018-12-31 13:29

【摘要】：高频GNSS因为可以全天时、高精确度地收集监测站的位移数据,在煤炭安全生产中得到非常广阔的应用。采矿区地表变形GNSS观测的数据是是频率域中的低频与高频数据组合而成的时间序列数据[1]。现在对于GNSS观测数据的一些分析,是由时域分析和频域分析两大方面组合而成的。频域分析又叫做谱分析[2],本文从谱分析的角度出发,基于唐家会煤矿GNSS观测数据,通过傅立叶变换将其转换到频域空间,经过谱分析可提取各种周期项信息,分析这些周期变形的激发机制,同时对监测过程中数据突变产生奇异点的信号,通过使用小波,可以更好的监测煤矿安全开采过程中的变化状况。文中主要的研究内容如下:(1)总结了 GNSS变形监测的一些优点和使用现状,而且对当前使用比较广泛的数据处理方式方法进行了很好的的介绍。(2)联系唐家会煤矿自动化监测系统,全流程的叙述了其数据采集处理程序,并对GNSS系统进行了较为具体的的说明。(3)对唐家会煤矿监测数据进行低频谱分析。首先通过傅立叶变换将时域数据变换到频率域中,计算出其一些峰值对应频率及相关的振幅和初相,对一些影响比较大的量的全周期波以及半日波信号,运用多方面分析,而且针对半日波产生的一些因素不能很好的具体指出,建立四参数模型进行具体讨论。得出全周期波的影响因素主要是地下开采作业导致的,且变形的量与地下作业推进量成正相关。温度对沉降变化低频部分产生影响且影响量在5.5cm左右,多路径影响可以忽略不计,只在降雨天气产生一定的影响。(4)对监测数据的高频部分的奇异点利用小波对出奇异点进行精确定位,得出影响的数据点位为[123,213,402,501,781,1110,1440,2150,3010,3415,3790,4875,5415,5670,6820,8255,9700,10290]这些点位所在的时间都分布煤矿0时到上午8时,8时到14时的共5个放炮工作段,找出导致产生奇异点的影响因子为煤矿的放炮作业。
[Abstract]:High frequency GNSS can collect displacement data of monitoring station with high accuracy all day, so it is widely used in coal safety production. The GNSS data of surface deformation in mining area are time series data combined with low frequency and high frequency data in frequency domain [1]. Some analysis of GNSS observation data is made up of time domain analysis and frequency domain analysis. Frequency domain analysis is also called spectrum analysis [2]. From the point of view of spectrum analysis, based on the GNSS observation data of Tangjiahui Coal Mine, the frequency domain analysis is transformed into frequency domain space by Fourier transform, and the information of various periodic items can be extracted by spectrum analysis. The excitation mechanism of these periodic deformation is analyzed, and the singularity signal of the data sudden change in the monitoring process is analyzed. By using wavelet, the change of coal mine safety mining process can be better monitored. The main research contents are as follows: (1) some advantages and application status of GNSS deformation monitoring are summarized. In addition, the widely used methods of data processing are introduced. (2) the whole process of data acquisition and processing is described by connecting with the automatic monitoring system of Tangjiahui Coal Mine. The GNSS system is explained in detail. (3) the low frequency spectrum of the monitoring data of Tangjiahui Coal Mine is analyzed. Firstly, the time domain data is transformed into the frequency domain by Fourier transform, and some peak corresponding frequencies and related amplitudes and initial phases are calculated. And some factors produced by semidiurnal wave can not be well pointed out, and the establishment of four-parameter model is discussed in detail. It is concluded that the influence factors of the whole period wave are mainly caused by the underground mining operation, and the amount of deformation is positively related to the underground operation propulsive quantity. The influence of temperature on the low frequency part of settlement change is about 5.5cm, and the multipath effect can be ignored. (4) the singularity of the high frequency part of the monitoring data is accurately located by wavelet transform. The data points of influence are: [123C 2130402N 501N 781n 111010 1440402150U 301010 3415U 3790 4875N 5415N 56706820 8255970010290] these points are located in 5 working sections of coal mine from 0: 00 to 8: 00 and 8: 00 to 14:00. It is found that the influence factor of the singularity point is the firing operation of coal mine.
【学位授予单位】：安徽理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P228.4

【相似文献】