煤矿井下人员精确定位系统数据压缩

发布时间：2018-07-14 18:43

【摘要】：煤矿井下人员定位系统是煤矿企业实现安全生产的重要保证之一。随着采煤企业规模的不断扩大化,由定位设备采集的井下人员位置数据也在快速增长。为了实现对矿井人员更加规范与合理的安全管理能力,煤矿井下人员定位系统势必要存储和传输更加海量的复杂数据。通过对海量人员位置数据的压缩优化,不仅可以有效减少系统存储和传输海量数据的负担,降低带宽资源的消耗,还可以大量缩减人员位置数据的冗余度,对井下各类工种的行为轨迹分析和挖掘提供便利。由于传统的压缩算法没有结合定位数据的时空特性,只是从文本的角度来进行压缩,因此,基于通用型的数据压缩算法并无法达到较高的压缩比。通过分析定位数据的时空特性,论文将定位数据的压缩划分为对轨迹类数据压缩问题。目前,轨迹压缩算法多是针对于车辆定位数据,通过对部分特征点的保留而实现对轨迹数据的粗粒度压缩。该算法虽然能够实现极高的压缩比,但是无法保证压缩质量。本文通过对现有轨迹压缩算法的研究与对比,并结合煤矿井下人员定位系统的定位原理和人员位置数据的特点,从而实现对人员定位数据的压缩。首先,综合分析了相关轨迹算法的优势和缺陷,进而选取基于路网特征的算法思想实现了以巷道信息来描述人员轨迹信息的初步压缩方案。其次,选取人员在巷道中的折返、停留信息作为人员运动特征的关键信息。通过对井下人员在巷道中折返点以及长时间停留点的分析和判定,实现了对井下人员在巷道内的具体运动情况的记录,从而完成人员位置数据的压缩。最后,论文采用时间比例等距划分的方式,逐步对相邻关键点之间信息进行重构,实现了数据的解压缩。相较于简单的基于路网特征的轨迹压缩算法,应用改进算法对井下人员位置数据压缩,不但可以获得较高的压缩率,而且可以显著降低重构数据的误差,提高压缩数据的质量和应用价值。结合井下人员的运动受巷道分布情况限制的特性,人员的运动轨迹会遵循特定的规律。对此,提出基于样本轨迹的轨迹压缩算法。在试验中,利用若干样本轨迹对海量新轨迹进行重构,实现了用样本轨迹对海量轨迹数据的压缩,获得较高的压缩率。
[Abstract]:The personnel positioning system in coal mine is one of the important guarantee for coal mine enterprises to realize safe production. With the expansion of the scale of coal mining enterprises, the location data of underground personnel collected by positioning equipment are also growing rapidly. In order to achieve a more standardized and reasonable safety management ability for mine personnel, the mine personnel positioning system is bound to store and transmit more massive complex data. By optimizing the compression of massive human location data, it can not only effectively reduce the burden of storing and transmitting massive data, reduce the consumption of bandwidth resources, but also greatly reduce the redundancy of personnel location data. It is convenient to analyze and excavate the behavior track of all kinds of underground work. Because the traditional compression algorithm does not combine the spatio-temporal characteristics of the location data, but only from the point of view of the text, the data compression algorithm based on general purpose can not achieve a higher compression ratio. By analyzing the spatial and temporal characteristics of the location data, the compression of the location data is divided into the compression of the locus data. At present, trajectory compression algorithms are mostly aimed at vehicle location data, and the coarse granularity compression of track data is realized by preserving some feature points. Although the algorithm can achieve extremely high compression ratio, it can not guarantee the compression quality. Based on the research and comparison of the existing algorithms of trajectory compression, combined with the positioning principle of the personnel positioning system in coal mine and the characteristics of the personnel position data, the compression of the personnel location data can be realized in this paper. Firstly, the advantages and disadvantages of the related trajectory algorithm are analyzed synthetically, and then the algorithm based on the road network features is selected to realize the preliminary compression scheme of describing the human trajectory information by the laneway information. Secondly, the key information of the movement of the personnel is to select the reentry and stay information in the laneway. Through the analysis and judgment of the entry point and the long stay point in the roadway, the record of the specific movement of the downhole personnel in the tunnel is realized, and the data of the personnel position is compressed. Finally, the paper reconstructs the information between the adjacent key points step by using the method of time proportional isometric partition, and realizes the decompression of the data. Compared with the simple path compression algorithm based on road network features, the improved algorithm can not only obtain a higher compression ratio, but also reduce the error of reconstruction data significantly. Improve the quality and application value of compressed data. Combined with the characteristic that the movement of underground personnel is restricted by the distribution of roadway, the movement track of personnel will follow a specific law. A trajectory compression algorithm based on sample trajectory is proposed. In the experiment, several sample trajectories are used to reconstruct the massive new trajectories, and the compression ratio of the massive trajectory data is obtained by using the sample trajectories.
【学位授予单位】：太原科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD76

【参考文献】