基于TLBO-LOIRE的回采工作面瓦斯涌出量预测

发布时间：2018-01-23 18:23

本文关键词： 瓦斯涌出量预测 TLBO LOIRE 参数优化　出处：《应用基础与工程科学学报》2017年05期 　论文类型：期刊论文

【摘要】：瓦斯涌出量是瓦斯防治与管理、矿井通风系统设计的重要基础数据,准确地预测瓦斯涌出量对于煤矿安全生产有着极其重要的指导意义与应用价值.但工作面瓦斯涌出规律复杂,在检测、数据采集过程中不可避免地会混入异常噪声,直接影响着瓦斯预测的准确性.本文采用l1正则化异常值隔离与回归方法(LOIRE)对煤矿回采工作面瓦斯涌出量及其相关影响因素的统计样本数据库进行计算分析,隔离样本的异常噪声干扰,利用教与学算法(TLBO)优化回归参数,建立了回采工作面瓦斯涌出量的优化预测模型,并对煤矿现场数据进行分析预测,结果表明3个回采工作面的瓦斯涌出量预测误差分别为3.04%、0.33%和2.36%,平均相对误差仅为2.36%.TLBO-LOIRE优化预测方法,预测准确性高,能够满足井下瓦斯防治的工程需要,对其它工程领域的数据预测同样适用.
[Abstract]:The gas emission is gas control and management, the important basic data for the design of ventilation system, accurate prediction of gas emission has an important guiding significance and application value for the coal mine safety production. But the working face gas emission law in complex detection, data collection process will inevitably mixed with abnormal noise, direct effect the accuracy of the prediction of gas. This paper uses the isolation and regression method of abnormal L1 regularization (LOIRE) calculation and analysis of statistical sample database factors on gas emission quantity of coal mining face and related effects, interference of abnormal noise isolation samples, the use of teaching and learning algorithm (TLBO) optimization and regression parameters, optimize emission gas coalface prediction model, and analyzed the mine site data, the result shows that the emission prediction error of 3 working face gas 3.04%, 0.33% and 2.36%, respectively. The average relative error is only 2.36%.TLBO-LOIRE, and the prediction method is optimized. The prediction accuracy is high, which can meet the engineering needs of underground gas prevention and control. It is also applicable for other engineering data prediction.

【作者单位】：安徽理工大学安徽省矿山机电装备协同中心;香港中文大学工程学院;香港大学电机电子工程系;
【基金】：国家自然科学基金项目(41542002,51475001)
【分类号】：TD712.5
【正文快照】： 瓦斯是一种从煤和围岩中逸出的甲烷、二氧化碳和氮气等组成的混合气体,当空气中瓦斯含量达到一定范围时,遇火会引起爆炸,造成严重安全事故.按瓦斯涌出量和瓦斯涌出形式来进行分类,中国的高瓦斯矿井约占矿井总量的35%,这些高瓦斯矿严重威胁着采煤工作现场的生产和井下人员的人

【相似文献】