基于基因表达式编程的煤矿地表变形预测研究
发布时间：2018-08-19 06:33
【摘要】：煤矿地表变形是一种普遍的灾害现象,其影响因素非常复杂,不仅包括煤炭的地下开采,煤矿的地质条件以及地下水的分布情况也都会对煤矿的地表变形产生一定的影响,准确地预测煤矿地表变形可以有效避免灾害的发生。目前传统的预测方法都存在预测精度低的问题。因此,如何有效地预测煤矿地表变形对避免灾害的发生有重大的意义。本文首先介绍了煤矿地表变形预测及基因表达式编程的国内外研究现状,阐述了煤矿地表变形特点及变形监测相关理论;然后,针对煤矿地表变形监测数据含有高频噪声的特点,设计了基于Fibonacci数列的加权预处理程序对煤矿地表变形监测原始数据进行了平滑预处理;另外,根据基因表达式编程原理,在Visual Studio编程环境下,利用C#编程语言对基于基因表达式编程的煤矿地表变形预测模型进行一系列程序设计;最后,利用某煤矿前20期的地表变形原始数据作为训练数据,通过设定一定参数,构建了基于基因表达式编程的煤矿地表预测模型,并对后5期的变形数据进行了预测分析;同时把该预测模型与另外建立的基于灰色GM(1,1)预测模型进行精度的对比分析。计算结果表明,利用GEP得到的预测值和实际值相差在4mm到9mm之间,而利用GM(1,1)得到的预测值和实际值相差都在10mm左右,由此可以得出基于基因表达式编程的煤矿地表预测模型能够有效地对煤矿地表变形进行预测,为煤矿地表变形预测提供了一种新的方法。
[Abstract]:The surface deformation of coal mine is a kind of universal disaster phenomenon, and its influencing factors are very complex. Not only the underground mining of coal, but also the geological conditions and the distribution of underground water of coal mine will have certain influence on the surface deformation of coal mine. Accurate prediction of coal mine surface deformation can effectively avoid the occurrence of disasters. At present, the traditional forecasting methods all have the problem of low prediction precision. Therefore, how to effectively predict the surface deformation of coal mines is of great significance to avoid disasters. This paper first introduces the research status of coal mine surface deformation prediction and gene expression programming at home and abroad, expounds the characteristics of coal mine surface deformation and the relevant theory of deformation monitoring. Aiming at the feature of high frequency noise in surface deformation monitoring data of coal mine, a weighted preprocessing program based on Fibonacci sequence is designed for smoothing the original data of surface deformation monitoring in coal mine, in addition, according to the principle of gene expression programming, In the Visual Studio programming environment, a series of programs are designed to predict the surface deformation of coal mine based on genetic expression programming language, and finally, the original data of surface deformation in the first 20 periods of a coal mine are used as the training data. The prediction model of coal mine surface based on gene expression programming is constructed by setting certain parameters, and the deformation data of the latter five periods are predicted and analyzed. At the same time, the accuracy of the prediction model is compared with that based on the grey GM (1 ~ 1) prediction model. The results show that the difference between the predicted value and the actual value obtained by GEP is between 4mm and 9mm, while the difference between the predicted value and the actual value by using GM (1 / 1) is about 10mm. It can be concluded that the prediction model of coal mine surface deformation based on genetic expression programming can effectively predict the surface deformation of coal mine, which provides a new method for the prediction of coal mine surface deformation.
【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD325
[Abstract]:The surface deformation of coal mine is a kind of universal disaster phenomenon, and its influencing factors are very complex. Not only the underground mining of coal, but also the geological conditions and the distribution of underground water of coal mine will have certain influence on the surface deformation of coal mine. Accurate prediction of coal mine surface deformation can effectively avoid the occurrence of disasters. At present, the traditional forecasting methods all have the problem of low prediction precision. Therefore, how to effectively predict the surface deformation of coal mines is of great significance to avoid disasters. This paper first introduces the research status of coal mine surface deformation prediction and gene expression programming at home and abroad, expounds the characteristics of coal mine surface deformation and the relevant theory of deformation monitoring. Aiming at the feature of high frequency noise in surface deformation monitoring data of coal mine, a weighted preprocessing program based on Fibonacci sequence is designed for smoothing the original data of surface deformation monitoring in coal mine, in addition, according to the principle of gene expression programming, In the Visual Studio programming environment, a series of programs are designed to predict the surface deformation of coal mine based on genetic expression programming language, and finally, the original data of surface deformation in the first 20 periods of a coal mine are used as the training data. The prediction model of coal mine surface based on gene expression programming is constructed by setting certain parameters, and the deformation data of the latter five periods are predicted and analyzed. At the same time, the accuracy of the prediction model is compared with that based on the grey GM (1 ~ 1) prediction model. The results show that the difference between the predicted value and the actual value obtained by GEP is between 4mm and 9mm, while the difference between the predicted value and the actual value by using GM (1 / 1) is about 10mm. It can be concluded that the prediction model of coal mine surface deformation based on genetic expression programming can effectively predict the surface deformation of coal mine, which provides a new method for the prediction of coal mine surface deformation.
【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD325
【参考文献】
相关期刊论文 前10条
1 袁广祥;赵海军;;金属矿山开采地表岩移的灰色预测模型[J];地下空间与工程学报;2011年01期
2 蒋建平;陈功奇;章杨松;;偏最小二乘回归在地表沉陷预测中的应用[J];重庆大学学报;2010年09期
3 王莉;赵彬;;基于时间序列的胶结充填体变形预测[J];化工矿物与加工;2010年01期
4 姜s，
本文编号：2190944
本文链接：https://www.wllwen.com/kejilunwen/kuangye/2190944.html
