概率积分法参数求取和模型修正方法研究及程序实现

发布时间：2018-06-21 19:16

本文选题：地表移动 + 预计参数　；参考：《安徽理工大学》2017年硕士论文

【摘要】：煤矿生产近两年有所缓慢,但煤炭作为我国主要能源的地位是暂时无法改变的。煤炭开采过程中所造成的地表移动影响着矿区的生产安全,对矿区地表的移动和变形进行准确的预计是必不可少的,对采区上方地表的变形预计工作尤为重要,准确的模型参数决定着预计结果的精度。目前,对于边缘收敛快的缺陷没有得到彻底解决。本文选取的工程实例为五沟矿1013工作面及皖北矿区其他9个工作面。介绍了基于概率积分的最小二乘法、模矢法、遗传算法和经验公式求参四种方法,以皖北矿区其他9个工作面为例,得到了综合求参经验公式。分别采用最小二乘法、模矢法、遗传算法和经验公式求参四种不同的方法对五沟矿1013工作面进行求参,结果表明遗传算法的预计精度高于其他三种预计方法,共性问题是预计结果具有边缘收敛快的缺陷。本文提出了新的修正模型,针对五沟矿1013工作面,基于遗传算法,分别利用修正模型和经典模型进行求参,并对结果进行了比较分析。本文利用MATLAB软件的开发平台,对四种求取预计参数的方法分别进行编程,求取地表移动预计参数,对修正模型进行编程计算。预计结果基本解决了边缘收敛快的缺陷,修正后的遗传算法模型在求取参数精度上具有更好的适用性,对其他相似地质采矿条件的矿区地表移动预计工作具有一定的指导意义。
[Abstract]:Coal mine production has been slow in the past two years, but the position of coal as the main energy of our country is temporarily unable to change. The surface movement caused by the coal mining process affects the production safety of the mining area. It is necessary to accurately predict the movement and deformation of the surface of the mining area, and the work on the surface deformation above the mining area is especially expected. It is important that the accurate model parameters determine the accuracy of the predicted results. At present, the defects of fast convergence of the edge have not been solved thoroughly. The engineering example selected in this paper is the 1013 working face of the five ditch and the other 9 other working faces in the northern Anhui mining area. The four methods, taking the other 9 working faces in the northern Anhui mining area as an example, obtained the comprehensive reference empirical formula. Using the least square method, the model vector method, the genetic algorithm and the empirical formula, four different methods were used to calculate the 1013 working faces of the five gully mine. The results show that the prediction accuracy of the genetic algorithm is higher than the other three prediction methods. A new correction model is proposed in this paper. Based on the genetic algorithm, a new correction model is proposed for the five groove mine 1013 working face. The results are compared and analyzed by using the modified model and the classical model respectively. This paper uses the opening platform of the MATLAB software for four methods to obtain the estimated parameters, respectively. The predicted parameters of the surface movement are calculated and the modified model is programmed and calculated. The predicted results basically solve the defects of the fast convergence of the edge. The modified genetic algorithm model has a better applicability in obtaining the precision of the parameters, and it has some guidance for the surface movement prediction in other mining conditions with similar geological conditions. Righteousness.
【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】：TD325.2

【相似文献】