摘要
矿山开采引起地表和建筑物变形,对矿区的安全生产产生不利影响,需要对矿区的地表沉降监测和预测。为提高预测模型的质量,文中在分析GM(1,1)模型的基础上,对GM(1,1)模型的参数估计方法进行了改进,用普通最小一乘法和折扣最小一乘法代替了原来的最小二乘法,并以煤矿的主井沉降监测为例,进行了分析、预测,验证了该改进方法的有效性,取得了满意的效果。
The coal mining-induced surface and buildings deformation always brings about unfavorable influence for safe production in the mining area,so it is necessary to monitor and predict the subsidence of the surface there. In the paper,in order to improve quality of the predictive model,improvement is made for the parameter estimation method of the GM(1,1) model based on analysis of this model,that is to say,the original least-squares method is replaced with the common least-absolute value method and the method of discount least-absolute value. The analysis and prediction made through taking the subsidence monitoring of the main shaft at a mine for example prove that the improved method is effective and satisfactory results could be obtained.
出处
《矿山测量》
2008年第3期23-25,28,共4页
Mine Surveying
基金
山东省自然科学基金资助项目(2004ZX31)
