基于小波去噪的非等间隔变形监测数据GM(1,1)预测方法研究

摘要

在小波去噪和非等间隔GM(1,1)的基础上提出基于小波去噪的非等间隔GM(1,1)组合模型的预测方法,并且通过矿区实测数据模拟实验对三种方法进行对比分析.结果表明,小波去噪的模拟精度高于非等间隔GM(1,1),组合模型的预计精度高于单独使用非等间隔GM(1,1),验证了该方法的实用性和可靠性.%This paper presents a prediction model of unequal interval GM (1,1) combined with wavelet denoise based on the ordinary unequal interval GM (1,1) and wavelet denoising,and compared analysis with the three methods by simulated experiment of mine da-ta. The results show that the simulation precision of wavelet denoising is higher than that of unequal interval GM (1,1) and the pre-diction accuracy of the combined model is higher than that of using unequal interval GM (1,1) alone, which verifies the practicability and reliability of the proposed method.