论文部分内容阅读
针对传统灰色GM(1,1)预测模型在建筑物变形监测预报中的拟合精度较差、预测精度较低和预测时间较短的问题,文中以传统GM(1,1)、线性回归和马尔科夫模型为理论基础,构建了灰线性马尔科夫预测模型,并结合某建筑物变形监测的观测数据,运用新陈代谢的计算模式进行预测。结果表明,灰线性马尔科夫预测模型的拟合精度和预测精度优于单一的灰色GM(1,1)预测模型和线性回归预测模型,灰线性马尔科夫预测模型具有预测精度高、预测时间长和稳定性高的优势。
Aiming at the poor fitting accuracy, low prediction accuracy and short forecasting time of traditional gray GM (1,1) prediction model in building deformation monitoring and prediction, this paper uses the traditional GM (1,1), linear regression and Markov model as the theoretical basis, the construction of a gray linear Markov forecasting model, combined with deformation monitoring of a building observation data, the use of metabolic prediction model to predict. The results show that the fitting accuracy and prediction accuracy of the gray linear Markov forecasting model are better than the single gray GM (1,1) forecasting model and the linear regression forecasting model. The gray line Markov forecasting model has the advantages of high prediction accuracy, Long and high stability advantages.