论文部分内容阅读
对于大型和高耸建筑,为了确保施工和安全运营,定期进行沉降监测是一项非常必要和重要的工作。但是由于受各种主、客观条件的限制和影响,沉降监测实测资料有时会出现间断。为了分析沉降规律,需要对间断资料进行插补。文章考虑到建筑物沉降监测非等时间间隔的特点,以相邻观测时间间隔为权,直接生成1-WA-GO序列,建立灰色系统非等间隔GM(1,1)模型,并将该模型应用于某建筑物沉降监测间断资料的插补工作中。当拟合值误差较大时,应用1阶残差序列建模,对原模型进行修正,提高了模型的拟合精度,得到了较满意的监测间断插补值。
For large and towering buildings, periodic settlement monitoring is a very necessary and important task to ensure construction and safe operation. However, due to the restrictions and influences of various kinds of main and objective conditions, the measured data of settlement monitoring sometimes appear intermittent. In order to analyze the settlement law, we need to interpolate the discontinuous data. Taking into account the characteristics of unequal time intervals of building settlement monitoring, this paper directly generates 1-WA-GO sequence by taking the adjacent observation time interval as the right, builds a gray system non-interval GM (1,1) model, Applied to a building settlement monitoring intermittent data interpolation work. When the fitting error is large, the first-order residual sequence modeling is applied to correct the original model, which improves the fitting accuracy of the model and obtains a more satisfactory discontinuity interpolation value.