非等间距WGM-AR模型在基坑周边建筑物沉降预测中的应用

Non-equal interval model of subsidence prediction in surrounding buildings around excavation

非等间距GM(1,1)模型在不等时间间隔序列的趋势分析和预测方面具有重要作用,在此基础上,提出一种基于非等间距加权GM(1,1)模型和自回归AR(p)模型相结合的非等间距加权灰色自回归模型(非等间距WGM-AR模型)。将基坑周边建筑物沉降监测数据视为具有确定趋势的非等时间序列,对序列进行平滑处理,利用非等间距加权GM(1,1)模型提取该时序中的确定性趋势项,用自回归AR(p)模型分析生成的等间距序列中的随机项,并采用内插法得到沉降监测序列的随机项。将组合模型与非等间距GM(1,1)模型计算结果对比分析,结果表明,组合模型具有更高的预测精度,在基坑周边建筑物沉降预测中具有较高的应用价值。

The non-equal interval model plays an important role in trend analysis and forecasting ofsequences with different intervals. Based on this, a combined model which is called non-equal intervalweighted grey-autoregressive model （Non-equal interval WGM_AR model） based on non-equal intervalweighted model and auto-regressive model are proposed. On the condition that the settlement monitor datafrom surrounding buildings around excavation are regarded as non-stationary time series with the definitetrend. Smooth over the data, adopt non-equal interval weighted model to extract the term of definitetrend, use auto-regressivemodel to analyze the random part of equal interval sequence, and obtain therandom entry of settlement monitor data by adopting interpolation. By anal y＇zing comparative results of thecombined model and nonequal interval model, the results show that the accuracy of combined model ishigh which can be wildly applied into the subsidence prediction in surrounding buildings aroundexcavation.