基于时序模型的变形预报研究

Model of time series analysis for deformation forecast

时序模型是一种有效的变形预报方法,但在很多变形预报时序模型应用中,存在建模过程不严密、缺乏统计检验和模型验证等问题。针对这些问题,本文首次在变形预报中采用LM检验来判断序列的相关性,ADF检验判断序列的平稳性,自相关图和偏自相关图初步确定模型类型和阶数以及AIC准则来确定最终模型,确保建模的每一过程都有严格的统计意义。文中利用实测沉降数据,建立了预报模型,并利用模型进行预报;最后将预报结果与实测数据比较,验证了预报模型的有效性,证明本文建模方法科学合理,能用于实际变形预报。

Time-series model is an effective method for deformation forecast.But in much application of time-series model for deformation forecast,the modeling process is not strict,lacking of statistics test and model test.So for the first time the paper cited LM test to determine the correlation of sequence,ADF test to determine the stability of sequence,the autocorrelation figure and partial autocorrelation figure to determine the type and the order of model preliminarily,and AIC criterion to determine the final model in deformation forecast,to assure every modeling process have strict statistical significance.With a grou Pof deformation monitoring data,it established the forecast model and forecasts.Comparing forecast results with monitoring data,it verified the effectiveness of forecast model,could be used in actual deformation forecast.