Settlement prediction model modeling method considering data timeliness

Abstract

The invention discloses a settlement prediction model modeling method considering data timeliness. The method comprises the following steps: (1) constructing a linear settlement prediction model taking a function of time t as an independent variable; (2) obtaining a coefficient vector estimated value and a weight matrix W (n) of the prediction model by using a least square method based on settlement observation data in the first n periods; (3) establishing a prediction model error equation for settlement observation data in the mth (m > n) period; (4) establishing a prediction model virtual error equation by utilizing a coefficient vector estimation value; introducing a weight reduction coefficient k, and correcting the weight matrix W (m-1) into W '(m-1); (5) simultaneously establishing the error equations in the steps (3) and (4), and calculating to obtain an updated coefficient vector estimated value and a weight matrix W (m) of the coefficient vector estimated value; and (6) continuously increasing settlement observation data, and repeating the steps (3), (4) and (5) to realize synchronous updating of the settlement prediction model along with engineering progress.

Description

technical field [0001] The invention belongs to the field of surveying and mapping, and relates to a settlement prediction model modeling method in consideration of data timeliness. Background technique [0002] After the completion of the project, the settlement caused by the consolidation effect of the soil and changes in other factors is called the post-construction settlement. Accurate prediction of post-construction settlement can provide an important reference for engineering safety assessment and maintenance program design. For engineering projects requiring long-term maintenance such as tunnels, roads, and bridges, the importance of settlement prediction is more prominent. [0003] At present, post-construction settlement is mainly predicted by prediction models, and common models include hyperbolic method, exponential curve method, Hoshino method, etc. Regardless of the settlement prediction model, its stochastic model is generally determined based on the accuracy...

AI Technical Summary

Problems solved by technology

[0004] Technical problem: In view of the deficiencies of the prior art, the present invention proposes a settlement prediction model modeling method that takes into account the timeliness of data, which is used to solve the unreasonable defect of the random model used to predict post-construction settlement in actual civil engineering applications

Images