CN-115906672-B - Method and system for predicting response of historical building structure under atmospheric temperature field

CN115906672BCN 115906672 BCN115906672 BCN 115906672BCN-115906672-B

Abstract

The invention discloses a method and a system for predicting historical building structure response under an atmospheric temperature field, wherein the method comprises the steps of obtaining atmospheric temperature field information, structure temperature field information and structure response information in a certain period of time; the atmospheric temperature field information, the structure temperature field information and the structure response information are respectively used as training sets to establish and train an atmospheric temperature prediction structure temperature model and a structure temperature prediction structure response model based on an artificial intelligent algorithm, future structure temperature field information is predicted on the atmospheric temperature prediction structure temperature model according to the future atmospheric temperature field information and the historical structure temperature field information, and future structure response information of a historical building structure in a certain period of time is predicted on the structure temperature prediction structure response model according to the future structure temperature field information and the historical structure response information. The invention improves the prediction precision and accuracy of the structural response of the historical building along with the change of the atmospheric temperature, and has wide application range.

Inventors

ZHANG YINGNAN
HE JIAO
ZHANG SHUKAI
GU ZHIWANG
LI HAIQING
LIU XIAO
MIAO FENG
WANG WEIMAO
SUN SHENPENG
MA YUEYANG

Assignees

上海建工四建集团有限公司

Dates

Publication Date: 20260505
Application Date: 20230106

Claims (10)

1. A method for predicting response of a historic building structure in an atmospheric temperature field, comprising: Acquiring atmospheric temperature field information, structure temperature field information and structure response information of an area where a historical building structure is located in a past period of time; taking the atmospheric temperature field information and the structural temperature field information in a certain period of time as training sets, and establishing and training an atmospheric temperature prediction structural temperature model based on an artificial intelligent algorithm according to the structural temperature field information acquired along with the change of the atmospheric temperature field information in a certain period of time; taking the structure temperature field information and the structure response information in the period of time as a training set, and establishing and training a structure temperature prediction structure response model based on an artificial intelligence algorithm according to the structure response information acquired along with the change of the structure temperature field information in the period of time; predicting future structure temperature field information in a future period of time on the atmospheric temperature prediction structure temperature model according to the input future atmospheric temperature field information in the future period of time and the history structure temperature field information in a past period of time before the future period of time, and inputting the predicted future structure temperature field information into the structure temperature prediction structure response model; Future structural response information of the historical building structure within a future period of time is predicted on the structural temperature prediction structural response model based on the future structural temperature field information predicted by the structural temperature prediction structural response model and the historical structural response information within a past period of time prior to the future period of time.
2. The method for predicting historic building structure response under an atmospheric temperature field according to claim 1, Combining and superposing an atmospheric temperature prediction structure temperature model and a structure temperature prediction structure response model to form an atmospheric temperature field prediction structure response model; The method comprises the steps of predicting future structure temperature field information in a future period of time according to the input future atmospheric temperature field information in the future period of time and the historical structure temperature field information and the historical structure response information in a past period of time before the future period of time on an atmospheric temperature field prediction structure response model, and predicting the future structure response information of a historical building structure in the future period of time according to the future atmospheric temperature field information and the historical structure temperature field information.
3. The method according to claim 1, wherein the atmospheric temperature field information of the area in which the historic building structure is located is obtained by weather forecast or historic weather record over a certain period of time, the structural temperature field information under the atmospheric temperature field information over the certain period of time is obtained by a temperature sensor provided on the historic building, and the structural response information under the structural temperature field information over the certain period of time is obtained by a structural response sensor provided on the historic building.
4. The method for predicting the response of a historical building structure in an atmospheric temperature field according to claim 1, wherein the atmospheric temperature prediction structural temperature model is built and trained according to structural temperature field information obtained along with the change of the atmospheric temperature field information and the thickness of cloud layer in a certain period of time in the past.
5. The method for predicting the response of a historical building structure under an atmospheric temperature field according to claim 1, wherein the atmospheric temperature prediction structural temperature model is built and trained according to structural temperature field information acquired along with the change of the atmospheric temperature field information and rainfall condition in a certain period of time in the past.
6. The method for predicting response of a historical building structure under an atmospheric temperature field according to claim 1, wherein the atmospheric temperature prediction structural temperature model is built and trained according to structural temperature field information acquired along with the change of the atmospheric temperature field information, the cloud layer thickness and the rainfall condition in a certain period of time in the past.
7. The method for predicting historical architectural structure responses in an atmospheric temperature field according to claim 1, wherein the model is trained using SGD or Adam algorithm.
8. The method for predicting historic building structure responses under an atmospheric temperature field according to claim 1, wherein the model types of the atmospheric temperature prediction structure temperature model and the structure temperature prediction structure response model are DNN, RNN, LSTM or a transducer model.
9. A system for predicting response of a historic building structure in an atmospheric temperature field, comprising: The information acquisition module is used for acquiring the atmospheric temperature field information, the structure temperature field information and the structure response information of the area where the historical building structure is located in a certain period of time in the past; The atmospheric temperature prediction structure temperature module takes atmospheric temperature field information and structure temperature field information in a certain period of time in the past as training sets, establishes and trains an atmospheric temperature prediction structure temperature model based on an artificial intelligence algorithm according to the structure temperature field information acquired along with the change of the atmospheric temperature field information in the certain period of time in the past; The structure temperature prediction structure response module takes the structure temperature field information and the structure response information in the period of time as a training set, establishes and trains a structure temperature prediction structure response model based on an artificial intelligence algorithm according to the structure response information acquired along with the change of the structure temperature field information in the period of time, and predicts future structure response information of a historical building structure in a period of time in the future according to the future structure temperature field information predicted by the structure temperature prediction structure response model and the historical structure response information in a period of time before the period of time in the future on the structure temperature prediction structure response model.
10. A system for predicting response of a historic building structure in an atmospheric temperature field, comprising: The information acquisition module is used for acquiring the atmospheric temperature field information, the structure temperature field information and the structure response information of the area where the historical building structure is located in a certain period of time in the past; The atmospheric temperature prediction structure temperature module takes the atmospheric temperature field information and the structure temperature field information in a certain period of time as a training set, and establishes and trains an atmospheric temperature prediction structure temperature model based on an artificial intelligence algorithm according to the structure temperature field information obtained along with the change of the atmospheric temperature field information in a certain period of time; The structure temperature prediction structure response module takes the structure temperature field information and the structure response information in the period of time as a training set, and establishes and trains a structure temperature prediction structure response model based on an artificial intelligence algorithm according to the structure response information acquired along with the change of the structure temperature field information in the period of time; The atmospheric temperature field prediction structure response module is used for combining and superposing an atmospheric temperature prediction structure temperature model and a structure temperature prediction structure response model to form a composite prediction model, and predicting future structure temperature field information in a future period of time according to the input future atmospheric temperature field information in the future period of time and historical structure temperature field information and historical structure response information in a past period of time before the future period of time on the atmospheric temperature field prediction structure response model, and predicting future structure response information of a historical building structure in the future period of time according to the future atmospheric temperature field information and the historical structure temperature field information.

Description

Method and system for predicting response of historical building structure under atmospheric temperature field Technical Field The invention relates to the field of civil engineering, in particular to a method and a system for predicting historical building structure response under an atmospheric temperature field. Background In the use process of the existing buildings such as historical buildings, the existing buildings are affected by adverse factors such as sun exposure, wind erosion, rain, foundation settlement, rust of steel structures, shrinkage of concrete and the like, and are also affected by stable loads such as temperature change and the like. At present, a part of historical building structures are provided with corresponding monitoring equipment, so that physical quantities such as displacement, stress and strain are monitored on the historical building structures. Existing monitoring data shows that the structural response of a relevant historical building is affected by the temperature conditions of the structural surface, which are also correlated to the atmospheric temperature field. The existing structure response process based on temperature prediction is often based on a finite element method, a structure temperature-stress model is established, and the structure response is predicted by the temperature through the finite element method. However, as the historical building is long in age, the structure of the historical building changes with time to a certain extent, the problem of inaccurate modeling exists in a method of predicting structural response by temperature through finite elements, and in addition, calculation of a single finite element calculation model is long in time consumption, and more calculation resources are needed for carrying out the finite element calculation again under different temperature working conditions. Disclosure of Invention The invention aims to provide a method and a system for predicting historical building structure response under an atmospheric temperature field, which are used for solving the problems that modeling from traditional finite element analysis temperature to a structure response method cannot be updated in real time along with time, calculation is time-consuming, repeated calculation is needed under different working conditions, and more calculation resources are needed. In order to solve the technical problems, the invention provides a method for predicting the response of a historical building structure under an atmospheric temperature field, which comprises the following steps: Acquiring atmospheric temperature field information, structure temperature field information and structure response information of an area where a historical building structure is located in a past period of time; taking the atmospheric temperature field information and the structural temperature field information in a certain period of time as training sets, and establishing and training an atmospheric temperature prediction structural temperature model based on an artificial intelligent algorithm according to the structural temperature field information acquired along with the change of the atmospheric temperature field information in a certain period of time; taking the structure temperature field information and the structure response information in the period of time as a training set, and establishing and training a structure temperature prediction structure response model based on an artificial intelligence algorithm according to the structure response information acquired along with the change of the structure temperature field information in the period of time; predicting future structure temperature field information in a future period of time on the atmospheric temperature prediction structure temperature model according to the input future atmospheric temperature field information in the future period of time and the history structure temperature field information in a past period of time before the future period of time, and inputting the predicted future structure temperature field information into the structure temperature prediction structure response model; Future structural response information of the historical building structure within a future period of time is predicted on the structural temperature prediction structural response model based on the future structural temperature field information predicted by the structural temperature prediction structural response model and the historical structural response information within a past period of time prior to the future period of time. Further, the invention provides a method for predicting the response of a historical building structure under an atmospheric temperature field, Combining and superposing an atmospheric temperature prediction structure temperature model and a structure temperature prediction structure response model to form an atmospheric temperature field prediction structu