CN-122020438-A - Load and environmental factor coupled reinforced concrete bridge service life prediction method

CN122020438ACN 122020438 ACN122020438 ACN 122020438ACN-122020438-A

Abstract

The invention relates to the technical field of new generation information, in particular to a service life prediction method of a reinforced concrete bridge with load and environmental factor coupling, which is used for realizing high-frequency real-time acquisition and encryption transmission of load and environmental data through an intelligent sensing node network, generating a lightweight coupling feature vector by utilizing edge calculation to support real-time damage evolution simulation of a digital twin model, and simultaneously polymerizing multi-source bridge data to optimize a global service life prediction model on the premise of protecting data privacy through a federal learning framework.

Inventors

ZHU LI
LI JIAHUAN
WANG ZHAO
YANG ZHICHENG
ZHAO GUANYUAN
LI PEI

Assignees

北京交通大学
北京市市政工程研究院

Dates

Publication Date: 20260512
Application Date: 20251203

Claims (8)

1. A method for predicting service life of reinforced concrete bridge by coupling load and environmental factors is characterized in that, Comprises the following steps: the method comprises the steps of acquiring load data and environment data in real time through an intelligent sensing node network deployed at a key part of a bridge, and encrypting and transmitting the load data and the environment data to a preprocessing module; The preprocessing module performs preprocessing and feature extraction on the edge calculation level of the collected original data to generate a lightweight coupling feature vector; Based on the generated lightweight coupling feature vector, driving a digital twin model of the bridge to perform damage evolution simulation under the coupling effect of load and environmental factors, and outputting the degradation state and future trend of key performance indexes of the bridge; Utilizing a federal learning framework to aggregate anonymized data of a plurality of bridges, optimizing a global life prediction model, and calculating residual service life and reliability indexes based on the degradation state; and generating a visual report and maintenance decision advice according to the service life evaluation result.
2. The method for predicting the service life of a reinforced concrete bridge coupled with load and environmental factors according to claim 1, The specific mode for encrypting and transmitting the collected data to the preprocessing module is as follows: When receiving the transmission task, the system firstly applies the key manager to the key generator to generate the encryption key and the decryption key, and distributes the encryption key and the decryption key to the encryption module and the decryption module respectively. Then, the encryption module encrypts the acquired data by using the encryption key and transmits the ciphertext to the preprocessing module. The preprocessing module decrypts the ciphertext through the decryption key in the decryption module to obtain the original data, and then preprocessing and feature extraction operations are completed.
3. The method for predicting the service life of a reinforced concrete bridge coupled with load and environmental factors according to claim 2, The intelligent sensing node network comprises a load monitoring sensor and an environment monitoring sensor; The load monitoring sensor comprises a strain sensor and an acceleration sensor and is used for monitoring static and dynamic loads; The environment monitoring sensor comprises a temperature sensor, a humidity sensor and a chloride ion concentration sensor and is used for monitoring environmental corrosion factors; the intelligent sensing node has edge computing capability and can perform primary data cleaning and caching.
4. The method for predicting the service life of a reinforced concrete bridge coupled with load and environmental factors according to claim 3, The specific process for generating the lightweight coupling feature vector by preprocessing and feature extraction of the edge computing layer of the collected original data comprises the following steps: Extracting characteristic quantities related to vehicle traffic from the load data by using a time-frequency analysis algorithm; performing space-time alignment and feature coding on the environmental data, and extracting temperature and humidity gradient and corrosion medium concentration change rate indexes; And fusing the extracted load with the environment characteristic vector to generate a comprehensive characteristic vector representing the load-environment coupling effect.
5. The method for predicting the service life of a reinforced concrete bridge coupled with load and environmental factors according to claim 4, The digital twin model is a physical mechanism model constructed based on BIM and finite element analysis, and is fused with a data driving model; The digital twin model receives a lightweight coupling feature vector, and dynamically updates environmental boundary conditions and load input in the model through a parameterized interface; and the coupling damage simulation adopts a parallel calculation algorithm to calculate the concrete carbonization depth, the steel bar corrosion rate and the evolution process of fatigue damage under the action of coupling factors.
6. The method for predicting the service life of a reinforced concrete bridge coupled with load and environmental factors according to claim 5, The data driving model fused in the digital twin model is a deep learning model, and training data of the deep learning model integrates historical monitoring data and a numerical simulation result through a transfer learning technology so as to quickly adapt to bridge characteristics of different areas.
7. The method for predicting the service life of a reinforced concrete bridge coupled with load and environmental factors according to claim 6, The federal learning framework operates in such a way that local data and model parameters of each bridge are aggregated after encryption, a global prediction model is updated, and the optimized model parameters are issued to each local node; The life evaluation not only outputs the expected value of the residual service life, but also outputs the reliability index based on the probability statistical method, and the early warning is triggered when the reliability index is lower than a preset threshold.
8. The method for predicting the service life of a reinforced concrete bridge coupled with load and environmental factors according to claim 7, The decision support comprises the steps of generating a decision candidate set containing different maintenance opportunities and schemes based on the residual service life and reliability indexes; and performing simulation deduction on the candidate maintenance scheme in the digital twin model to evaluate the long-term effect of the candidate maintenance scheme, so as to recommend an optimal decision.

Description

Load and environmental factor coupled reinforced concrete bridge service life prediction method Technical Field The invention relates to the technical field of new generation information, in particular to a service life prediction method for a reinforced concrete bridge with load and environmental factor coupling. Background Reinforced concrete bridges serve as backbones of modern traffic infrastructure, and long-term safety and durability are directly related to public safety and economic development. Under the long-term coupling action of the vehicle load and the external environment (such as temperature, humidity, chloride ion corrosion and the like), the performance of the bridge structure is inevitably degraded, and the accurate prediction of the residual service life is important for making a scientific and reasonable maintenance strategy and guaranteeing the safety of the structure. Traditional prediction methods rely mainly on periodic manual detection and limited sensor data, combined with empirical mathematical models or simplified physical models for prediction. In recent years, with the development of sensing technology, a structural health monitoring system is widely applied to large bridges, and more abundant information is provided for performance evaluation by collecting data such as stress, strain, vibration and the like. However, the existing prediction method is difficult to realize high-frequency real-time coupling analysis of load and environmental factors, and lacks an effective mechanism for collaborative learning of multi-source data on the premise of protecting data privacy. Disclosure of Invention The invention aims to provide a service life prediction method for a reinforced concrete bridge with load and environmental factors coupled, which solves the technical problems that the prediction method in the prior art is difficult to realize high-frequency real-time coupling analysis of the load and the environmental factors and lacks an effective mechanism for collaborative learning of multi-source data under the premise of protecting data privacy. In order to achieve the above purpose, the invention provides a method for predicting the service life of a reinforced concrete bridge by coupling load and environmental factors, which comprises the following steps: the method comprises the steps of acquiring load data and environment data in real time through an intelligent sensing node network deployed at a key part of a bridge, and encrypting and transmitting the load data and the environment data to a preprocessing module; The preprocessing module performs preprocessing and feature extraction on the edge calculation level of the collected original data to generate a lightweight coupling feature vector; Based on the generated lightweight coupling feature vector, driving a digital twin model of the bridge to perform damage evolution simulation under the coupling effect of load and environmental factors, and outputting the degradation state and future trend of key performance indexes of the bridge; Utilizing a federal learning framework to aggregate anonymized data of a plurality of bridges, optimizing a global life prediction model, and calculating residual service life and reliability indexes based on the degradation state; and generating a visual report and maintenance decision advice according to the service life evaluation result. The specific mode for encrypting and transmitting the collected data to the preprocessing module is as follows: When receiving the transmission task, the system firstly applies the key manager to the key generator to generate the encryption key and the decryption key, and distributes the encryption key and the decryption key to the encryption module and the decryption module respectively. Then, the encryption module encrypts the acquired data by using the encryption key and transmits the ciphertext to the preprocessing module. The preprocessing module decrypts the ciphertext through the decryption key in the decryption module to obtain the original data, and then preprocessing and feature extraction operations are completed. The intelligent sensing node network comprises a load monitoring sensor and an environment monitoring sensor; The load monitoring sensor comprises a strain sensor and an acceleration sensor and is used for monitoring static and dynamic loads; The environment monitoring sensor comprises a temperature sensor, a humidity sensor and a chloride ion concentration sensor and is used for monitoring environmental corrosion factors; the intelligent sensing node has edge computing capability and can perform primary data cleaning and caching. The specific process for generating the lightweight coupling feature vector comprises the following steps of: Extracting characteristic quantities related to vehicle traffic from the load data by using a time-frequency analysis algorithm; performing space-time alignment and feature coding on the environmental d