Abstract:The paper presents an effective approach to the damage identification and discrimination of the continuous bridges with the static data based on the static strain.For the research purpose,we have chosen the pseudo strain energy as the damage indicator,which is more sensitive to the bridge damage than the strain when used for damage identification.In order to overcome the local minima problem of the BP neural network and optimize the weights and thresholds of the BP neural network,we have suecessfully developed a PSO-BP prediction model by using the improved PSO.Furthermore,we have simulated the damage levels of the bridges and managed to apply the model to the continuous bridge damage identification by declining the element rigidity,and working out the static response data of the damaged bridges,such as the static strain used for training and testing the PSO-BP prediction model.What is more,to verify the effectiveness of the method,we have also established a finite dement model of three span continuous bridges based on the Midas/civil approach.Seeing the complexity of the calculation model,we have also selected the vulnerable sections of the bridge as the objective of our study.Besides,we have also adopted the pseudo strain energy ratios of the above said areas to be the input parameters for the neural network,with the validity of the model being analyzed on the Matlab.The results of our exploration show that the PSO-BP neural network has been highly competitive and compatible in identifying and determining either the single location and the multiple locations damage.Not only can it identify the spots of the structural damage but also determine the degrees of damage.What is more,it can identify the maximum relative error of the damage degree both in the single locations and the multiple locations individually to an accuracy of 2.8％ and 3.14％.Therefore,the proposed method can be expected to produce more satisfactory results in single location identification.In addition,for the damage severity identification,the network interpolation capability also proves superior to the demanded extent.As compared with the traditional BP neural network,the said prediction model can be said to have achieved a more precisely predictable capability in general.Thus,it can be seen that,although the method is proposed for the continuous bridges,it can also be entitled with a referential significance for other similar bridge systems.

Remaining Pre-Stress Identification of Diseased Pre-Stressed Concrete Bridges Based on Neural Network

The Application of a Damage Detection Method Using Artificial Neural Network and Train-Induced Vibrations on a Simplified Railway Bridge Model

DAMAGE LOCATING OF CABLE-STAYED BRIDGES BASED ON DYNAMIC MEASUREMENT AND NEURAL NETWORK TECHNIQUE

Damage identification and discrimination method for the continuous bridges based on the strain and PSO-BP neural network

Damage identification method of prestressed concrete beam bridge based on convolutional neural network

Examination of load-deformation characteristics of long-span bridges in harsh natural environments based on real-time updating artificial neural network

Research on Methods of Damage Diagnosis Based on Bridges with Dynamic Characteristics

A study on the application of an improved adaptive neural network in prestressed bridge engineering inspection

Research on Intelligent Inspection Method of Prestressed Bridge Engineering Based on Machine Learning

Determination of Bridge Prestress Loss under Fatigue Load Based on PSO-BP Neural Network

Step-by-step Detection for Damage in Cable-stayed Bridges Based on Neural Network Technology

Computer Vision-Based Bridge Damage Detection Using Deep Convolutional Networks with Expectation Maximum Attention Module

Application of Neural Network in Reliability Analysis of Prestressed Concrete Bridges

Study on Modern Bridge Structure Health Monitoring System Based on Damage Identification

Finite Element–based Data-Driven Method to Detect Multiple Damages of 1D Beam Model and 2D Slab Model of Bridges-a Theoretical and Experimental Study

Bridge Structural Damage Identification Based on Genetic Optimization Neural Networks Algorithm

Neural Networks-Based Damage Detection for Bridges Considering Errors in Baseline Finite Element Models

A Simplified ANN-based Evaluation Method for Creep Sensitive Girder Bridge Identification at Preliminary Design Stage

Enhanced ANN-based ensemble method for bridge damage characterization using limited dataset

A stress-development prediction method and its application to stress assessment of existing bridges

Damage Identification Of Simply Supported Bridge Based On Rbf Neural Network