Nondestructive Testing Of Rebar Inside With Concrete Based On Soft-Measurement Technology

Whether the reinforced concrete structure is stable or not depends on whether the quality of the rebar used inside the structure meets the national standards.In related reports,the collapse of the building due to the poor quality of the rebar or using of small-diameter rebar illegally has occurred.Therefore,the detection of the internal rebar of the structure without damaging the structural quality,which the targets mainly include the orientation of the steel bar,the diameter of the steel bars,and the thickness of the protective layer is one of the most important issues in the construction industry that needs to be solved at present in China.This thesis is based on the eddy current testing method in non-destructive testing,and designed the detection scheme for the above detection indicators.The main research contents include:1?The plan for the non-destructive testing system of reinforcing steel was determined.According to the characteristics of reinforced concrete structures,the advantages and disadvantages of several common non-destructive testing methods were analyzed.The eddy current testing method was selected for in-depth research,and a pulsed eddy current array was used as an advanced technology for the design of the project;2?Based on pulsed eddy current array theory,a new coil detection device was designed,and the coil array sensor model designed in this thesis was theoretically analyzed.The hardware part selects STM32 chip as the overall control and operation component of the system.There were two parts software systems.The lower computer is mainly designed for C program based on STM32,and the overall system control was completed.The upper computer part adopts Qt for interface designed and completed the display and storage of data;3?the establishment of mathematical models.After collecting and corresponding the sample data through a large number of experiments,the mathematical models was established,which the detection coil voltage signal as an independent variable and The bar diameter and protective layer thickness as dependent variables,based on BP neural network algorithm,support vector machine regression algorithm and partial least-squares regression algorithm,respectively.By comparing the three models of detection error and variance and other indicators,selected the optimal model for design.Mathematical modeling experiments show that the partial least-squares regression model has large variance and cannot meet the detection accuracy requirements of the thesis among the three models;the BP neural network model and the support vector machine regression model have similar results and can be obtained in most cases,but the larger error points occurred in the experiment.In order to solve the large error points,this thesis improved the algorithm based on the BP neural network model and designed an improved BP neural network model.Experiments show that the model has a good detection effect,and effectively avoid the emergence of larger error points.Compared with traditional rebar detection methods,the subject has the advantages of no damage to the measured object and high accuracy of the detection results.Combined with the pulsed eddy current array detection module and the modeling algorithm,the diameter of the rebar and the thickness of the protective layer can be detected at the same time,making up for the shortcomings of similar products in the current market.