Crack Monitoring And Identification Of Small Branch Pipes In Nuclear Power Plants Based On Distributed FBG Sensing

As an important part of the pressure-bearing boundary of a nuclear power plant,the structural integrity of small branch pipes is an important safeguard for the safe operation of nuclear power plants.However,in actual operation,small branch pipe failures occur from time to time,with cracking defects being the most frequent.The sprouting and expansion of cracks will pose a great threat to the pipeline or even the entire pressure-bearing system.Therefore,it is necessary to adopt a method to carry out crack detection for small branch pipes to identify the extent and location of cracks in a timely and accurate manner at the early stage of crack initiation and expansion,so as to ensure the safe operation of small branch pipes.This study focuses on the mechanism analysis,simulation and damage identification of small branch tubes in nuclear power plants,taking advantage of distributed "multi-point measurement" to obtain the circumferential strain distribution characteristics of the centre of circumferential cracks and its neighbours,and combining multiple fractals and their improvement methods to establish the mapping law between crack extent,location and characteristic parameters.The final result is the detection,extent and location identification of cracks in small branch pipes.The main research work is as follows:(1)Simulation of small branch pipes with different crack levels in nuclear power plants based on ANSYS finite element software.Based on the crack generation mechanism of small branch pipes and the theory of pipe circumferential strain monitoring,the ANSYS software was used to simulate different crack levels in small branch pipes,and the distribution characteristics of circumferential strain on the outer surface of the crack centre and its neighbours were investigated to obtain the best monitoring distance,which provided guidance for the construction of the experimental platform and the deployment of sensors.(2)A small centrifugal pump-pipe circulation experimental platform was designed and built,and the circumferential strain monitoring range obtained from the simulation data was used to guide the establishment of a sensor placement spacing scheme.In the small branch pipe crack monitoring experiments,the distributed Fiber Bragg Grating(FBG)strain sensors were also used to obtain the multi-point circumferential strain signals of the small branch pipe.The experimental results are within a reasonable error range with the simulation results,which verifies the correctness of the basic theory of monitoring crack defects in small branch pipes based on circumferential strain.(3)Damage identification and localisation of small branch pipes with different degrees and locations of cracks is carried out by using the improved MFDFA method.The experimental pipe model of small branch pipe is used as the object of analysis,and the multiple fractal characteristics of the small branch pipe crack signal are combined to compare the advantages of multiple fractal compared with single fractal method,and the combination of detrended fluctuation analysis and multiple fractal theory is used to extract the multiple fractal feature parameters and select the two feature parameters which are most sensitive to the crack degree and,so as to identify the crack damage degree.The multifractal detrended fluctuation analysis method is then improved by combining the features of empirical modal decomposition and least squares methods to provide higher sensitivity and better performance for the identification of crack levels.Moreover,the improved algorithm is used to compare the signal of a missing FBG with the distributed full signal of the multifractal spectrum,which can accurately locate the crack location and provide a basis for the identification and location of the crack degree of the actual small branch pipe.