Research And Design Of LBB System For Main Pipeline Of Nuclear Power Plant

The main pipeline of a nuclear power plant is an important part of the primary coolant pressure boundary.The integrity of nuclear power plant main pipeline is significance for the safe and reliable operation of nuclear power system.In order to avoid the corresponding nuclear power safety accidents caused by the leakage of the main pipeline,a large number of safety facilities have been built in the design and construction stage of the nuclear power system with the double end shear fracture of the main pipeline as the design basis accident.Complexity and design build costs of nuclear power systems are increased,due to the presence of these safety facilities.It is proved that the possibility of double-end shear fracture of main pipeline is very small by multi-year experience of nuclear power operation.If the cracks and small leakage in the main pipeline of the nuclear power plant can be detected in time and effective measures can be taken,a large number of safety facilities for the double end shear fracture accident can be avoided,the safe and stable operation of the nuclear power plant can be ensured while reducing the complexity and construction cost of the nuclear power system.Based on the above considerations,the acoustic emission LBB system for the main pipeline is studied and designed,which aims to ensure that the cracks and small size leakage caused by the pipeline are monitored in time,so that the occurrence of the cracks and double end shear accidents are avoided.First of all,the mechanical analysis and acoustic emission signal simulation of heat leg of main pipe of AP1000 reactor are carried out.The finite element method of the ANSYS software platform is used to analyze the stress of the main pipeline without cracks,the most prone to defects on the pipeline is determined.A semi-ellipsoidal crack and an elliptical cross-sectional crack((i.e.leakage)on the inner surface of the pipeline are established for fracture mechanics analysis.The acoustic emission signals of the main pipe in the three states of no crack,crack and leakage are obtained by transient fluid structure coupling and transient dynamic analysis.Through the preliminary time-frequency analysis,the defect information of the main pipeline is mainly concentrated in the high-frequency section.Secondly,the method of acoustic emission signal processing of the main pipeline is studied.The improved threshold wavelet and wavelet packet algorithms are used to denoise the noise-added acoustic emission signals,and the two denoising algorithms are evaluated through three evaluation criteria: SNR,similarity and mutual information.Finally,wavelet packet is selected to denoise the acoustic emission signal.The wavelet packet-optimized MF-DFA combined feature extraction method is used to extract the energy spectrum and waveform characteristic parameters of the acoustic emission signal.Based on the LVQ neural network,50 groups of samples in three states are learned,and the classification effect of the LVQ neural network is tested by 70 test samples different from the training samples.The number of correctly judged groups reached 70,which proved that the classification effect can meet the classification requirements.Lastly,the quantitative analysis of the signal energy and the geometric size of the pipeline leakage proves that there is a linear relationship between the logarithm of them.Based on the cross-correlation,the propagation velocity of acoustic emission wave in the pipeline is fitted,and the velocity of acoustic emission wave is between the S-wave velocity and the P-wave velocity calculated theoretically.The leakage signals of different axial positions are used to verify the wave velocity and time difference obtained by this method.Under the simulation experiment conditions in this paper,the average value of the axial positioning error of this method is 6.16%,which meets the requirements of rg1.45 for LBB system positioning.It is verified that the signal energy of the monitoring point near the leakage point in the radial direction is significantly higher than that of other monitoring points in the same radial interface,which can be used to locate the pipeline leakage in the radial direction.