基于小波分析的粗晶材料超声无损评价信号处理技术

噪声是降低超声检测可靠性的因素之一。本文利用小波分析对粗晶材料超声检测中由晶界散射引起的结构噪声进行了分析处理。和小波变换相比,小波包变换频域分辨率高,在信号的分解,降噪和重构时性能比前者好。对奥氏体不锈钢中人工缺陷的Ａ型超声信号的处理结果表明,结果噪声被有效地滤除,且不损失缺陷信号,信噪比明显提高。     

粗晶材料超声检测信号处理

粗晶材料超声检测时,严重的结构噪声使信噪比很低。为了提高信噪比,增加粗晶材料超声检测的可靠性,人们在数字信号处理方法中做了大量工作。信号平均、滤波、卷积、频谱分析及时频分析等信号处理方法都获得过应用。小波分析是一种时频分析技术,具有自适应的带宽,适合于时变的超声缺陷信号的处理;用于粗晶材料的信号处理时,比其它方法获得的信息要丰富得多,且能大幅度提高信噪比。     

Noise Reduction for Low-Dose Single-Slice Helical CT Sinograms

Helical computed tomography (HCT) has several advantages over conventional step-and-shoot CT for imaging a relatively large object, especially for dynamic studies. However, HCT may increase X-ray exposure significantly. This work aims to reduce the radiation by lowering X-ray tube current (mA) and filtering low-mA (or dose) sinogram noise of HCT. The noise reduction method is based on three observations on HCT: (1) the axial sampling of HCT projections is nearly continuous as detection system rotates; (2) the noise distribution in sinogram space is nearly a Gaussian after system calibration (including logarithmic transform); and (3) the relationship between the calibrated data mean and variance can be expressed as an exponential functional across the field-of-view. Based on the second and third observations, a penalized weighted least-squares (PWLS) solution is an optimal choice, where the weight is given by the mean-variance relationship. The first observation encourages the use of Karhunen-Loève (KL) transform along the axial direction because of the associated correlation. In the KL domain, the eigenvalue of each principal component and the derived data variance provide the signal-to-noise ratio (SNR) information, resulting in a SNR-adaptive noise reduction. The KL-PWLS noise-reduction method was implemented analytically for efficient restoration of large volume HCT sinograms. Simulation studies showed a noticeable improvement, in terms of image quality and defect detectability, of the proposed noise-reduction method over the Ordered-Subsets Expectation-Maximization reconstruction and the conventional low-pass noise filtering with optimal cutoff frequency and/or other filter parameters.     

Modeling of eddy current probe response for steam generator tubes

Sample calculations were performed with a three-dimensional (3D) finite-element model to describe the response of an eddy current (EC) probe to defects in steam generator (SG) tubing. Such calculations could be very helpful in understanding and interpreting EC probe response to complex tube/defect geometries associated with the inservice inspection (ISI) of SG tubes. The governing field equations are in terms of coupled magnetic vector and electric scalar potentials in conducting media and of total or reduced scalar potentials in nonconducting regions. To establish the validity of the model, comparisons of the theoretical and experimental responses of an absolute bobbin probe are given for two types of calibration standard defects. Simulation results are also presented on the effect of ligament size in axial cracks on bobbin probe response.     

An investigation of two-phase flow instability using wavelet signal extraction technique

When the oscillation periods of the instability of two-phase flow are sought with traditional methods of signal analysis, generally the Fourier transform must be employed and then the oscillation periods will be gotten at the location of the local maximum amplitude of frequency transform. However, Fourier transform will be difficult to clearly analyze the unsteady signals especially when the signals include many peaks and the noise interference is not generated by white noise in many areas of practical engineering like the oscillation of the instability of two-phase flow. The most effective solving method for the difficulty of Fourier transform is to analyze the signals directly in time domain. Wavelet analysis is able to search out the periods from time domain directly. It also has more excellent local characteristics than Fourier analysis in the both of time and frequency domains. In this paper, not only is a direct detecting method of the oscillation periods successfully applied based on the wavelet signal extraction techniques, but also the oscillation of density wave type of TYPE I is found as a kind of oscillations with a high-frequency harmonization.     

Flow measurements by noise analysis of thermocouple signals from the BOCA experiment at the Studsvik R2 reactor

At the Studsvik research reactor R2, a Boiling Capsule (BOCA) is used for long-term irradiation of BWR and PWR fuel rods. The BOCA experiment consists of a pressurised container that can hold a number of fuel rods in a bundle type configuration. The water flow inside the tube is driven by natural circulation. The coolant flow rate is not normally measured in the BOCA rig. Only thermocouples, measuring the water temperature at pertinent locations, are located inside the pressure tube. To confirm calculated values of the flow rate, transit time determination through the cross-correlation technique has been implemented.

Campaigns of noise measurements have been performed at five different occasions. The measurement campaigns have included 10 thermocouples at 3–4 different power levels. The results for the flow rate range between 0.15 and 0.35 m/s depending on reactor power level. The statistical accuracy of the results has also been evaluated. This paper shows that signal processing of thermocouple signals can be used to obtain rather accurate values of the flow rate in BOCA.     

Development of probability of detection functions for evaluating integrity of axial outside diameter stress corrosion cracks in steam generator

The reliability of an eddy current testing (ECT) inspection system depends upon the inspection technique and quality of analyst. In evaluating the integrity of a steam generator (SG) tube, degradation detection and sizing accuracy are considered performance measures of the nondestructive evaluation (NDE) system. A probability of detection (POD) model serves as a functional measure of the ability of an NDE system to detect degradation. It is one of the inputs in the operational assessment, and it is used to estimate the degradation during service via ECT of the SG tube. In this study, the POD functions of the inspection technique and analyst were obtained to quantitatively analyze the ECT bobbin probe for axial outside diameter stress corrosion cracks in SG tubes. This should serve to evaluate the integrity of the SG tubes. The depth and amplitude of defects were used as parameters of the POD model. Hit (detection) and miss (no detection) binary data obtained from destructive and nondestructive inspection of cracked tubes were also used.     

Time–frequency analysis of nonstationary complex magneto-hydro-dynamics in fusion plasma signals using the Choi–Williams distribution

The Choi–Williams distribution is applied to the time–frequency analysis of signals describing rapid magneto-hydro-dynamic (MHD) modes and events in tokamak plasmas. A comparison is made with Soft X-ray (SXR) signals as well as Mirnov signal that shows the advantages of the Choi–Williams distribution over both continuous wavelets transform scalogram and the short-time Fourier transform spectrogram. Examples of MHD activities in HT-7 and EAST tokamak are shown, namely the onset of coupling tearing modes, high frequency precursors of sawtooth, and low frequency MHD instabilities in edge localized mode (ELM) free in H mode discharge.     

Estimation and reduction of errors in flow measurements which use cross-correlation techniques

Propagation of inherent inhomogeneities or tracers in flowing liquids is used to measure flow. The peak value obtained from a cross-correlation analysis of downstream and upstream sensor signals is used to determine the tracer transit time and hence the flow. Performance of a “clamp-on” ultrasonic cross-correlation flowmeter developed by Canadian General Electric under contract from Atomic Energy of Canada Limited (AECL) was evaluated at the Chalk River Nuclear Laboratories of AECL. During this work, the effects of: (1) cross-correlation parameters, (2) filtering of correlator output and (3) use of an add-only polarity coincidence correlator on the standard deviation of transit time measurements were investigated both theoretically and experimentally. Comparison of ultrasonic cross-correlation flowmeter readings against gravimetric flow measurements gave a standard deviation of 2% of reading.     

A novel auto-correlation function method for the determination of the decay ratio in BWR stability analysis

A novel auto-correlation function (ACF) method has been investigated for determining the oscillation frequency and the decay ratio in BWR stability analyses. The neutron signals are band-pass filtered to separate the oscillation peak in the power spectral density (PSD) from background. Two linear second-order oscillation models are considered. These models, corrected for signal filtering and including a background term under the peak in the PSD, are then least-squares fitted to the ACF of the previously filtered neutron signal, in order to determine the oscillation frequency and the decay ratio. Our method uses fast Fourier transform techniques with signal segmentation for filtering and ACF estimation. Gliding ‘short-term’ ACF estimates on a record allow the evaluation of uncertainties. Numerical results are given which have been obtained from neutron data of the recent Forsmark I and Forsmark II NEA benchmark project. Our results are compared with those obtained by other participants in the benchmark project.     

Development and Systematic Application of Surface Inspection Methods Used for In-Service Inspection of Reactor Pressure Vessel

Both from the mechanics of fracture and from actual instances of defects observed in reactor pressure vessels, it is indicated that greater importance should be attached to surface than to internal defects in the in-service inspection of these components.

In the JPDR, the reactor pressure vessel has undergone ISI three times since 1968, with emphasis placed on surface inspection, and using both remote bore scope and remote Smeck methods. The two methods gave the same results on all three occasions, so that both methods can be considered effective.

With the remote Smeck method, the lower limit of detectable defect size was found to be less than ±lmm, and the reproducibility of defect position, measured on a mock-up nozzle was better than ±3 mm. For the bore scope a newly devised “shadowing technique” is described, which appreciably enhances its ability to detect and to accurately observe surface defects.

The results of ISI in JPDR indicate that significant improvements in detection ability and accuracy could be expected by systematic application of ISI methods.     

Development of Acoustic Leak Detection and Localization Methods for Inlet Piping of Fugen Nuclear Power Plant

The development work carried out on Fugen NPP is focused on detection of a small leakage on the reactor's inlet feeder pipes at an early stage by an acoustic leak detection method with usage of high-temperature resistant microphones. Specifically, the leak rate of 0.046 m³/h has been chosen as a target detection capability for this system. A cross-correlation technique has been studied for leak detection under low signal-noise ratios. The study shows that the sound diffusion on piping causes distortion of leak signals that results in their low correlation. A leak-location estimator and multi-channel correlation value, associated with estimated leak position, have been employed to detect such low-correlated leak signals. A method based on cross-correlation of signal spectral components has been proposed to deal with non-stationary leak signals. Joint-Time-Frequency-Analysis has been applied to analyze such signals, whilst a Wavelet decomposition technique has been used to extract their short-term spectral fluctuations. Since the spectral components are less affected by signal distortion, they provide higher correlation value and can be applied for leak detection under lower signal-noise ratios. The possibility of detecting and locating a small leakage by the methods proposed has been demonstrated by a number of simulation tests conducted on the Fugen NPP site.     

Implementing method of optimum front-end conditioner based on Butterworth filter

The front-end conditioner is an essential part of digital systems of nuclear spectrometer, which functions in two ways： （1） prevents saturation of the subsequent ADC; （2） limits the bandwidth of frequency to realize anti-aliasing. To realize the above-mentioned functions, an optimum front-end conditioner for a resistive feedback charge-sensitive preamplifier is designed. In the conditioner, the pole-zero compensation （P/Z compensation） technique was used to effectively filter signals from the preamplifier. The Butterworth filter was improved after the pole-zero position was optimally set up to shape the wave of output, which tallied with the whole system. The front-end conditioner can resolve the aberration of waveform of nuclear signals in a regular Butterworth filter. Compared with the traditional triple-pole filtering circuitry, the circuitry of this conditioner is more compact and flexible. Moreover, its output waveform is more symmetrical and the signal-to-noise ratio （SNR） is higher. The improvement in the resolution of spectrometer is also significant.     

Time dependence of linear stability parameters of a BWR

In this work we present several tools to study the time dependence of the linear stability parameters of a BWR using neutron noise analysis. Particularly, we have studied the variation in time of the fundamental frequency of a signal using the short-time Fourier transform and we have compared this method with the calculation of a time dependent Power Spectral Density (PSD) function. The temporal variation of the decay ratio is analysed using a method based on an autoregressive model to fit the different blocks of the signal. The performance of the tools presented is compared analysing analytic signals and a real signal of Forsmark 1&2 Stability Benchmark.     

Validation of reactor noise linear stability methods by means of advanced stochastic differential equation models

The aim of this paper is to show a validation method of a stability monitor using a BWR model with multiple Wiener noise sources, of additive and multiplicative nature. This model is solved using the modern methods to integrate stochastic differential equation systems, that are based on the stochastic Îto-Taylor expansion, and developed by Kloeden and Platen (1995), Kloeden et al. (1994). The synthetic signals generated with this BWR reduced order model with multiple Wiener processes are then used to obtain what are the optimal ways of filtering the signals for the different methods to estimate the decay ration (DR) and the natural frequency (ω) of the system. Also, for each DR estimation method, we study what is the optimal combination of algorithms to obtain the order and coefficients of the AR model that yields the best prediction of the reactor stability parameters for a broad range of DR values.     

Steam generator small bore piping socket weld inspection using the phased array ultrasonic technique

As nuclear power plants age, the likelihood of failures in the small bore piping used in those plants caused by exposure to mechanical vibrations during plant operations increases. While small bore piping failures rarely cause plant shutdown, the management of small piping has been a keen area of interest since their repair or maintenance requires a reactor trip. Steam generator (SG) drain pipe socket welds are small diameter piping connections (nominal pipe schedule 3–4 inches) susceptible to mechanical vibration. SG drain pipe socket weld failures have caused coolant leakage. Therefore, more reliable inspection technologies for small bore piping need to be developed to detect problems at an early stage and prevent pipe failures. This research aims to improve the reliability and accuracy of small bore piping inspections through the design, manufacture and application of a new phased array ultrasonic testing technique and inspection system for SG drain line socket welds.     

On the space cross-correlation of noise sources in boiling water reactors

In this paper, an attempt has been made to investigate the space cross-correlation of boiling noise in boiling water reactors by assuming the slip ratio to be unity and power profile to be constant along the core height of the reactor. Two cases have been considered: One, no flow fluctuations (ΔV = 0) caused by boiling noise and two, finite flow fluctuations (ΔV ≠ 0) caused by the boiling noise. It has been found that finite range space cross-correlation of noise sources exists only for ΔV ≠ 0 case and not for ΔV = 0 case. Auto power spectral density of steam content fluctuations, Δα have a break frequency which is directly related with the flow velocity and attenuation coefficient, μ of the cross-correlation of noise sources in BWRs. Normalized root mean square value of Δα is more sensitive to μ in the upper half of the core for μ ≤ 10. For $\text{μ ? 20}$, it is more sensitive to μ in the lower half of the core.     

The transport and diffusion theory of a line source in an infinite half-space with internal reflection

A closed form analytical solution is obtained for a three-dimensional transport theory problem, namely that of a line source in a half-space in one-speed transport theory with an internally reflecting surface. The theory is developed by using Fourier transforms in the transverse directions and a Laplace transform in the axial direction, together with the Wiener–Hopf technique. Both specular and diffuse internal reflection are considered and it is shown that a closed form solution is not available for the specular case but is available for the diffuse case. The surface particle scalar intensity and current are obtained and their sensitivity to absorption and the reflection coefficient are assessed. We also obtain a number of analytic asymptotic estimates for the intensity valid at large distances from the source. Diffusion theory is also employed and compared numerically with the transport solution. In general, diffusion theory gives very satisfactory results within its regions of limitation. We also offer this paper as an exercise in analytical transport theory; an aspect of reactor physics which is not often seen in recent times.     

The use of the noise analysis for the detection of the pellet- cladding mechanical interaction

The pellet-cladding mechanical interaction (PCMI) has been classically studied with the cladding elongation signal versus the rod power, but a different method can be used to evaluate the PCMI. It is based on the noise analysis techniques and uses the coherence function between the neutron flux and the cladding elongation noise signals detecting PCMI. This technique has been studied in the Halden experimental reactor. The mean coherence versus average linear heat rate shows two different regions that can be fitted with two straight lines, with the crossing point between them being taken as the threshold for PCMI. In addition, noise analysis in the time domain allows us to measure an elongation response time to a power step. This result can be applied to power instability events in BWRs.     

Artificial neural networks for neutron source localization within sealed tanks

A modular back-propagation ANN has been implemented for the non-destructive localization of a source of Even Plutonium Isotopes (EPI) contained in sealed tanks. The ANN has been trained on data obtained from a simulation of a well counter (filtered and Fourier transformed signals of the neutron detectors surrounding the well counter) for known positions of the EPI. After training, the ANN can predict the position of EPI within sealed tanks from the corresponding detector signals. The introduction of median and majority ANNs has been found to significantly improve the accuracy of prediction. Furthermore, these ANNs perform in a satisfactory manner when noise is injected to the detector signals; prediction is corrupted in a manner which is directly related to the extent and amount of noise.

The motivation for using back-propagation ANNs is twofold: on one hand (theoretical importance), they are capable of learning to approximate complex functions such as the strongly non-linear relation that exists between the neutron detector signals and the EPI position; on the other hand, they accomplish on-line localization which is of practical interest.