Timothy Lardner

Robust defect detection in ultrasonic nondestructive evaluation (NDE) of difficult materials

In this paper, we present a novel and flexible method for reliable and robust defect detection in difficult materials. It is well known in the literature that the interaction between ultrasonic beams and the insonified medium is a highly nonlinear process, which potentially exhibits distinctive frequency-dependent properties for defects and random reflectors with a degree of randomness. Instead of investigating the structure and pattern of the spectrum of an individual echo, the proposed method focuses on the distinction between the ensembles of defect signals and clutter noise. A training process is used to establish the statistical analysis, based on which a hypothesis test is then applied to received echoes to detect defects. The approach is expected to be adaptive to the material microstructure and characteristics due to the statistical training. Experiments with a 5MHz transducer on austenitic steel samples from a coal fired power station are conducted. Austenitic steel is highly scattering and attenuating, and the method demonstrates accurate and reliable defect detection. When applied to A-scan waveforms, the grain noise is significantly reduced while defect signals are enhanced, and the signal-to-noise ratio (SNR) is improved by about 20dB. As a result, the defect is more visible and can be readily identified in B-scan images. Initial results indicate that this method is robust and delivers good performance without additional calibration and compensation.

A new speckle noise suppression technique using cross-correlation of array sub-apertures in ultrasonic NDE of coarse grain materials

In ultrasonic NDE, materials with coarse grains are difficult to inspect because a significant portion of energy will be reflected by the grains which cause speckle noise when imaging. Coarse grain materials are widely used in industry therefore a suitable inspection technique must be developed. This paper presents Spatially Averaged Sub-Array Correlation Imaging (SASACI), an advanced method that reduces speckle noise. SASACI was validated through comparison with the Total Focusing Method. A mean noise reduction of over 20dB SNR was observed when inspecting a highly scattering nickel alloy at 5MHz with a 128 element array.

Robust frequency diversity based algorithm for clutter noise reduction of ultrasonic signals using multiple sub-spectrum phase coherence

This paper presents a robust frequency diversity based algorithm for clutter reduction in ultrasonic A-scan waveforms. The performance of conventional spectral-temporal techniques like Split Spectrum Processing (SSP) is highly dependent on the parameter selection, especially when the signal to noise ratio (SNR) is low. Although spatial beamforming offers noise reduction with less sensitivity to parameter variation, phased array techniques are not always available. The proposed algorithm first selects an ascending series of frequency bands. A signal is reconstructed for each selected band in which a defect is present when all frequency components are in uniform sign. Combining all reconstructed signals through averaging gives a probability profile of potential defect position. To facilitate data collection and validate the proposed algorithm, Full Matrix Capture is applied on the austenitic steel and high nickel alloy (HNA) samples with 5MHz transducer arrays. When processing A-scan signals with unrefined parameters, the proposed algorithm enhances SNR by 20dB for both samples and consequently, defects are more visible in B-scan images created from the large amount of A-scan traces. Importantly, the proposed algorithm is considered robust, while SSP is shown to fail on the austenitic steel data and achieves less SNR enhancement on the HNA data.

Using phase information to enhance speckle noise reduction in the ultrasonic NDE of coarse grain materials

Materials with a coarse grain structure are becoming increasingly prevalent in industry due to their resilience to stress and corrosion. These materials are difficult to inspect with ultrasound because reflections from the grains lead to high noise levels which hinder the echoes of interest. Spatially Averaged Sub-Aperture Correlation Imaging (SASACI) is an advanced array beamforming technique that uses the cross-correlation between images from array sub-apertures to generate an image weighting matrix, in order to reduce noise levels. This paper presents a method inspired by SASACI to further improve imaging using phase information to refine focusing and reduce noise. A-scans from adjacent array elements are cross-correlated using both signal amplitude and phase to refine delay laws and minimize phase aberration. The phase-based and amplitude-based corrected images are used as inputs to a two-dimensional cross-correlation algorithm that will output a weighting matrix that can be applied to any conventional image. This approach was validated experimentally using a 5MHz array a coarse grained Inconel 625 step wedge, and compared to the Total Focusing Method (TFM). Initial results have seen SNR improvements of over 20dB compared to TFM, and a resolution that is much higher.

Automated full matrix capture for industrial processes

Full matrix capture (FMC) ultrasound can be used to generate a permanent re-focusable record of data describing the geometry of a part; a valuable asset for an inspection process. FMC is a desirable acquisition mode for automated scanning of complex geometries, as it allows compensation for surface shape in post processing and application of the total focusing method. However, automating the delivery of such FMC inspection remains a significant challenge for real industrial processes due to the high data overhead associated with the ultrasonic acquisition. The benefits of NDE delivery using six-axis industrial robots are well versed when considering complex inspection geometries, but such an approach brings additional challenges to scanning speed and positional accuracy when combined with FMC inspection. This study outlines steps taken to optimize the scanning speed and data management of a process to scan the diffusion bonded membrane of a titanium test plate. A system combining a KUKA robotic arm and a reconfigurable FMC phased array controller is presented. The speed and data implications of different scanning methods are compared, and the impacts on data visualization quality are discussed with reference to this study. For the 0.5 m2 sample considered, typical acquisitions of 18 TB/m2 were measured for a triple back wall FMC acquisition, illustrating the challenge of combining high data throughput with acceptable scanning speeds.

Fast ultrasonic phased array inspection of complex geometries delivered through robotic manipulators and high speed data acquisition instrumentation

Performance of modern robotic manipulators has enabled research and development of fast automated non-destructive testing (NDT) systems for complex geometries. This paper presents recent outcomes of work aimed at removing the bottleneck due to data acquisition rates, to fully exploit the scanning speed of modern 6-DoF manipulators. State of the art ultrasonic instrumentation has been integrated into a large robot cell to enable fast data acquisition, high scan resolutions and accurate positional encoding. A fibre optic connection between the ultrasonic instrument and the server computer enables data transfer rates up to 1.6GB/s. Multiple data collection methods are compared. Performance of the integrated system allows traditional ultrasonic phased array scanning as well as full matrix capture (FMC). In FMC configuration, linear scan speeds up to 156mm/s with 64 pulses per frame are achieved - this speed is only constrained by the acoustic wave propagation in the component. An 8x increase of the speed (up to 1.25m/s) can be achieved using multiple transmission elements, reaching the physical limits for acceptable acoustic alignment of transmission and reception paths. Scan results, relative to a 1.2m × 3m carbon fibre sample, are presented.

Inspection design using 2D phased array, TFM and cueMAP software

A simulation suite, cueMAP, has been developed to facilitate the design of inspection processes and sparse 2D array configurations. At the core of cueMAP is a Total Focusing Method (TFM) imaging algorithm that enables computer assisted design of ultrasonic inspection scenarios, including the design of bespoke array configurations to match the inspection criteria. This in-house developed TFM code allows for interactive evaluation of image quality indicators of ultrasonic imaging performance when utilizing a 2D phased array working in FMC/TFM mode. The cueMAP software uses a series of TFM images to build a map of resolution, contrast and sensitivity of imaging performance of a simulated reflector, swept across the inspection volume. The software takes into account probe properties, wedge or water standoff, and effects of specimen curvature. In the validation process of this new software package, two 2D arrays have been evaluated on 304n stainless steel samples, typical of the primary circuit in nuclear plants. Thick section samples have been inspected using a 1MHz 2D matrix array. Due to the processing efficiency of the software, the data collected from these array configurations has been used to investigate the influence sub-aperture operation on inspection performance.

Data-driven analysis of ultrasonic inspection data of pressure tubes

Abstract Pressure tubes are critical components of CANDU reactors and other pressurized heavy water–type reactors because they contain the nuclear fuel and the coolant. Manufacturing flaws as well as defects developed during in-service operation can lead to coolant leakage and can potentially damage the reactor. The current inspection process of these flaws is based on manually analyzing ultrasonic data received from multiple probes during planned, statutory outages. Recent advances in ultrasonic inspection tools enable the provision of high-resolution data of significantly large volumes. This highlights the need for an efficient autonomous signal analysis process. Typically, automation of ultrasonic inspection data analysis is approached by knowledge-based or supervised data-driven methods. This work proposes an unsupervised data-driven framework that requires no explicit rules or individually labeled signals. The framework follows a two-stage clustering procedure that utilizes the Density-Based Spatial Clustering of Applications with Noise density-based clustering algorithm and aims to provide decision support for the assessment of potential defects in a robust and consistent way. Nevertheless, verified defect dimensions are essential in order to assess the results and train the framework for unseen defects. Initial results of the implementation are presented and discussed, with the method showing promise as a means of assessing ultrasonic inspection data.

Wavelet analysis of poorly-focused ultrasonic signal of pressure tube inspection in nuclear industry

Pressure tube fabrication and installment challenges combined with natural sagging over time can produce issues with probe alignment for pressure tube inspection of the primary circuit of CANDU reactors. The ability to extract accurate defect depth information from poorly-focused ultrasonic signals would reduce additional inspection procedures, which leads to a significant time and cost saving. Currently, the defect depth measurement protocol is to simply calculate the time difference between the peaks of the echo signals from the tube surface and the defect from a single element probe focused at the back-wall depth. When alignment issues are present, incorrect focusing results in interference within the returning echo signal. This paper proposes a novel wavelet analysis method that employs the Haar wavelet to decompose the original poorly focused A-scan signal and reconstruct detailed information based on a selected high frequency component range within the bandwidth of the transducer. Compared to the original signal, the wavelet analysis method provides additional characteristic defect information and an improved estimate of defect depth with errors less than 5%.

A design methodology for 2D sparse NDE arrays using an efficient implementation of refracted-ray TFM

An algorithm has been developed that allows extremely efficient calculation of the total time of flight of an acoustic ray through two layer media, taking into account the effects of refraction through a 3D non-planar surface. The approach has been implemented on GP-GPU hardware, and embedded within the Total Focussing Method (TFM) imaging algorithm. This new software module supports arbitrary location of probe elements, array element directivity, arbitrary curved interface between two media, arbitrary transmit/receive sequences and any 1D/2D/3D image size for reconstructing the ultrasonic image from raw RF ultrasonic data. This allows calculation of the Point Spread Function of the probe at a range of points in its field of view, in a practical timeframe. Methodology, performance benchmarks, and possible applications are discussed.