Filtern
Dokumenttyp
- Beitrag zu einem Tagungsband (23)
- Vortrag (9)
- Zeitschriftenartikel (4)
- Posterpräsentation (2)
Schlagworte
- Simulation (14)
- Radiography (7)
- Modeling (5)
- Spectrum (5)
- Electron transport (4)
- Digital radiography (3)
- Elektronentransport (3)
- Probability of detection (3)
- X-ray (3)
- X-ray imaging (3)
Eingeladener Vortrag
- nein (9)
This work presents an investigation of the accuracy of a radiographic simulation model applied to subsea pipeline inspections. Experimental measurements of a sample in a water tank are used to develop a set of calibrated simulation parameters for the modelling software aRTist. Image quality parameters such as signal-tonoise ratio, contrast and basic spatial resolution are compared with the aim of matching simulated values to experimental results. With this method signal-to-noise ratio was successfully matched while differences were still found in contrast-to-noise ratio comparisons. This means that measurements depending on absolute intensity are not accurate enough, however wall thickness measurements in tangential images, which are not based on absolute intensity, were found to produce similar results in simulated and experimental cases. The differences in contrast and intensity are thought to be due to detector backscatter and additional scatter from out-of-setup objects within the exposure bay, due to a lack of source collimation. These would affect the experimental results but were not included in the simulated setup. This was investigated by including different proportions of peripheral water and other objects in the modelled setup and examining the effect on Image quality parameters. Results show that this additional scatter has a significant impact on the radiograph, particularly on image contrast, and is therefore the likely cause of differences between experimental and simulated images. This implies that it will be very difficult to completely match simulated to experimental results, as including all possible scattering objects in the model would be very complex. An improvement could be made by using real subsea data to estimate this additional scattering, which could then be used to calibrate the model. However there would still be significant uncertainty in the ability of the model to accurately produce realistic intensity and contrast.
This work presents the use of limited experimental measurements to develop a set of calibrated Simulation parameters that can then be used for reliable simulation of subsea pipeline inspections. The modelling software aRTist is used as the simulation tool, and the calibration is through comparison with experimental images of a well characterised sample in a water tank. Image Quality parameters such as signal-to-noise ratio, contrast and basic spatial resolution are compared with the aim of matching simulated values to experimental results. Currently the model is partially calibrated, with signal-to-noise ratio successfully matched while differences are still found in contrast-to-noise ratio comparisons. This means that measurements depending on absolute intensity are not accurate enough in the simulation at this stage. However, the simulation is found to be accurate for wall thickness measurements in tangential images, which are not based on absolute intensity, with simulated and experimental cases producing similar results.
One essential step on the way towards accurate quantitative simulation of radiographic testing is an accurate description of the utilized energy spectrum of X-ray photons. For use in general purpose simulation tools, the spectra of X-ray tubes have to be described by a model covering at least the intended range of applications. This range includes transmission tubes as well as direct beam tubes with varying angles of incidence and emission, for a number of typical target materials. In radiographic testing acceleration voltages frequently reach up to 450 kV for direct beam targets and up to 225 kV for transmission targets, with even higher voltages available or being developed. Currently used models are unable to cover the whole range of configurations.
Here a model is presented that employs a unified approach for simulating the photon energy spectra for transmission and direct beam targets composed of arbi-trary homogeneous materials. In order to achieve this, a detailed model of electron transport within the target is employed. The validity of the developed model is shown through comparisons with Monte Carlo simulations as well as measurements for a number of different configurations.
In this contribution, we discuss the influence of scattered radiation on materials’ effective attenuation coefficients at higher X-ray energies. The selected X-ray spectra for the dual-energy experiments correspond to 3 MV and 7.5 MV acceleration potential of the used betatron. Experiments were performed on a test phantom containing step wedges of different low- and high-Z materials. We evaluated the ratio between low- and high-energy X-ray attenuation coefficients quantitatively based on simulated poly-energetic high-energy X-ray source spectra and the detector sensitivity using the “analytical Radiographic Testing inspection simulation tool” (aRTist) developed at BAM. Furthermore, the influence of scattered radiation is evaluated using an efficient Monte-Carlo simulation. The simulation results are compared quantitatively with experimental investigations. Finally, important applications of the proposed technique in the context of aviation security are discussed.
Modelling becomes more and more important in modern NDE. It is increasingly used to optimize techniques for complex applications, to support the preparation of written procedures, and for education purposes. To describe the complete chain of RT, the model includes simulating all necessary properties of X- or Gamma-ray sources, the interaction of photons with material with special attention to scattered radiation, the detection process, and the complete geometrical RT setup handling arbitrary parts or constructions. Depending on the given inspection problem and the influencing factors that should be addressed by the simulation, an appropriate physical model has to be chosen to describe the underlying interaction mechanisms. The simulator aRTist combines analytical and Monte Carlo methods to efficiently model the radiation transport such that transmission as well as scatter techniques can be modelled. In this contribution we Focus on Monte Carlo Simulation of scatter contribution within aRTist. Examples for RT/tomographic applications and back-scatter techniques are presented to demonstrate the usability of the presented simulation tool for a broad range of radiological applications.
A fast quantitative model for the energy spectra of radiation emitted by X-ray tubes is described, handling thin as well as thick plane targets of arbitrary materials. The developed model is based on fundamental interaction cross sections, describing electron and photon transport via numerical integration of discretized distributions. While the focus lies on bremsstrahlung production, modeling of characteristic radiation is fully integrated. The model does not include any free parameters. The validity of the model is shown through comparison with measurements and Monte Carlo simulations for several combinations of target material and acceleration potential between 30 kV and 450 kV.
Modelling becomes more and more important in modern NDE. It is increasingly used to optimize techniques for complex applications and to support the preparation of written procedures. Hence, computer simulation nowadays has to be able to handle all significant properties of an NDE system with sufficient accuracy. In case of radiographic applications this includes models of the radiation source, of the interaction of radiation with materials, of the detection process, and the geometrical description of the part or the construction. Depending on the given inspection problem and the influencing factors that should be addressed by the simulation an appropriate physical model has to be chosen to describe the underlying interaction mechanisms. The simulator aRTist combines analytical and Monte Carlo methods to efficiently model the radiation transport. Simulation is a powerful tool to separate different influencing factors. Here we focus on Monte-Carlo simulation of scatter contributions within aRTist.
Radiography benchmark 2014
(2015)
The purpose of the 2014 WFNDEC RT benchmark study was to compare predictions of various models of radiographic techniques, in particular those that predict the contribution of scattered radiation. All calculations were carried out for homogenous materials and a mono-energetic X-ray point source in the energy range between 100 keV and 10 MeV. The calculations were to include the best physics approach available considering electron binding effects. Secondary effects like X-ray fluorescence and bremsstrahlung production were to be taken into account if possible. The problem to be considered had two parts. Part I examined the spectrum and the spatial distribution of radiation behind a single iron plate. Part II considered two equally sized plates, made of iron and aluminum respectively, only evaluating the spatial distribution. Here we present the results of above benchmark study, comparing them to MCNP as the assumed reference model. The possible origins of the observed deviations are discussed.
Correctly modeling the continuous photon spectrum of X-ray tubes requires detailed knowledge of the probability distribution of electron properties at the time of X-ray photon creation, in particular electron energy, depth within the target, and direction of movement. Semi-analytical X-ray spectrum models frequently assume a very simplified or even uniform distribution of electron direction. In the case of thick targets and small deviations from normal incidence this is a useful approximation. For thin targets or large deviations from normal incidence the correct distribution of electron directions becomes more important. As calculation speed is an important aspect of semi-analytical models compared to Monte Carlo simulations, fast evaluation of the distribution of electron properties is highly desirable. The approach presented here numerically evaluates the evolution of a discrete probability distribution of electron properties due to single electron scatter interactions within a plane target. This allows capturing the important aspects of the electron distribution while achieving runtimes of a few seconds up to a minute on a standard office PC.
Applications of RT Modelling
(2014)