458 research outputs found
An evaluation of Finite-Difference and Finite-Integration Time-Domain modelling tools for Ground Penetrating Radar antennas
Get PDFThe development of accurate and realistic models of Ground Penetrating Radar (GPR) antennas is being driven by research into quantitative amplitude information from GPR, improved GPR antenna designs, and better-performing forward simulations that can feed into inversion algorithms. The Finite-Difference Time-Domain (FDTD) method and Finite-Integration technique (FIT) are popular numerical methods for simulating electromagnetic wave propagation. Time-Domain methods are particularly well-suited to modelling ultra-wideband GPR antennas as a broad range of frequencies can be modelled with a single simulation. We present comparisons using experimental and simulated data from a Geophysical Survey Systems 1.5 GHz antenna and a MALÅ Geoscience 1.2 GHz antenna. The antennas were investigated in free space and over a lossy dielectric environment with a target. For the simulations we used a commercial solver - Computer Simulation Technology Microwave Studio (CST) - and a free open-source FDTD solver - gprMax. For each test scenario, phase and amplitude information from the antenna responses were compared. Generally, we found very good agreement between the experimental data and the two simulations
Numerical modelling of high-frequency ground-penetrating radar antennas
Get PDFGround-Penetrating Radar (GPR) is a non-destructive electromagnetic investigative
tool used in many applications across the fields of engineering and
geophysics. The propagation of electromagnetic waves in lossy materials is
complex and over the past 20 years, the computational modelling of GPR has
developed to improve our understanding of this phenomenon.
This research focuses on the development of accurate numerical models of
widely-used, high-frequency commercial GPR antennas. High-frequency, highresolution
GPR antennas are mainly used in civil engineering for the evaluation
of structural features in concrete i. e., the location of rebars, conduits, voids
and cracking. These types of target are typically located close to the surface
and their responses can be coupled with the direct wave of the antenna. Most
numerical simulations of GPR only include a simple excitation model, such as an
infinitesimal dipole, which does not represent the actual antenna. By omitting
the real antenna from the model, simulations cannot accurately replicate the
amplitudes and waveshapes of real GPR responses.
Numerical models of a 1.5 GHz Geophysical Survey Systems, Inc. (GSSI) antenna
and a 1.2 GHz MALÅ GeoScience (MALÅ) antenna have been developed.
The geometry of antennas is often complex with many fine features that must be
captured in the numerical models. To visualise this level of detail in 3d, software
was developed to link Paraview—an open source visualisation application which
uses the Visualisation Toolkit (VTK)—with GprMax3D—electromagnetic simulation
software based on the Finite-Difference Time-Domain (FDTD) method.
Certain component values from the real antennas that were required for the
models could not be readily determined due to commercial sensitivity. Values
for these unknown parameters were found by implementing an optimisation
technique known as Taguchi’s method. The metric used to initially assess
the accuracy of the antenna models was a cross-corellation of the crosstalk
responses from the models with the crosstalk responses measured from the real
antennas. A 98 % match between modelled and real crosstalk responses was
achieved.
Further validation of the antenna models was undertaken using a series of
laboratory experiments where oil-in-water (O/W) emulsions were created to
simulate the electrical properties of real materials. The emulsions provided
homogeneous liquids with controllable permittivity and conductivity and enabled
different types of targets, typically encountered with GPR, to be tested. The laboratory setup was replicated in simulations which included the antenna
models and an excellent agreement was shown between the measured and
modelled data. The models reproduced both the amplitude and waveshape of
the real responses whilst B-scans showed that the models were also accurately
capturing effects, such as masking, present in the real data. It was shown
that to achieve this accuracy, the real permittivity and conductivity profiles of
materials must be correctly modelled.
The validated antenna models were then used to investigate the radiation
dynamics of GPR antennas. It was found that the shape and directivity of
theoretically predicted far-field radiation patterns differ significantly from real
antenna patterns. Being able to understand and visualise in 3d the antenna
patterns of real GPR antennas, over realistic materials containing typical
targets, is extremely important for antenna design and also from a practical
user perspective
Deep learning processing and interpretation of ground penetrating radar data using a numerical equivalent of a real GPR transducer
Get PDFGround-Penetrating Radar (GPR) is a popular non-destructive electromagnetic (EM) technique that is used in diverse applications across different fields, most commonly geophysics and civil engineering. One of the most common applications of GPR is concrete scanning, where it is used to detect structural elements and support the assessment of its condition. However, in any GPR application, the data have no resemblance to the characteristics of targets of interest and a means of extracting information from the data regarding the targets is required.
Interpreting the GPR data, to infer key properties of the subsurface and to locate the targets is a difficult and challenging task and is highly dependent on the processing of the data and the experience of the user. Traditional processing techniques have some drawbacks, which can lead to misinterpretations of the data in addition to the interpretation being subjective to the user. Machine learning (ML) has proven its ability to solve a variety of problems and map complex relationships and in recent years, is becoming an increasingly attractive option for solving GPR and other EM problems regarding processing and interpretation. Numerical modelling has been extensively used to understand the EM wave propagation and assist in the interpretation of GPR responses. If ML is combined with numerical modelling, efficient solutions to GPR problems can be acquired.
This research focuses on developing a numerical equivalent of a commercial GPR transducer and utilising this model to produce realistic synthetic training data sets for deep learning applications. The numerical model is based on the high-frequency 2000 MHz "palm" antenna from Geophysical Survey Systems, Inc. (GSSI). This GPR system is mainly used for concrete scanning, where the targets are located close to the surface. Unknown antenna parameters were found using global optimisation by minimising the mismatch between synthetic and real responses. A very good match was achieved, demonstrating that the model can accurately replicate the behaviour of the real antenna which was further validated using a number of laboratory experiments. Real data were acquired using the GSSI transducer over a sandbox and reinforced concrete slabs and the same scenarios were replicated in the simulations using the antenna model, showing excellent agreement.
The developed antenna model was used to generate synthetic data, which are similar to the true data, for two deep learning applications, trained entirely using synthetic data. The first deep learning application suggested in the present thesis is background response and properties prediction. Two coupled neural networks are trained to predict the background response given as input total GPR responses, perform background removal and subsequently use the predicted background response to predict its dielectric properties. The suggested scheme not only performs the background removal processing step, but also enables the velocity calculation of the EM wave propagating in a medium using the predicted permittivity value. The ML algorithm is evaluated using a number of synthetic and measured data demonstrating its efficiency and higher accuracy compared to traditional methods. Predicting a permittivity value per A-scan included in a B-scan results in a permittivity distribution, which is used along with background removal to perform reverse-time migration (RTM). The proposed RTM scheme proved to be superior when compared with the commonly used RTM schemes.
The second application was a deep learning-based forward solver, which is used as part of a full-waveform inversion (FWI) framework. A neural network is trained to predict entire B-scans given certain model parameters as input for reinforced concrete slab scenarios. The network makes predictions in real time, reducing by orders of magnitude the computational time of FWI, which is usually coupled with an FDTD forward solver. Therefore, making FWI applicable to commercial computers without the need of high-performance computing (HPC). The results clearly illustrate that ML schemes can be implemented to solve GPR problems and highlight the importance of having a digital representation of a real transducer in the simulations
Comparison of Time-Domain Finite-Difference, Finite-Integration, and Integral-Equation Methods for Dipole Radiation in Half-Space Environments
Get PDFIn this paper we compare current implementations of commonly used numerical techniques - the Finite-Difference Time-Domain (FDTD) method, the Finite-Integration Technique (FIT), and Time-Domain Integral Equations (TDIE) - to solve the canonical problem of a horizontal dipole antenna radiating over lossless and lossy half-spaces. These types of environment are important starting points for simulating many Ground Penetrating Radar (GPR applications which operate in the near-field of the antenna, where the interaction among the antenna, the ground, and targets is important. We analysed the simulated current at the centre of the dipole antenna, as well as the electric field at different distances from the centre of the antenna inside the half-space. We observed that the results from the simulations using the FDTD and FIT methods agreed well with each other in all of the environments. Comparisons of the electric field showed that the TDIE technique agreed with the FDTD and FIT methods when observation distances were towards the far-field of the antenna but degraded closer to the antenna. These results provide evidence necessary to develop a hybridisation of current implementations of the FDTD and TDIE methods to capitalise on the strengths of each technique
Some Uninterpreted to the New Method and Unpublished Compositions in Post-Byzantine Psaltic Manuscripts. Katavasiai for the Sunday before Christmas, Composed by Priest Antonios the Nomophylax
Get PDFThe three teachers (Chrysanthos, Gregorios and Chourmouzios) established and taught the New Method of ecclesiastical music (1814-1820) and interpreted an enormous corpus of Byzantine and post-Byzantine compositions to it. However, while studying the psaltic manuscripts we also detect many musical works of well-known composers which hadn’t been transcribed to this analytical musical notation by them, nor by their posteriors. In the paper with the aforementioned title I will present some interesting compositions of this category and I will focus on a specific work of Antonios the priest which I transcribed in the nowadays musical notation
A Novel Piecewise Linear Recursive Convolution Approach for Dispersive Media Using the Finite-Difference Time-Domain Method
Get PDFTwo novel methods for implementing recursively the convolution betweenthe electric field and a time dependent electric susceptibility function inthe finite-difference time domain (FDTD) method are presented. Both resultingalgorithms are straightforward to implement and employ an inclusive susceptibilityfunction which holds as special cases the Lorentz, Debye, and Drude mediarelaxations. The accuracy of the new proposed algorithms is found to be systematicallyimproved when compared to existing standard piecewise linear recursive convolution(PLRC) approaches, it is conjectured that the reason for this improvementis that the new proposed algorithms do not make any assumptions about thetime variation of the polarization density in each time interval; no finitedifference or semi-implicit schemes are used for the calculation of the polarizationdensity. The only assumption that these two new methods make is that the firsttime derivative of the electric field is constant within each FDTD time interval
Numerical modelling of ground penetrating radar for optimization of the time-zero adjustment and complex refractive index model
Get PDFTime-zero adjustment or the true ground surface for Ground Penetrating Radar (GPR) applications is a very important aspect and an essential
factor in order to carry out accurate shallow depth measurements. As
the transmitted and received signals from GPR antennas are affected
by the presence of different materials with various dielectric constants
and electromagnetic properties adjusting the time-zero appropriately is
important. This study uses a realistic Three Dimensional (3D) numerical
model of a GPR transducer in order to examine where is the best location
for time-zero on a GPR trace. It is shown that in order to establish a
robust and consistent time-zero position careful consideration is needed
also of the way the two-way travel time of the reflected GPR wavelet is
estimated as well. Starting with a simple homogeneous model with a set
of different targets a better process of time-zero adjustment and time
picking of the GPR wavelets is put forward that is verified using further
more complex and realistic heterogeneous models. Further verification
is obtained by using experimental data.
Estimating the permittivity of heterogeneous mixtures based on the
permittivity of their individual components is of high importance with
many applications in GPR and in electrodynamics-based sensing in
general. The Complex Refractive Index Model (CRIM) is the most
mainstream approach for estimating the bulk permittivity of heterogeneous materials and has widely been applied for GPR applications. The
popularity of CRIM is primarily based on its simplicity while its accuracy
has never been rigorously tested. In the current study, an optimized
shape factor is derived that is fine-tuned for modelling the dielectric
properties of concrete. The bulk permittivity of concrete is expressed
with respect to its components i.e, aggregate particles, cement particles,
air-void and volumetric water fraction. Different combinations of the
above materials are accurately modelled using the Finite-Difference
Time-Domain (FDTD) method. The numerically estimated bulk permittivity is then used to fine-tune the shape factor of the CRIM model.
Then, using laboratory measurements it is shown that the revised CRIM
model over-performs the default shape factor and provides with more
accurate estimations of the bulk permittivity of concrete.
Numerical modelling of a heterogeneous concrete model and a bowtie
antenna with a separate transmitter and receiver that are able to move
independently are also presented in this study. Both models are used for
the optimisation of the time-zero position and the CRIM model shape
factor
Realistic numerical modelling of ground penetrating radar for landmine detection
Get PDFGround-Penetrating Radar (GPR) is a popular non-destructive geophysical
technique with a wide range of diverse applications. Civil engineering, hydrogeophysics,
forensic, glacier geology, human detection and borehole geology are
some of the fields in which GPR has been applied with successful or promising
results. One of the most mainstream applications of GPR is landmine detection.
A lot of methods have been suggested over the years to assist the landmine
detection issue. Metal detectors, trained rats or dogs, chemical methods and
electrical resistivity tomography are –amongst others– some of the suggested
techniques. The non-destructive nature of GPR makes it an attractive choice for
a problem such as demining in which contact to the ground is not allowed. The
main advantage of GPR is its ability to detect both metallic and non-metallic
targets. Furthermore, GPR can provide an insight regarding the nature of the
target (e.g. size, burial depth, type). From the above, it is evident that GPR
can potentially reduce the false alarms emerging from small metallic objects
(e.g. bullets, wires, etc.) usually encountered in battle-fields and industrialised
areas. Combining the robustness of the metal detector with the resolution of
GPR results in a reliable and efficient detection framework which has been
successfully applied in Cambodia and Afghanistan.
Despite the promising, and in some cases impressive results, aspects of
GPR can be further improved in an effort to optimise GPR’s performance and
decrease its limitations. The validation of a GPR system is usually achieved
through the so called Receiver Operation Characteristics (ROC) which depicts
the probability of detection with respect to the false alarm rate. ROC is a
highly nonlinear function which is sensitive to the environment as well as to
the antenna unit.
Landmines are typically small objects, often less than 10 cm diameter, which
are shallow buried, usually in less than 10 cm depth, and sometimes almost
exposed. In order for the landmines to be resolved, high frequency antennas
are essential. The latter are sensitive to soil’s inhomogeneities, rough surface,
water puddles, vegetation and so on. Apart from that, the near field nature of
the problem makes the antenna unit part of the medium which contributes to
the unwanted clutter. The above, outlines the multi-parametric nature of the
problem for which no straightforward approach has yet to be proposed.
Numerical modelling is a practical and solid approach to understand the
physical behaviour of a system. In the case of GPR for landmine detection,
numerical modelling can be a practical tool for designing and optimising
antennas in synthetic but nonetheless realistic conditions. Apart from that,
evaluation of a processing method only to a specific environment is not a
robust approach and does not provide any evidence for its wider inclusivity and
limitations. However, evaluation in different conditions can become costly and
unpractical. Numerical modelling can tackle this problem by providing data for
a wide range of scenarios. An extensive database of simulated responses, apart
from being a practical testbed, can be also employed as a training set for machine
learning. A multi-variable problem like demining, in order to be addressed using
machine learning, requires a large amount of data. These must equally include
all possible different scenarios i.e. different landmines, in different media with
stochastically varied properties and topography. Additionally, different heights
of the antenna and different depths of the landmines must also be examined.
Numerical modelling seems to be a practical approach to achieve an equally
distributed and coherent dataset like the one briefly described above.
Numerical modelling of GPR for landmine detection has been applied in
the past using generic antennas in simplified and clinical scenarios. This
approach can be used in an educational context just to provide a rough
estimation of GPR’s performance. In the present thesis a realistic numerical
scheme is suggested in which, simplifications are kept to a minimum. The
numerical solver, employed in the suggested numerical scheme, is the Finite-
Difference Time-Domain (FDTD) method. Both the dispersive properties and
the Absorbing Boundary Condition (ABC) are implemented through novel
and accurate techniques. In particular, a novel method which implements an
inclusive susceptibility function is suggested and it is shown that surpasses the
performance of the previous approaches while retaining their computational
efficiency. Furthermore, Perfectly Matched Layer (PML) and more specifically
Convolutional Perfectly Matched Layer (CPML) is implemented through a novel
time-synchronised scheme which it is proven to be more accurate compared to
the traditional CPML with no additional computational requirements.
An accurate numerical solver, although essential, is not the only requirement
for a realistic numerical framework. Accurate implementation of the geometry
and the dielectric properties of the simulated model is highly important,
especially when it comes to high-frequency near-field scenarios such as GPR
for landmine detection. In the suggested numerical scheme, both the soil’s
properties as well as the rough surface are simulated using fractal correlated
noise. It is shown, that fractals can sufficiently represent Earth’s topography
and give rise to semi-variograms often encountered in real soils. Regarding the
dielectric properties of the soils, a semi-analytic function is employed which
relates soil’s dielectric properties to its sand fraction, clay fraction, sand density,
bulk density and water volumetric fraction. Subsequently, the semi-analytic
function is approximated using a Debye function that can be easily implemented
to FDTD. Vegetation is also implemented to the model using a novel method
which simulates the geometry of vegetation through a stochastic process. The
experimentally-derived dielectric properties of vegetation are approximated
–similarly to soil’s dielectric properties– with a Debye expansion. The antenna
units tested in the numerical scheme are two bow-tie antennas based on
commercially available transducers. Regarding the targets, three landmines are
chosen, namely, PMN, PMA-1 and TS-50. Dummy landmines are used in order
to obtain their geometrical characteristics and comparison between measured
and numerically evaluated traces are used to tune the dielectric properties of
the modelled landmines. Lastly, water puddles are realistically implemented in
the model in an effort to realistically simulate high-saturated scenarios.
The proposed numerical scheme has been employed in order to test and
evaluate widely used post-processing methods. The results clearly illustrate
that post-processing methods are sensitive to the antenna unit as well as the
medium. This highlights the importance of an accurate numerical scheme as a
testbed for evaluating different GPR systems and post-processing approaches
in wide range of scenarios.
Using an equivalent 2D numerical scheme –restricted to 2D due to computational
constrains– preliminary results are given regarding the effectiveness of
Artificial Neural Network (ANN) subject to an adequate and equally distributed
database. The results are promising, showing that ANN can be successfully
employed for detection as well as classification using only a single trace as
input. A basic requirement to do so is a representative training set. This can
be synthetically generated using a realistic numerical framework. The above,
provide solid arguments for further expanding the proposed machine learning
scheme to the more computationally demanding 3D case
gprMax: Open source software to simulate electromagnetic wave propagation for Ground Penetrating Radar
Get PDFAbstractgprMax is open source software that simulates electromagnetic wave propagation, using the Finite-Difference Time-Domain (FDTD) method, for the numerical modelling of Ground Penetrating Radar (GPR). gprMax was originally developed in 1996 when numerical modelling using the FDTD method and, in general, the numerical modelling of GPR were in their infancy. Current computing resources offer the opportunity to build detailed and complex FDTD models of GPR to an extent that was not previously possible. To enable these types of simulations to be more easily realised, and also to facilitate the addition of more advanced features, gprMax has been redeveloped and significantly modernised. The original C-based code has been completely rewritten using a combination of Python and Cython programming languages. Standard and robust file formats have been chosen for geometry and field output files. New advanced modelling features have been added including: an unsplit implementation of higher order Perfectly Matched Layers (PMLs) using a recursive integration approach; diagonally anisotropic materials; dispersive media using multi-pole Debye, Drude or Lorenz expressions; soil modelling using a semi-empirical formulation for dielectric properties and fractals for geometric characteristics; rough surface generation; and the ability to embed complex transducers and targets.Program summaryProgram title: gprMaxCatalogue identifier: AFBG_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AFBG_v1_0.htmlProgram obtainable from: CPC Program Library, Queen’s University, Belfast, N. IrelandLicensing provisions: GNU GPL v3No. of lines in distributed program, including test data, etc.: 627180No. of bytes in distributed program, including test data, etc.: 26762280Distribution format: tar.gzProgramming language: Python.Computer: Any computer with a Python interpreter and a C compiler.Operating system: Microsoft Windows, Mac OS X, and Linux.RAM: Problem dependentClassification: 10.External routines: Cython[1], h5py[2], matplotlib[3], NumPy[4], mpi4py[5]Nature of problem: Classical electrodynamicsSolution method: Finite-Difference Time-Domain (FDTD)Running time: Problem dependentReferences:[1]Cython, http://www.cython.org[2]h5py, http://www.h5py.org[3]matplotlib, http://www.matplotlib.org[4]NumPy, http://www.numpy.org[5]mpi4py, http://mpi4py.scipy.or
Rethinking construction cost overruns: an artificial neural network approach to construction cost estimation
Get PDFThe main concern of a construction client is to procure a facility that is able to
meet its functional requirements, of the required quality, and delivered within an
acceptable budget and timeframe. The cost aspect of these key performance
indicators usually ranks highest. In spite of the importance of cost estimation, it is
undeniably neither simple nor straightforward because of the lack of information
in the early stages of the project. Construction projects therefore have routinely
overrun their estimates.
Cost overrun has been attributed to a number of sources including technical error
in design, managerial incompetence, risk and uncertainty, suspicions of foul play
and even corruption. Furthermore, even though it is accepted that factors such as
tendering method, location of project, procurement method or size of project
have an effect on likely final cost of a project, it is difficult to establish their
measured financial impact. Estimators thus have to rely largely on experience and
intuition when preparing initial estimates, often neglecting most of these factors
in the final cost build-up. The decision-to-build for most projects is therefore
largely based on unrealistic estimates that would inevitably be exceeded.
The main aim of this research is to re-examine the sources of cost overrun on
construction projects and to develop final cost estimation models that could help
in reaching more reliable final cost estimates at the tendering stage of the project.
The research identified two predominant schools of thought on the sources of
overruns – referred to here as the PsychoStrategists and Evolution Theorists.
Another finding was that there is no unanimity on the reference point from which
cost performance could be assessed, leading to a large disparity in the size of
overruns reported. Another misunderstanding relates to the term “cost overrun”
itself.
The experimental part of the research, conducted in collaboration with two
industry partners, used a combination of non-parametric bootstrapping and
ensemble modelling with artificial neural networks to develop final project cost
models based on about 1,600 water infrastructure projects. 92% of the validation
predictions were within ±10% of the actual final cost of the project. The models
will be particularly useful at the pre-contract stage as they will provide a
benchmark for evaluating submitted tenders and also allow the quick generation
of various alternative solutions for a construction project using what-if scenarios.
The original contribution of the study is a fresh thinking of construction “cost
overruns”, now proposed to be more appropriately known as “cost growth” based
on a synthesises of the two schools of thought into a conceptual model. The
second contribution is the development of novel models of construction cost
estimation utilising artificial neural networks coupled with bootstrapping and
ensemble modelling
- …
