Elsevier

Construction and Building Materials

Volume 164, 10 March 2018, Pages 888-898
Construction and Building Materials

Virtual Special Issue Ground-Penetrating Radar and Complementary Non-Destructive Testing Techniques in Civil Engineering
Comparison of different radar technologies and frequencies for road pavement evaluation

https://doi.org/10.1016/j.conbuildmat.2018.01.124Get rights and content

Highlights

  • Study explores on real life situation on road trying to quantify the surface porosity of asphalt.

  • Results shows importance of representative volume element.

  • Results shows the sensitivity of measurements on aggregate permittivity variation.

  • As an outcome, aggregate permittivity variation masks the porosity variation.

Abstract

Experimental asphalt pavement field measurements have been performed with a commercial Ground Penetrating Radar (GPR) system and with three dedicated, new prototype radars operating at 1–2 GHz, 12–18 GHz and 32 GHz frequency. The aim of these measurements is to find the surface or near-surface permittivity of worn asphalt and to investigate the suitability of various radar technologies for this task. Our experiments were supplemented by drill core samples extracted at selected interesting locations on a test road. As expected, there is hardly any correlation between the results of the four different radars along the same test track, and drill core results complicate the situation further. Taking 10 and 90 percentile cumulative probabilities, the commercial GPR data has the smallest εr′ span from 5 to 5.6, the 1–2 GHz prototype system indicates εr′ between 4.5 and 5.5, our 12–18 GHz system 4.5 to 6.5 and the 32 GHz fixed frequency reflectometer 2.5–4.5. Only the 32 GHz measurements show clearly different mean permittivity values for the visually different pavement surfaces. Test results, however, suggest that extracting air void variation from highly inhomogenous asphalt pavement needs stochastic approach instead of trying to do some deterministic calibrations with pavement cores.

Introduction

Non-destructive and non-contacting pavement quality analysis has been carried out in production use with Ground Penetrating Radar (GPR) systems for about 20 years. Some of the first practical implementations are documented in Maser and Scullion [1] Chen et al. [2]. Typical measurements have included studies of compaction ratio [3] and asphalt layer thickness. More recently, new methods for in situ asphalt mixture density prediction have been evaluated [4]. Most of these parameters are post-processed from radar pulse reflection amplitudes observed at permittivity discontinuities [5] and propagation time delays, often computed from raw data with sophisticated software packages. Measurements with contacting dielectric probes have also been tested at low VHF frequencies [6], but those methods are seldom suitable for long road surveys.

In a number of countries, including Finland, GPR road measurements have become routine. More recently, for example in Fauchard et al. [8], research has indicated concerns regarding the obtainable uncertainty of relative permittivity and depth resolution with GPR devices when employing them in applications for thin pavement surface layers. The GPR has been used, for instance in Finland, to gather quantitative data for quality control measurements to assess pavement compaction and density. Also, radar calibration procedures have aroused discussion, see for example Scheer [9] and Knott et al. [10].

Another important issue is the effect of pavement rock characteristics [11] on the permittivity of asphalt. The variation of the permittivity of rock aggregate in asphalt may mask the variation of porosity. This may prevent meaningful quality measurements. Asphalt is a granular composite material composed of aggregate, bitumen and air. Therefore, the permittivity of asphalt is the effective bulk permittivity of all these constituents. As such, the volume of air or porosity is hard to detect or measure, as it only constitutes less than ten percent of the total volume. Consequently, spatial resolution is crucial for obtaining meaningful results. Authors have studied this issue extensively [20], [21], [23], [24], [25], [26], [27], [28] but a reasonably accurate model to predict the relative effective bulk permittivity of asphalt volume is yet to be developed. The need to develop such a model is thus the primary motivation for these measurements and the experiment. Even though we have used frequencies as high as 32 GHz and have selected the pavement test sites based on visual observations of surface conditions, the intent is not to gather information of the surface characteristics per se, but from the bulk volume of asphalt.

Interesting studies, mainly at the very high frequency of 94 GHz, have been made in Sarabandi et al. [12], where the authors have developed a theoretical model and performed reflectivity measurements of real asphalt, but mostly with grazing angles, which are applicable to automotive anti-collision radars. Results are thus not directly suitable for the problem in our experiments. However, the general problem of road layer structure and granularity of medium is identical.

The purpose of the measurements presented here was to evaluate the suitability and general performance of different radar technologies and frequency bands for road pavement quality assessment analysis, possibly even as a multi-radar configuration. The frequency band of 1–32 GHz has been chosen in order to investigate the effect of the frequency on the spatial resolution of the radar’s performance. Studies on this issue [19], [22] clearly indicate the importance of using higher frequencies to confine the radar pulse to the upper asphalt layers to obtain the density of the 40–50 mm thick surface layer. However, when using high frequencies, the radar is sensitive to the granularity of the material, which distorts measurement results in this particular case. Therefore, a compromise may be needed in choosing the best frequency range for the measurements.

Section snippets

Test road

The total length of our test road is 275 m divided into four consecutive visually different pavement sections, later labeled from A to D. Visual differences suggest changes in material composition, which in turn translate to differences in mechanical and electrical properties. The pavement of section A is more than 10 years old whereas section B was re-paved in 2014. Sections C and D are part of an old 1970s highway, but their pavement ages are unknown. No accurate mix design data of the

Radar equipment

Four different radar systems, having varying frequency ranges and operating principles, were used. Taking into account the known physical and electrical properties of asphalt pavement, it was clear from the very beginning that high microwave and millimeter wave frequencies would certainly behave differently to, say, typical L-band, but in this experiment we wanted to get initial real-life measurement data at this higher band and at the same time test if it is possible to create background data

Results with radars

Firstly, it is advantageous to have an overall view of the obtained permittivity (ε′ only) along the entire test track. This is shown in Fig. 2, where we also indicate the drill core locations and the road section boundaries.

The commercial GPR data has only one notable permittivity step, which occurs at the boundary between sections C and D and here its result exceeds all other measured values during this campaign. Otherwise, the GPR-measured permittivity has no clear response to the pavement’s

Drill core samples

Before taking pavement cores, the road surface was photographed at the six selected test points. These pictures are collected in Fig. 5.

Drilled cores (diameter D = 100 mm) were first measured in their full length (whole core as one piece) and then all pavement layers were separated by sawing and measured individually. Only results for the whole core and for the surface layer are given in Table 4, while Fig. 7 shows air void contents for each individual layer. Pavement section A had only one

Discussion

Fig. 7 compares asphalt layers at each core location and shows the measured density of the different pavement layers visually. The required density for dense graded asphalt concrete (AC) pavements in Finland is between 1% and 5% of air voids from the total volume. Based on laboratory measurements, asphalt layers were very dense having less than 2.5% of air. Only drill core 4 (base layer) and core 5 (surface layer) had air void contents above 4%. Vertical dotted lines in Fig. 7 illustrate the

Concluding remarks

This experiment confirms that there are significant differences in measuring road pavement surface characteristics with different radar technologies and frequency bands. The low frequency band radars seem to be too insensitive to detect the variation of the volume of air in dense graded asphalt pavement. However, it is very difficult to handle the granularity-related scattering at 12–18 GHz. The lowest millimeter wave band close to 32 GHz looks attractive, but additionally we have to consider

What is already known in this area?

Asphalt mixture can be modelled with EM methods by knowing aggregate and bitumen permittivity. The frequency affects spatial resolution of measured permittivity. High frequencies cause scattering.

What does this study add to the literature?

Explores the frequency effect on real life situation on road trying to quantify the surface porosity of asphalt. Shows the sensitivity of measurements on RVE and aggregate permittivity variation.

Acknowledgements

Authors wish to thank the Finnish Road Administration for providing funding for this research. This work benefited from networking activities carried out within the EU funded COST Action TU1208 “Civil Engineering Applications of Ground Penetrating Radar”.

References (28)

  • G. Olhoeft

    Electrical properties of rocks

  • K. Sarabandi et al.

    Modeling and measurements of scattering from road surfaces at millimeter-wave frequencies

    IEEE Trans. Antennas Propag.

    (1997)
  • A. Martinez, A. Byrnes, Modeling dielectric-constant values of geologic materials: an aid to ground-penetrating radar...
  • M. Hallikainen et al.

    Microwave dielectric behavior of wet soil – part I: empirical and experimental observations

    IEEE Trans. Geosci. Remote Sens.

    (1985)
  • View full text