National Academies Press: OpenBook
« Previous: Front Matter
Page 1
Suggested Citation:"Summary." National Academies of Sciences, Engineering, and Medicine. 2020. Evaluating Mechanical Properties of Earth Material During Intelligent Compaction. Washington, DC: The National Academies Press. doi: 10.17226/25777.
×
Page 1
Page 2
Suggested Citation:"Summary." National Academies of Sciences, Engineering, and Medicine. 2020. Evaluating Mechanical Properties of Earth Material During Intelligent Compaction. Washington, DC: The National Academies Press. doi: 10.17226/25777.
×
Page 2
Page 3
Suggested Citation:"Summary." National Academies of Sciences, Engineering, and Medicine. 2020. Evaluating Mechanical Properties of Earth Material During Intelligent Compaction. Washington, DC: The National Academies Press. doi: 10.17226/25777.
×
Page 3
Page 4
Suggested Citation:"Summary." National Academies of Sciences, Engineering, and Medicine. 2020. Evaluating Mechanical Properties of Earth Material During Intelligent Compaction. Washington, DC: The National Academies Press. doi: 10.17226/25777.
×
Page 4
Page 5
Suggested Citation:"Summary." National Academies of Sciences, Engineering, and Medicine. 2020. Evaluating Mechanical Properties of Earth Material During Intelligent Compaction. Washington, DC: The National Academies Press. doi: 10.17226/25777.
×
Page 5
Page 6
Suggested Citation:"Summary." National Academies of Sciences, Engineering, and Medicine. 2020. Evaluating Mechanical Properties of Earth Material During Intelligent Compaction. Washington, DC: The National Academies Press. doi: 10.17226/25777.
×
Page 6
Page 7
Suggested Citation:"Summary." National Academies of Sciences, Engineering, and Medicine. 2020. Evaluating Mechanical Properties of Earth Material During Intelligent Compaction. Washington, DC: The National Academies Press. doi: 10.17226/25777.
×
Page 7

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Evaluating Mechanical Properties of Earth Material During Intelligent Compaction Satisfactory pavement performance can only be assured with an appropriate process control to ensure that compacted materials meet proper density and stiffness require- ments. The primary tool currently used for quality management of earthwork and unbound aggregates is the nuclear density gauge (NDG) to ensure appropriate density and moisture content. Measurement of moisture content and dry density, even though quite practical and straightforward, does not directly tie the construction quality with the mechanistic- empirical design processes where stress and modulus are employed. With the recent popularity of the mechanistic pavement design procedures, research efforts have been undertaken to understand and develop procedures for implementing modulus-based quality control procedures of compacted geomaterials. These procedures involve the use of in-situ nondestructive testing (NDT) devices that estimate the stiffness parameters of a constructed pavement structure. A shortcoming of NDT spot testing is that weak areas may be missed. If implemented properly, intelligent compaction (IC) can provide quality control over 100% of compacted materials. Furthermore, the uniformity of compacted earthwork can be realistically assessed with accelerometer-based IC measurement values (ICMV). Another possible benefit of using IC is the instant identification of weak areas that need to be reworked. IC technology consists of a vibratory roller equipped with accelerometers mounted on the drum’s axle, a global positioning system (GPS), and an onboard computer reporting system that displays IC measurements in real time. Despite the tremendous efforts that have been made to investigate the application of IC technology in construction quality control, knowledge gaps still prevent the use of IC technology for construction acceptance of geomaterials. These knowledge gaps include (1) the need to relate the design parameters to the construction quality control parameters and the in situ moisture content, and (2) the absence of rational means of relating different proprietary ICMVs reported by different roller vendors. To address these issues, field-calibrated numerical models are needed that can be used for the proper evaluation and acceptance of the compacted geomaterials. A realistic numerical model for a roller-soil system can be combined with a state-of- the-art inverse or backcalculation algorithm to provide reliable layer-specific ICMV for construction quality control and potential acceptance. This process must be robust and practice-ready, however, so that departments of transportation (DOTs) can readily incorporate it in their IC specifications. This report details the findings of NCHRP Project 24-45, “Evaluating Mechanical Properties of Earth Material During Intelligent Compaction,” which was undertaken to investigate methods to evaluate mechanical properties of geomaterials using IC technology and to develop generic specifications for the application of IC in quality management of soil and aggregates base materials. S U M M A R Y

2 Evaluating Mechanical Properties of Earth Material During Intelligent Compaction Objective The main objective of this research was to develop procedures to estimate the mechanical properties of the geomaterials using IC technology in a robust manner so that DOTs can incorporate it in their specifications. Summary of Activities The research process emphasized practicality and sought to establish field processes and analysis algorithms suitable for considering the attributes of a diverse range of geo- materials. The research team started by documenting, synthesizing, prioritizing, and conducting gap analyses on the following topics: • National and international state of the practice and implementation of quality control/ quality assurance (QC/QA) with IC technology; • Different approaches to incorporate the unsaturated soil mechanics concepts in the process; • Numerical models that simulate the IC roller compaction process during proof-mapping; • Available backcalculation algorithms, including those that make use of artificial neural networks (ANNs) and genetic algorithms (GAs) to extract the mechanical properties (stiffness and/or modulus) of compacted materials; and • Means of rapidly and robustly measuring and reporting the mechanical properties of layers. Based on this information, the research team drafted a systematic process for imple- menting roller IC technology (illustrated in Figure S-1) that would allow the team to focus on: • The most relevant parameters that should be considered and researched, • Their practical and desirable tolerances, • Means of rapidly measuring output parameters, and • Means of analyzing the field results in a rapid and robust manner that balances the risks of highway agencies and the contractors. Numerical models with different levels of complexity were developed to simulate realistically and efficiently the geomaterials’ response under vibratory roller compaction with a focus on modeling the mapping operations for quality management. The proposed numerical models were evaluated in terms of execution time and results accuracy with respect to the field-measured responses. Those models were used to develop and evaluate different backcalculation techniques to extract the mechanical properties from IC roller measurements during the mapping process. To implement this step, the research team developed an extensive database comprising a wide range of layer properties and thicknesses. A sensitivity analysis and a parametric study were conducted to select the structural and stiffness parameters that significantly influence the pavement responses. These parameters were used to develop and evaluate inverse algorithms that would be robust enough to extract the layer mechanical properties from the IC measurements, in conjunction with additional spot testing using NDT devices. The recommended specifications for the extraction of the mechanical properties were developed and pilot tested through field testing of seven active earthwork construction projects in Minnesota, Ohio, and Texas on granular soils, fine-grained soils, cement- stabilized soils, and unbound aggregate base materials commonly used in subgrade, subbase, and base course construction. Data from rollers manufactured by different manufacturers

Summary 3 were used as part of the development of the forward and inverse models and the valida- tion of the proposed approach for extracting the mechanical properties of compacted materials. The sites were instrumented with embedded geophones to investigate the relationship between roller measurements and the developed numerical models for cali- bration purposes. Relationships between ICMVs and measurements from lightweight deflectometer spot tests were investigated and used for calibrating the numerical models and the inverse solver for extracting layer modulus. The following sections summarize the findings obtained as part of the different tasks completed in this research study. Numerical Modeling Different studies evaluating the experimental data collected with instrumented rollers revealed complex nonlinear roller vibration behaviors, such as loss of contact between the drum and the soil and nonlinear behavior of the materials subjected to compaction. For representative prediction of the responses, it was found necessary to use a three-dimensional (3D) nonlinear finite element (FE) model to simulate the proof-mapping process of single-layer and two-layer geosystems, with an automatic surface-to-surface contact model to account for the soil-drum interaction. A comprehensive database of cases with different input parameters was assembled for single-layer and two-layer geosystems and various drum dimensions with different operating conditions. Different levels of complexity were introduced into the model to numerically assess the impact of the vibratory conditions and to consider both linear and Select Material Type for All Layers Simulate Roller Measurements • Determine Target Field Measurements Value (ICMVTarget) for each Layer • Determine Target NDT Value (Optional) Estimate Properties of All Layers • Physical: Thickness • Index: Gradation and Atterberg Limits • Mechanical: Resilient Modulus Parameters, Strength • Moisture-Density: In Situ, OMC Select Roller Parameters • Model: Drum Dimensions and Weight • Vibration Parameters: Amplitude, Frequency, Speed Pre-Map Layer of Interest • Extract Statistical Information about ICMV • Conduct Spot Test with Modulus-based NDT Devices (to Extract Layer Modulus) Map Compacted Layer • Extract Statistical Information about ICMV and ∆ICMV • Provide Average Stiffness and Uncertainty Related to Estimated Properties Post-Processing to Extract Layer Mechanical Properties • Conduct Spot Test with Modulus-based NDT Devices • Conduct Tests to Characterize Moisture Variation • Use a Robust Inverse Algorithm to Extract Modulus Perform Compaction to Achieve Target • Review ICMV Color-Coded Map to Ensure DOT Acceptance Figure S-1. Implementing roller IC technology (generic flowchart).

4 Evaluating Mechanical Properties of Earth Material During Intelligent Compaction nonlinear geomaterial constitutive models on the pavement responses. Correlations were established among the responses of the different models and the obtained relationships were evaluated to simplify the modeling. Mapping of ICMV IC measuring systems collect vibration data at discrete points at the edge of the roller drum during IC proof mapping. Commercially available IC systems incorporate various processes to extrapolate the measured ICMV data points over the width of the roller prior to generating the color-coded maps. In this study, the research team developed a system to evaluate the vibration characteristics of the IC rollers and the response of ground layers during the IC operations. The roller vibration data was collected with up to two accelerom- eters mounted at the drum of a roller while a GPS unit mounted on the roller monitored the roller’s location. Another component consisted of 3D geophones embedded at different depths into the earthwork. The geophones were connected to a data acquisition system to measure the responses of the ground layers during mapping operations. The impacts of the data collection rate and reduction processes on the collected IC measurements were evaluated to set appropriate levels of spatial density and resolution. To make the process more practical, the collected IC data was partitioned into virtual sublots equal to the width of the roller and the length equal the minimum length of the compacted section that was practical to rework. For mapping ICMV, all ICMV measure- ments falling inside a sublot were averaged to obtain representative ICMVs. This approach can accommodate the inherent uncertainties related to the accuracy of the GPS devices and the precise position of the moving roller in a straightforward and transparent manner. To ensure uniformity throughout the site, a color-coded map representing the coeffi- cient of variation (COV) of the ICMVs within each sublot should accompany the map- ping of ICMVs. Such COV color-coded maps allow the identification of sublots where the repre sentative ICMVs are no longer reliable due to construction- or equipment-related issues. The traditional approach to color coding relies on the standard deviation of the site; however, this approach loses its effectiveness as the site becomes more variable (yields a higher standard deviation). The research team used an enhanced approach, establishing a color-coded criterion under which any sublot with representative ICMVs greater than the average ICMV of the lot was considered relatively stiffer and assigned a color (green). Sublots with average ICMV of less than 75% of the average ICMV of the lot were considered as less stiff (not inferior quality) and were represented in a second color (red). Sublots with ICMVs of 75% of the average ICMV of the lot or higher, and lower than the average ICMV of the lot, were considered moderately stiff and were represented in a third color (yellow). For any given site, the resulting color-coded map might not contain any red areas as long as the work is uniform enough to yield a COV of less than 25%. Again, sublots with ICMVs less than 75% of the average ICMV of the lot do not necessarily imply that they do not meet stiffness requirements; this threshold implies only that their stiffness is less than that of other sublots. Using the information acquired with the data acquisition system, additional maps were generated for real-time quality control of collected IC data. The mapping of operating frequency, roller speed, line passes, number of discrete measurements per sublot, and amplitude (i.e., surface displacement) served for conducting quality control of the ICMVs acquired during mapping. This information also was to be used as input into the back- calculation algorithms for the backcalculation of layer moduli.

Summary 5 Field Evaluation Four test cells at the MnROAD test track facility were instrumented. Test sections were built with full-scale construction equipment to simulate normal highway construction. Each cell consisted of distinct combinations of subgrade and unbound aggregate base materials. All sections were proof mapped using an IC roller equipped with a data acquisi- tion system consisting of accelerometers mounted at the edge of the drum and a GPS system to measure the location of the roller. This procedure allowed the comparison of the collected IC data with the roller’s IC measurements obtained from the Controller Area Network (CAN) bus. Both measurements were found to be in agreement. A second data acquisition system collected soil displacements from the embedded geophones during the roller’s mapping operations. This information was used for further comparison and calibration purposes of the numerical model responses. During IC mapping, data acquisition for each cell was conducted on top of the subgrade (single-layer system) and on top of the constructed base layer (two-layer system). All test sections were evaluated using NDT devices to measure modulus-based properties of the compacted materials. NDT testing consisted of the use of a light weight deflectometer (LWD), falling weight deflectometer (FWD), and dynamic cone penetrometer (DCP) to estimate the modulus/stiffness of compacted base and subgrade sections implemented in equally spaced spots along the length and width of the test sections. Similarly, a nuclear density gauge (NDG) was employed to evaluate the moisture-density properties of compacted geomaterials. Soil samples were transported to the laboratory to measure their in-place moisture content, index properties, and to perform resilient modulus tests. The implemen- tation of dense spot tests along the test section allowed the research team to address the field variability of the material properties and how this affects the measurements collected by the IC roller. Note: In the literature, the term resilient modulus has various abbreviations (e.g., MR, MR, and mr). All three abbreviations may appear in figures or equations in NCHRP Research Report 933, reflecting their presentation in the original sources. To minimize confusion, the text typically spells out the term rather than using the abbreviation. Calibration of Numerical Models The developed numerical models were calibrated using field measurements acquired from embedded geophones during IC mapping of test sections at the Minnesota pavement test track (MnROAD) and a construction site in Texas. The pavement properties of the mapped pavement sections were used as input into the FE models used in this project, and the drum was simulated using the dimensions and operating characteristics of the IC roller used in the test sections. The Texas dataset consisted of a section mapped using IC rollers from various manufac- turers vibrating in stationary and moving conditions. The collective field and numerical datasets were used to develop adjustment factors after the comparison of field to numeri- cal pavement responses. Better correlations between the field measurements and their cor responding numerical model responses were observed when local adjustment relation- ships were obtained as opposed to a single global relationship obtained from field measure- ments and numerical model responses using only laboratory-obtained properties as inputs. The local calibration approach integrated the LWD test measurements with the resilient modulus test results. This approach incorporated the state of compaction of the layer and to some extent the variation in moisture content in the analysis.

6 Evaluating Mechanical Properties of Earth Material During Intelligent Compaction Extraction of Mechanical Properties A robust backcalculation technique to extract the mechanical properties without excessive processing time during the mapping of the compacted layers was developed. The stiffness (equivalent to a modulus of subgrade reaction) can be extracted directly in the frequency domain by obtaining the ratio of the complex amplitudes of the force imposed by the drum and the roller deflection at that location. This approach was integrated into the almost real-time analysis module that processed the measured datasets in the field as soon as the proof-mapping process was finished. In the case of two-layer geosystems, the roller measurements would provide a composite stiffness because only one piece of information was available at each sublot. To obtain an optimal predictive function for estimating the moduli of the subgrade and base, two approaches were evaluated for real-time backcalculation: a genetic programming (GP) method that makes use of GAs, and an ANN-based approach. For this purpose, a comprehensive dataset of single-layer and two-layer geosystems with different layer properties and base thicknesses was assembled using a stationary static nonlinear model. After a sensitivity analysis, the parameters that had a more significant impact on the pavement responses were identified and selected as inputs into the proposed GP- and ANN-based inverse solvers. Different inverse solvers, with differing numbers of input variables, were proposed for the various geosystem scenarios. The expectation was that the precision of the predictions would improve with the more complex solvers; however, the more complex inverse solvers would require more laboratory efforts to determine the needed input variables. The inverse solvers were evaluated to select backcalculation scenarios that were best suited for predicting layer moduli for both single-layer and two- layer systems. The predictive power of the inverse solvers improved when local adjust- ment factors were used. The predictions made using both the GP method and an ANN approach were compared to the results obtained from the FE models, and a decision was made to continue the research by refining and further testing the ANN-based model. Findings from Validation Process Visits were conducted to four additional test construction sites in Minnesota, Ohio, and Texas to evaluate and validate the practicality of the developed forward models and back- calculation algorithms under field conditions. These test sections were also instrumented using embedded geophones at different depths to record soil displacements during the mapping process for further improvement of the inverse solver algorithms. The impact of variability in the measurements toward the extraction of the mechanical properties of the compacted layers was also assessed. The results were used to further evaluate and improve the framework of the specification. In general, the conclusions drawn and lessons learned during the validation phase were reasonably similar to those obtained from the test sections that were evaluated during development. As part of the dissemination of the proposed specification, the following items should be strongly emphasized to the highway agencies: • The adoption of a specification to extract the mechanical properties of compacted materials using IC needs to be approached in the context of the levels of uncertainty associated to the uniformity of the compaction. • The most consistent results are obtained when proof mapping is carried out in conjunc- tion with the modulus-based measurements and when variability in the ICMVs is kept at less than 25%.

Summary 7 • Due to large diversity in construction practices and material types, the implementation of the draft specification requires more localized field studies by DOTs to adopt it to their local materials and construction practices. Based on this study and interaction with the highway agencies, the following comments and suggestions can be made: • This research study provides a critical review of the strengths and concerns about the implementation of a specification to extract the mechanical properties of compacted materials using IC technology. The research team attempted to highlight the complexi- ties that could arise and made an effort to address them in a comprehensive manner. • Even though this report emphasizes both the strengths and concerns with the proposed specification, the proposed specification is a big step toward higher quality highway construction. Recommended Specifications Following AASHTO PP 81-14 as a baseline, two proposed specifications were developed; one for stiffness-based acceptance, entitled “Proposed Standard Specification for Quality Management of Earthwork and Unbound Aggregates using Intelligent Compaction (IC),” and one for extraction of the modulus of compacted layers, entitled “Proposed Standard Specification for Extracting Modulus of Compacted Geomaterials Using Intelligent Compaction (IC).” These specifications are presented in Appendix A of this report. Two test methods that are proposed to accompany the specifications also are provided in Appendix A. The use of the stiffness-based specification serves as an almost real-time approach for determining mechanistic-based field target values for routine quality management purposes. The modulus-based specification, on the other hand, would be preferred if the goal of the highway agencies is to extract the moduli of the layers. For the implementation of the modulus-based specification, however, the highway agency should be prepared to conduct some laboratory testing up front and institute more rigorous process control during the compaction process. The specifications make use of an approach to parti- tion the lot into virtual sublots for mapping ICMVs. The sublot dimensions equal the width of the roller and the minimum length of the compacted section that is practical to rework, which is set at the discretion of the engineer. ICMVs falling inside a sublot are averaged to obtain representative ICMVs. This approach is proposed to accommo- date the inherent uncertainties related to the accuracy of the GPS devices and the precise position of the moving roller. It also facilitates identification of the less stiff areas. The modulus-based specification requires using IC to identify the sublots with more uniform ICMVs for conducting additional spot tests.

Next: Chapter 1 - Introduction »
Evaluating Mechanical Properties of Earth Material During Intelligent Compaction Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Satisfactory pavement performance can only be assured with appropriate process controls to ensure compacted materials meet proper density and stiffness requirements.

The TRB National Cooperative Highway Research Program's NCHRP Research Report 933: Evaluating Mechanical Properties of Earth Material During Intelligent Compaction details the development of procedures to estimate the mechanical properties of geomaterials using intelligent compaction (IC) technology in a robust manner so that departments of transportation can incorporate it in their specifications.

Appendix A, containing the proposed specifications and test methods, is included in the report. Appendices B through H appear in a supplementary file.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!