| In this study, prototype machine integrated compaction monitoring technologies were evaluated for self-propelled non-vibratory soil compactors. Monitoring technologies included roller walkout, rolling radius, machine energy, wheel slip, and machine vibration. In concept, the machine measurement value should be capable of predicting compaction properties in real time for 100% of the machine coverage area while displaying the results on a computer monitor inside the cab with global positioning system (GPS) based mapping. To move toward a production-level compaction monitoring technology, a detailed evaluation of these technologies including field experimentation and statistical analysis of the data was required. The statistical analysis of measurements included repeatability analysis, simple linear correlations analysis, and multiple variable model relationship analysis. The relationships between machine measurement values and soil compaction properties was verified using various in-situ point measurements to determine density, moisture content, strength, and stiffness properties of the soil. Several test strips with variable moisture contents, lift thicknesses, and sloping conditions were constructed as part of the evaluation. Relatively strong relationships were observed between machine measurement values and in-situ point measurements (coefficients of determination, R 2, generally above 0.7), especially when measurements of roller walkout and machine energy were utilized. Regression relationships were improved by adding moisture content as an independent variable, in some cases producing R2 values that exceed 0.80. A repeatability study on machine measurement values produced errors in the range of 20 to 40 percent. Findings from this study help to narrow down several roller integrated compaction monitoring technologies based on repeatability and correlation analysis to soil compaction properties. |
