Calibration of Roller-Integrated Machine Drive Power Measurements Using Soil Compaction Forecasting Model

Calibration of roller-integrated machine drive power (MDP) measurements is a necessary step for implementing the continuous compaction control technology into practice. State-of-the-practice calibration procedures use univariate or multivariate regression analysis of roller machine values with traditional compaction indices, such as modulus or soil unit weight. In this paper, a new calibration technique for estimating dry unit weight from MDP measurements was developed through investigation of the relationship between roller-integrated MDP and the inferred compactive effort associated with the corresponding moisture-density state of soil. Based on field data from eight test strips, linear trends were observed between compactive effort and cumulative MDP, and the proposed approach was able to detect how moisture content (a changing variable between test strips) substantially influences soil compactive effort and the soil–roller interaction affecting MDP measurements. Identifying slopes and intercepts of these moisture-dependent correlations allows for estimating roller-achieved compactive effort, and then converting estimated compactive effort to soil dry unit weight through a unified soil compaction model that is calibrated using traditional laboratory Proctor test data. The proposed forecasting calibration technique is based on accepted and defensible geotechnical principles on which earthworks practice is based, not relying solely on regression models. Observing the correlation of roller machine values and dry unit weight in the context of soil compaction behavior is expected to improve confidence in the continuous compaction control technology and ultimately result in more widespread acceptance and use.