ABSTRACT

This paper presents a new approach for calibration of Mechanistic-Empirical Pavement Design Guide (MEPDG) permanent deformation models using the Hamburg Wheel Rut Tester (HWRT) Data. The approach was applied on two Ontario mixes tested in HWRT to calibrate and validate the asphalt concrete (AC) rut models. Data from one Long-Term Pavement Performance (LTPP) programme test section in Ontario were also used for verification, calibration and validation of the AC and unbound pavement layer rut models. The section had two construction (CN) life spans, one before milling and overlay (CN1) and one after (CN2). The study used two finite element models and an Excel VBA sheet to run the verification, calibration and validation of the MEPDG rut models. Proposed calibration coefficients resulted in significant reduction in bias, sum of squared errors (SSE) and standard error of estimate (SEE) for rutting by at least 91% for the two HWRT mixes and by at least 67% for CN1 and CN2, when compared with the global and Ontario local calibration coefficients. The results support the conclusion that the approach of using laboratory wheel tracking device data for calibrating the MEPDG rut models is promising.

摘要

本文介绍了一种使用汉堡车轮车辙测试仪 (HWRT) 数据校准机械经验路面设计指南 (MEPDG) 永久变形模型的新方法。该方法应用于在 HWRT 中测试的两种安大略混合料，以校准和验证沥青混凝土 (AC) 车辙模型。来自安大略省一个长期路面性能 (LTPP) 计划测试部分的数据也用于 AC 和无约束路面车辙模型的验证、校准和验证。该部分有两个施工 (CN) 寿命，一个在铣削和覆盖之前 (CN1)，一个在铣削和覆盖之后 (CN2)。该研究使用两个有限元模型和一个 Excel VBA 表来运行 MEPDG 车辙模型的验证、校准和验证。拟议的校准系数导致偏差显着减少，与全球和安大略省当地校准系数相比，两种 HWRT 混合物车辙的平方误差和 (SSE) 和估计标准误差 (SEE) 至少降低 91%，CN1 和 CN2 至少降低 67%。结果支持以下结论：使用实验室车轮跟踪设备数据校准 MEPDG 车辙模型的方法是有前途的。