Pavement surface texture is an essential factor in influencing the sound emission of vehicles, and reducing noise pollution would be a big issue for all humans. According to the Economic Commission for Europe (ECE) R-117 requirement, all the vehicle tires vending in Europe should pass noise accreditation after September 2009, and the pavement surface of the test track should follow ISO 10844. The Mean Profile Depth (MPD) is adopted to be the texture parameter of the test track surface in ISO 10844. the MPD of the Test track should be controlled at specific intervals, and the material design should be a big challenge of pavement Engineering. This study aims to establish an MPD prediction function with material properties for pavement Engineers. In material, there were two fine aggregates, three gradation curves, three air voids of the specimen, and 18 mixture designs adopted in this study. There were three specimens in each mixture design; the total specimen number was 54 for MPD measurement. Fifty-four laboratory test specimens were compacted using a Superpave gyratory compactor (SGC). A High-Definition Pavement Texture Machine (HDPTM) measured the specimen surface MPD. The parameters in this study included fineness module (FM), fine aggregate angularity (FAA), air void (AV) of the specimen, and Gyration number (G) of SGC. Gene Expression Programming (GEP) was adopted to build a prediction model between MPD and various parameters. Based on the results of this study, the best MPD forecasting model using FAA, FM, and AV, the R2 of is 0.74, meaning the best MPD forecasting model could be a very nice tool to help pavement engineers to do MPD quality control on the test track for ISO 10844. In addition, GEP was an excellent method to discover the nonlinear relationship between MPD and various parameters of asphalt concrete, and it was worthy of further study.
Read full abstract