Abstract
Against the backdrop of national new infrastructure development and intelligent construction, intelligent roadbed compaction technology is burgeoning; however, the low correlation and reliability of traditional compaction degree characterization indicators hinder the promotion and application of intelligent compaction technology. To address these core challenges, this research focuses on the dynamic response mechanisms and compaction characteristics of roadbeds, systematically investigating the dynamic behavior of roadbed fill materials, multi-scale structural evolution mechanisms, and intelligent control methods. First, based on dynamic triaxial tests, the macroscopic dynamic response characteristics of roadbed fill materials under different working conditions and stress paths are studied, and a dynamic constitutive model suitable for intelligent compaction analysis is established. Second, utilizing discrete element simulation and CT three-dimensional reconstruction technology, the microscopic compaction mechanisms such as contact network evolution between soil particles, particle rearrangement, pore closure, and changes in local stress transmission paths under vibratory loads are revealed, the relationship between vibratory roller acceleration characteristics and the dynamic modulus of roadbed materials is established, and a dynamic modulus characterization model for back-calculating compaction state based on measured multi-dimensional acceleration signals is proposed. Subsequently, through full-scale model tests, the variation characteristics of dynamic modulus of roadbed materials under different compaction energy levels and porosity conditions are systematically investigated, a coupling mechanism model of "porosity—dynamic modulus—vibration response" is established, and a novel compaction quality evaluation system based on dynamic parameters is proposed. Furthermore, an artificial neural network model is constructed based on deep learning algorithms, integrating construction rolling paths, vibration parameters, soil properties, and sensor signals to achieve high-precision prediction of compaction quality and adaptive optimization and control of construction parameters, thereby developing an integrated intelligent roadbed compaction control system encompassing data acquisition, intelligent identification, and feedback control. The research findings will significantly enhance the scientificity, real-time capability, and intelligentization level of roadbed compaction quality control, providing key theoretical support and a technical pathway for the deep integration of intelligent construction technology in the field of transportation civil engineering.
Full Text
Preamble
Research and Development of Intelligent Compaction Technology for Road Subgrades
Jinhao Zang¹,², Jin Li¹,², Shen Zuo¹,²
¹Shandong Key Laboratory of Technologies and Systems for Intelligent Construction Equipment, Shandong Jiaotong University, Jinan, Shandong 250357, China
²School of Transportation and Civil Engineering, Shandong Jiaotong University, Jinan, Shandong 250357, China
Abstract
In the context of national new infrastructure development and intelligent construction, intelligent compaction technology for road subgrades is burgeoning. However, the low correlation and reliability of traditional compaction degree indicators have hindered the widespread adoption and application of intelligent compaction technology. To address this core challenge, this research focuses on subgrade dynamic response mechanisms and compaction characteristics, systematically investigating the dynamic behavior of subgrade fill materials, multi-scale structural evolution mechanisms, and intelligent control methods.
First, dynamic triaxial tests were conducted to study the macroscopic dynamic response characteristics of subgrade fill materials under various working conditions and stress paths, establishing a dynamic constitutive model suitable for intelligent compaction analysis. Second, discrete element simulation and CT three-dimensional reconstruction technology were employed to reveal microscopic compaction mechanisms under vibratory loads, including the evolution of inter-particle contact networks, particle rearrangement, pore closure, and changes in local stress transfer pathways. The relationship between vibratory roller acceleration characteristics and subgrade material dynamic modulus was established, and a dynamic modulus characterization model was proposed to infer compaction state from measured multi-dimensional acceleration signals.
Subsequently, full-scale model tests were performed to systematically investigate the variation characteristics of subgrade material dynamic modulus under different compaction energy levels and porosity conditions. A coupled mechanism model linking "porosity—dynamic modulus—vibration response" was established, and a novel compaction quality evaluation system based on dynamic parameters was proposed. Furthermore, an artificial neural network model was constructed using deep learning algorithms, integrating construction rolling paths, vibration parameters, soil properties, and sensor signals to achieve high-precision prediction of compaction quality and adaptive optimization of construction parameters. An integrated intelligent compaction control system for subgrades was developed, encompassing data acquisition, intelligent identification, and feedback control.
The research findings will significantly enhance the scientific rigor, real-time capability, and intelligence level of subgrade compaction quality control, providing critical theoretical support and a technical pathway for the deep integration of intelligent construction technologies in transportation civil engineering.
Keywords: Road subgrade; intelligent compaction; dynamic response; porosity; discrete element method; dynamic modulus