Performance study of vertical vibratory roller compaction of large thickness water stabilized layer

In road construction, the compaction of the large-thickness water-stable layer is mainly done in layers. However, the construction period is long, and the integrity of the water-stable layer after compaction is not good. To determine the maximum false pavement thickness and rolling combination that can be compacted at one time by LSV220 single drum vertical vibratory roller. In this paper, two test sections with a virtual pavement thickness of 100 cm were set up to bury the sensors in layers. Through the sensor and sand filling method, the maximum thickness of the compacted pavement is 47cm, and the combination of rolling is "one static compaction, one Weak vibration, and three Strong vibrations" The BP neural network was constructed using the experimental data was used to predict the compaction by sensor data. The error between the predicted and measured values of this neural network was verified to be 1.44%.