The Improved Ant Colony Algorithm to Predict Longitudinal Long-term Uneven Settlement of Shield Tunnel in Soft Soil
The dominant information with some regulations to follow which is obtained through analyzing in depth the law of the measured uneven settlement of shield tunnel in soft soil and its causes after long-term operation, and the auxiliary information with the strong randomness which should be taken in account, are used to tentatively improve the ant colony algorithm researched earlier to predict settlement. In the process of training the new model, the prediction accuracy is analyzed, and the rational thought to further improve the prediction model is discussed also. The results show: ①the dominant information is that the long-term longitudinal uneven settlement of shield tunnel in soft soil has something to do with the cumulative sedimentation magnitude of settlement trough bottom and to do with the distance between trough bottom and trough wall or trough top;②the new prediction model is provided with clear calculation process, simple formula, being easily used and high prediction accuracy; ③to establish the information function and heuristic function reflecting that uneven settlement changes with time and space in the prediction model is the key to improve the prediction model based on ant colony algorithm. (Abstract)
soft soil shield tunnei long-term uneven settlememt prediction model of settlement ant colony algorithm(key words)
Kai Wei Quan-mei Gong Shun-hua Zhou
Traction Power State Key Laboratory Southwest Jiaotong University Chengdu, China Key Laboratory of R Key Laboratory of Road and Traffic Engineering of Ministry of Education Tongji University Shanghai,
国际会议
成都
英文
1774-1778
2010-12-17（万方平台首次上网日期，不代表论文的发表时间）