为减少盾构施工过程中的人工干预，提高施工效率，解决人员误判或经验不足等情况导致的施工质量差、效率低，遇复杂地质难以及时调整，甚至引起施工事故等问题，结合盾构施工工艺、工程地质等因素，基于大数据与人工智能技术，研发一套基于MLP-ARX(多层感知机自回归)模型的盾构自动掘进控制系统，开发推进过程动态模型以及模型在线更新、参数优化的方法，在施工过程中能够及时、准确地自动调节盾构推进速度、分区压力和螺旋机速度等参数，并应用在深圳地铁14号线管线工程隧道。结果表明： 该系统可实现推进过程的自动控制，推动盾构施工自动化、信息化和智能化，极大地缓解盾构司机的操作压力，降低对施工情况误判的发生概率，提高施工质量和效率。

A set of automatic shield tunneling control systems based on a multilayer perceptron autoregressive model is developed based on shield tunneling technology, engineering geology, big data, and artificial intelligence technology to reduce the manual intervention in the construction process of shield tunneling, improve the construction efficiency, and solve the problems of poor construction quality and low efficiency caused by personnel misjudgment or lack of experience, difficult to adjust in time in case of complex geology, and construction accidents. Meanwhile, a dynamic model of the propulsion process and the method of online model update and parameter optimization are developed, which can timely, accurately, and automatically adjust the tunneling parameters, such as shield propulsion speed, zone pressure, and screw conveyor speed. The system has been applied in the pipeline tunnel of the Shenzhen metro line 14. The results show that the system can automatically control the propulsion process, promote the automation, information, and intelligence of shield tunneling, greatly reduce the labor intensity and the occurrence of misjudgment of construction conditions, and improve the quality and efficiency of shield tunneling.