Zhou, C (Zhou,C.); Ding, LY (Ding, L. Y.); He, R (He, R.)

下载地址：

http://www.sciencedirect.com/science/article/pii/S0926580513000290

Abstract：The excavationface stability is crucial for safety and risk management in slurry shieldtunneling, especially for the river-crossing tunnel. To avoid face collapse orblow-out, shield operators need to keep air chamber pressure balanced usingtheir own experience, which would be difficult, discontinuous and less reliablein the process of construction. Considering the disadvantage of the manualcontrol process, this paper presents a predictive control system for airchamber pressure in slurry shield tunneling using Elman neural network (ENN)model. It mainly contains a theoretical model, an ENN predictor and an ENNcontroller to set optimal control parameters automatically tracking the desiredair chamber pressure. Moreover, to improve the learning capability of ENNmodel, a particle swarm optimization (PSO) algorithm is implemented. Thissystem has been tested with collected data of slurry shield operationparameters in the Yangtze riverbed metro tunnel project in Wuhan, China.Analysis revealed that the predictive control system using PSO-based Elmanneural network model in this paper has the potential for enhancing facestability in slurry shield tunneling.

Keywords:Air chamber pressure;Elman neutral network; Particle swarm optimization; Predictive control system;Slurry shield parameters; Yangtze riverbed tunnel