Research on intelligent drainage control strategy of tunnel based on GABP

In this paper, the author uses the genetic algorithms (GA) to optimize the traditional BP neural network model, and establish the GABP neural network model. The Zhongshan Road underground passage of Wuchang Railway station is taken as the object of study to establish the tunnel drainage system based on the GABP neural network and does the simulation experiment. Compare to the traditional BP neural network, GABP can reduce the computational amount and training time, speed up convergence and improve forecast accuracy. Meanwhile, by adopting the method of combination of quality and quantity to do the system analysis and synthesis, the design of the controller will shake off restricts of system model, make the algorithm simple and the robustness great.