Optimization Model Design of Deep Foundation Pit Support Based on Improved Fish Swarm Algorithm

The optimization design of deep foundation pit supporting structure has always been a difficult and hot issue with complex technology and strong comprehensiveness. There are many factors involved in optimization, and the variables are discrete and the objective function is multi-peak. Using the usual optimization methods, the algorithm often falls into the local optimal solution. In the design of deep foundation pit, the designer usually needs to adjust the design parameters for many times and check again, so that the calculation results can meet the design requirements. This paper discusses the possibility of combining deep foundation pit support engineering design with improved artificial fish swarm algorithm (AFSA), and puts forward an optimization model of deep foundation pit support based on improved AFSA. Simulation results show that the proposed algorithm has obvious advantages over the compared methods in average error, prediction accuracy and corresponding speed. This method overcomes the shortcoming of traditional optimization method that it is easy to fall into the local optimal solution, and reduces the requirements for the properties of the objective function. Introducing the improved AFSA into the optimization design of deep foundation pit support engineering can effectively overcome the difficulties of traditional optimization model and provide a new idea for the optimization design of support engineering.