Research On Foundation Pit Deformation Prediction Based On GA-PSO Combination Optimized BP Neural Network Model

The rapid development of the national economy and urbanization has promoted the rapid development of infrastructure.The number of construction of foundation pits in high-rise buildings,underground malls,subway tunnels and other facilities has increased dramatically.The deepening of the pit excavation and the expansion of the area have also brought many hidden safety hazards.Therefore,the foundation pit is deformed during the entire construction and operation process.Monitoring is essential.Based on the deformation monitoring data of the foundation pit,the deformation characteristics of the foundation pit are analyzed,and a suitable deformation prediction model is established to predict the foundation pit deformation,which provides a reference for the safe construction decision of the foundation pit project and it is particularly important to prevent accidents.This paper takes the foundation pit monitoring project of a Square in Tai’an as the background,applies related theories and research methods,and builds a foundation pit deformation prediction model based on BP neural network based on actual field monitoring data to predict future development.The BP neural network model is deduced and analyzed,and then the genetic algorithm(GA)and particle swarm algorithm(PSO)are used to optimize the BP neural network model,and the GA-BP neural network model and the PSO-BP neural network model are constructed.Based on this,Combining the two algorithms,the GA-PSO-BP model is proposed,and the on-site monitoring data is predicted.Finally,the prediction results of the four models are compared and analyzed.The main research contents of this article are as follows:(1)The deformation foundation theory of foundation pit is analyzed,and the purpose,content and method of foundation pit monitoring are introduced systematically,which provides a reliable basis for subsequent theoretical research.(2)Explored the theory of BP neural network,introduced the learning process of BP neural network in detail,and pointed out the advantages and disadvantages of BP neural network.The genetic algorithm and particle swarm algorithm were introduced to optimize the BP neural network model.Finally,combining the advantages of genetic algorithm and particle swarm algorithm,a GA-PSO combination algorithm was constructed to optimize the BP neural network and used in deformation prediction.(3)By comparing and analyzing the measured data in this paper with the prediction values of the four prediction models,and selecting the three indicators of prediction results,MAE,MSE,and MAPE,to evaluate the superiority and reliability of the model,we can conclude that the BP neural network model is in the foundation The prediction accuracy of engineering examples is good.The prediction accuracy of the BP neural network model optimized by genetic algorithm and particle swarm optimization is higher.The prediction accuracy of the BP neural network model optimized by GA-PSO combination is the highest and the applicability is stronger.Use this combined model to effectively guide the construction of foundation pits.