Prediction Of Landslide Displacement Data Based On Particle Swarm Optimization Algorithm

The monitoring contents of landslide are mainly divided into two parts:displacement monitoring and crack monitoring.In recent years,experts at home and abroad have been devoted to the study of landslide displacement monitoring,and discovered that landslide displacement is affected to various factors such as formation lithology,rainfall amount,underground water level,earthquake and human activity.Once landslide occurs,it will bring incalculable loss to human life and production,so landslide prediction and forecast research is of great significance.At present,there are many prediction models of landslide displacement.The single prediction model has the disadvantages of slow operation speed and low prediction accuracy,and the prediction results are unstable and the reliability is poor.BP neural network,as a network system that mimics the characteristics of human brain,has strong nonlinear mapping ability and is often applied to landslide displacement prediction.But a single neural network model has shortcomings such as difficult to determine hidden layer and slow convergence speed and so on.The prediction accuracy is not ideal.In this paper,we use particle swarm optimization(PSO)to optimize the weights and thresholds of BP neural network,which not only optimizes network structure,but also improves convergence speed.Due to the premature convergence and easy to fall into the local extremum,the weight of standard PSO is improved in this paper,which are linear decrement inertia weight method,random inertia weight method and nonlinear inertia weight method.Dynamic weights can improve the global search ability of PSO by adjusting the speed and location of particles.Two improvements are carried out on the basis of the dynamic weight improvement method.Chaos theory is used to identify the location of the best particle in PSO,in order to avoid the particle swarm into local extremum,and improve the shortage of the combination of the standard PSO and the BP neural network.The improved particle swarm optimization BP neural network model is applied to the engineering practice of landslide displacement prediction.With the help of MATLAB programming,the prediction model is used to predict the landslide displacement data,and the predicted data obtained by different improved algorithms are compared and analyzed.It is concluded that the chaotic PSO is better for the prediction of the BP neural network.The conclusion is that the prediction accuracy is high.Experimental research shows that the model can be applied to landslide displacement prediction.