Dynamic Cost Estimation of Reconstruction Project Based on Particle Swarm Optimization Algorithm

Abstract

In order to predict the value of dynamic cost estimation of reconstruction project, this paper proposes the research on dynamic cost estimation of reconstruction project based on particle swarm optimization algorithm. Firstly, the applicability of example swarm optimization algorithm is introduced. The basic principle of particle swarm optimization algorithm is described, and PSO (particle swarm optimization) algorithm is used to optimize the super parameters of LS- SVM. With the help of spss20 0 to cluster the sample data to obtain similar engineering classes. In order to better verify the application effect of the optimized model in project cost prediction, BP neural network, LS - SVM and PSO - LSSVM are selected to simulate and predict the project cost. The results show that the relative errors of the three models are controlled within + 10%, which can meet the accuracy requirements of construction cost prediction in the early stage of construction. The relative error distribution interval predicted based on BP neural network model is [- 7.46%, 5.74%], and its range is 13.12%; The relative error distribution interval predicted based on LS SVM model is [- 8.12%, 6.17%], and its range is 14.22%; The relative error distribution interval predicted based on PSO -LSSVM model is [- 2.56%, 2.49%], and its range is 5.21%. The prediction model optimized by PSO algorithm is better than LS SVM model in prediction stability, and the prediction effect is more robust. In conclusion, the prediction model based on PSO optimized LS SVM has good guiding significance for the construction cost, and is more suitable for the prediction of the construction cost in the early stage of construction.