Research On Cost Forecast Of Construction Project Based On BSA-ELM Model

The development of construction industry has made a significant contribution to the national economy.However,due to its complex composition,large number of personnel,long working cycle and high cost basis,once a construction enterprise has a cost management problem,it will often cause an increase in costs,which will increase the financial burden of a larger enterprise.As a primary link to comprehensively strengthen enterprise cost management and has a positive effect on reducing cost.In order to better achieve cost management,the link of cost prediction research must be strengthened.This dissertation takes the cost prediction of construction project as the research object,and mainly studies the cost prediction of construction project according to its complex and unpredictable structure.Firstly,this dissertation explained the theoretical basis,and mainly expounds the related concepts of construction cost prediction in detail.Secondly,from the perspective of construction enterprises,the main factors that affect the cost of construction projects were identified and analyzed,and the cost impact index system of construction project was established according to principles and steps of establishing the index system.Finally,aiming at the randomness of parameter setting in the extreme learning machine(ELM),the bird swarm algorithm(BSA)was used to optimize the parameters,so as to improve the accuracy of cost prediction of construction projects.On this basis,the construction cost prediction model of BSA-ELM was established and the parameters of the model were analyzed and set in detail.In order to effectively validate the construction cost prediction model of BSA-ELM,it was applied to the actual construction project case,and the prediction result was compared with multiple prediction models.The results show that the construction cost prediction model of BSA-ELM is more accurate,which verifies the feasibility and effectiveness of the model and provides a reference for the related cost prediction research.