Research On The Optimization Of High-rise Structure Demolition Decision System On Entropy Weight TOPSIS Method

With the continuous progress of urbanization,some buildings(structures)have to be demolished due to their aging,functional degradation,environmental protection or other policies.However,there are few researches on non blasting demolition system at home and abroad.In addition,compared with other types of structures,high-rise structures are different in safety evaluation,demolition mode adaptability and project cost calculation due to their structural characteristics and construction difficulty.Based on the above reasons,this paper designs a high-rise structure demolition decision-making system by reading a large number of domestic and foreign demolition technology,neural network,decision-making system and other related literature.This software is based on the static demolition expert decision system V1.0of high-rise structures,and the following improvements are made: summarize the current new demolition methods,use the access2010 software to establish a database,and compile a case base for the demolition cost and construction period of high-rise structures;for dismantling the expert decision system,using MATLAB toolbox to predict the problem of computation is time-consuming and laborious,and BP algorithm model convergence speed is slow and local minimization is serious.On this basis,combined with adaptive differential evolution algorithm,the hybrid algorithm ade-bp algorithm is formed.Using the actual demolition project case,by determining the appropriate number of hidden layer nodes,using the G(1,1)model of gray theory to process the samples,the adaptive differential evolution algorithm to optimize the initial weight and threshold,the experiment shows that the prediction accuracy meets the requirements;Using the above improved algorithm,the cost and construction period of different demolition schemes for a specific demolition structure are predicted.The entropy weight method is used to determine the weight of each index in the demolition scheme.The TOPSIS method is used to rank the advantages and disadvantages of each demolition scheme,and the decision scheme system based on the entropy weight TOPSIS method model is established.This paper summarizes the principles,advantages and disadvantages of common index weight determination methods and comparative evaluation methods of various schemes,and demonstrates the operation process by using specific cases,and proves the effectiveness of the method;With the help of GUI tools in MATLAB,a friendly interface is designed.Among them,the interface content is divided into demolition method database,demolition cost and construction period case database,improved prediction algorithm and Entropy TOPSIS method scheme decision-making.The main conclusions are as follows:(1)By comparing the two norm curves of the relative error of different hidden layer nodes,it can be concluded that the hidden layer nodes are the key parameters affecting the BP neural network model;Compared with the original data and G(1,1)data processing,the performance of the model is improved.(2)The improved algorithm and the improved algorithm are compared and analyzed.It is proved that it is necessary to carry out the data processing,G(1,1)processing and ade optimization of BP neural network.The improved algorithm has a prediction accuracy of 5.13%,which meets the actual needs.(3)Aiming at the four factors that affect the demolition decision of high-rise structure,this paper uses the objective weighting method entropy weight method to determine the weight,combined with the ideal approximation method to quantitatively compare the gap between the various schemes,and uses the real case as a test,the results are in line with the real situation,which proves that this method is feasible in the demolition decision of high-rise buildings,The model can be used in the decision-making system of high-rise structure demolition.