Font Size: a A A

Expressway Engineering Investment Estimation Model Research Based On Single Hidden Layer Feed-forward Neural Network Limit Algorithm

Posted on:2018-12-07Degree:MasterType:Thesis
Country:ChinaCandidate:W DengFull Text:PDF
GTID:2359330518961657Subject:Management Science and Engineering
Abstract/Summary:PDF Full Text Request
Firstly,this paper analyzes the background and significance of the research,combing the domestic and foreign research status,according to the current situation of Expressway Project Investment Estimation,find the existing problems of Expressway Investment Estimation;This paper introduces the definition of Expressway Investment and the related Investment Estimation Theory,and compares the advantages and disadvantages of the existing investment estimation methods.At the same time,from the investment structure,main characteristics,related documents,the Grey Theory is used to analyze the characteristics parameters of Expressway Project Investment Budget,and establish the characteristic parameter system of investment estimation engineering.Combining the Grey Relational Analysis Method and Single Hidden Layer Feedforward Neural Network Limit Algorithm,the relevant model is established,seven engineering characteristics parameters are introduced into expressway concept,and the error is defined as the model,and the parameters need not repeated multiple training.finally,the proposed algorithm is scientifically validated to highlight its good application flexibility and adaptability.Compared with other network estimation methods,the Single Hidden Layer Feed-forward Neural Network Limit Algorithm can more accord with the requirements of expressway construction project,and its budget accuracy and widely applicability,has obvious advantages in the estimation process and speed.Using the superiority of the Single Hidden Layer Feed-forward Neural Network Limit Algorithm in calculation speed,provides a new estimation method for expressway investment estimation,which has important innovation significance.However,there are some deficiencies.In the collection of data,because of the application of Single Hidden Layer Feed-forward Neural Network Limit Algorithm is less,the reference material is not much,this gives some difficulty,the data may be collected is not comprehensive enough;After a lot of practice,engineering characteristics have a certain impact on the total value of its investment,but in the actual budget process,it will miss some factors,and then produce error;In the process of practice,the information,data,literature,general situation and other information(completed project,building project)related to expressway project are essential,which increases the workload of information collection.Therefore,the area covered by the case information may not be broad enough.
Keywords/Search Tags:Highway Engineering, Investment Estimation, Index, Grey Correlation, Neural Network
PDF Full Text Request
Related items