Font Size: a A A

Research On Construction Risk Management Of Deep Foundation Pit Engineering Based On BP Neural Network

Posted on:2017-09-29Degree:MasterType:Thesis
Country:ChinaCandidate:F Y NiuFull Text:PDF
GTID:2322330512958731Subject:Architecture and civil engineering
Abstract/Summary:PDF Full Text Request
In recent years,with the efforts of the country to promote urbanization continues to increase,a large number of population into the city led to the city,especially the city's core business circle construction land supply and demand has been in a more tense state.In order to make better use of scarce land resources,all kinds of high-rise,super high-rise buildings are gradually rise,and parking lot,shopping malls are also gradually transferred to the underground.The development and construction of the above construction,promote the rapid development of deep foundation pit project.However,theoretical research on geotechnical engineering need to be improved,and the deep foundation pit engineering is greatly affected by the natural environment and social environment,the construction itself has the characteristics of long time,high cost,complex technology,strong concealment,resulting in the whole construction process is full of uncertainty and risk.Therefore,the system and scientific evaluation of the risk of deep foundation pit in its construction process for the construction unit has a certain practical significance,and it is convenient to guide the construction of scientific.Taking the deep foundation pit engineering construction risk management as the breakthrough point,this paper analyzes the risk factors in the construction process of deep foundation pit engineering.Using WBS-RBS risk identification basic principles will be deep foundation pit construction risk and construction units,construction units,prospecting units,design units,supervision units of these five different organizations linked together,this is helpful to identify the risk factors in the construction process of deep foundation pit from the perspective of project participants,and combined with the construction process of deep foundation pit,the risk identification is carried out from five construction processes of earthwork excavation,slope support,foundation treatment,dewatering and drainage and foundation pit monitoring.Then use the principal component analysis(PCA)to do preliminary screening and reduce the complexity of risk matrix.And then use the data envelopment analysis(DEA)to do the two screening of risk factors,to further eliminate the non main factors,and enhance the monitoring of key risk factors.Using the analytic hierarchy process to construct the hierarchy of risk index system,reduce the complex relationship between the indicators,and the final screening of the risk indicators to empower.At last,the BP neural network is used to train the risk index and the sample set after two times to improve the precision of the artificial intelligence operation,and the construction risk grade of the deep foundation pit is obtained.Risk management of deep foundation pit engineering is affected by several factors,and the whole process is more complex,so look for the risk assessment of a scientific and suitable model plays a very important role in the evaluation results,which is conducive to improve the scientific risk management of deep foundation pit engineering construction.This paper attempts to combine the four different methods of the principal component analysis method,data envelopment analysis,AHP method and BP neural network to achieve the merits of complementary.And then it is proposed that the data envelopment analysis method is used to eliminate the redundant indexes,which greatly improves the accuracy and speed of the BP neural network algorithm.Finally,the risk evaluation model is used to evaluate the engineering practice of Qingdao Hengda Jinsha Deep Foundation Pit,and to test the scientific and applicability of the model.The results show that the evaluation results are in good agreement with the actual situation,so the comprehensive evaluation model of construction risk of deep foundation pit constructed in this paper has certain scientific and applicability.The model is established to enrich the risk assessment method of deep foundation pit engineering construction,and provides scientific theoretical guidance for the construction process of deep foundation pit engineering.At the same time,it also has a certain practical significance to make risk prevention measures and improve the safety of the whole construction process of deep foundation pit engineering.
Keywords/Search Tags:Deep foundation pit engineering, Risk management, BP Neural Network, Principal Component Analysis(PCA), Data Envelopment Analysis(DEA), Analytic Hierarchy Process(AHP)
PDF Full Text Request
Related items