Optimization Of Supporting Schemes For Deep Foundation Pit Based On Fuzzy Grey Relation Projection Method

Along with the development of China's construction industry, the construction of High-rise buildings and subways brought a lot of deep excavation. In the excavation, inappropriate choice for supporting projects led to many accidents. Therefore, the reasonable choice of deep foundation pit supporting schemes is very important.The evaluation and selection of deep foundation pit supporting schemes is a multi-objective and complex issues, which is influenced by many factors. Grey Relation Projection Method based on the grey system theory and vector projection principle is a multi-objective decision making method. It can comprehensively analyze the relationships among the evaluation index, reflect the impact of the factor index space, and avoid single direction deviation.The principle of grey relation projection method was outlined, and its advantages and disadvantages were analyzed and summarized. On this basis, combined with fuzzy theory to improve the gray relational projection method, a fuzzy gray relational projection method was put forward, and it was first applied for deep foundation pit decisions.Based on the analysis of the deep foundation pit program factors, the deep foundation pit selected index system was established, and all factors'content and physical meaning were described, and all index values were determined by using the statistical method. The optimization of supporting schemes is established based on the gray relation projection method. Meanwhile, the determination of deep foundation pit evaluation index selection was solved by the combination weighted method of Analytic Hierarchy Process (AHP) and information entropy, which won success in practice. The optimal solution is proposed based on discrimination sensitivity analysis method. The sensitivity analysis provides a scientific basis for the selecting of key parameters that is the values of coefficient of expert preferences in the combination weighted method. Compared with several other algorithms for the sensitivity, it shows the superiority of this method.