Research On Risk Management Of Deep Foundation Pit Construction Of High-rise Building Based On DS Theory-neural Network

With the acceleration of urbanization,the scarcity of urban space resources has intensified,and people's demand for urban space has promoted the development of deep foundation pit projects.Deep foundation pit engineering usually has problems such as limited construction space,complex and diverse hydrogeological conditions,harsh construction and surrounding environment,and many unforeseeable factors.In addition,the dynamic nature of risks in the construction process increases the possibility of safety accidents and the difficulty of construction risk management.How to conduct overall dynamic management of deep foundation pit construction risks is a very important and necessary research project.Based on the general process of deep foundation pit construction risk management,this paper studies the related theories and methods of deep foundation pit construction risk management,establishes the risk management system of deep foundation pit construction based on DS theory-neural network,and focuses on the dynamic management process of construction risk with trigger alarm and monitoring risk index as the core.Through the quantitative judgment of the risk level of deep foundation pit,the safety state of the foundation pit is evaluated,and at the same time,the source of risk is accurately located according to the dynamic monitoring system,so as to achieve the purpose of effective risk response.The results which this paper mainly research are as follows:(1)The group evaluation method was used to determine a reasonable and diversified expert group of 5 members.On the basis of collecting and sorting out literature data,engineering data and expert opinions,based on the WBS-RBS coupling matrix method,according to the construction process,the risk indicators are decomposed from the aspects of personnel,technology,management and environment safety risks,and the risk factors are comprehensively identified.There are 26 risk factor indicators on 3 floors,and a risk factor indicator system for deep foundation pit construction has been established.(2)A risk management system based on DS theory-neural network has been established,which takes risk indicators as the premise and includes four stages:risk assessment,risk dynamic monitoring,risk alarm and risk response.In the system,risk factor index₄)and monitoring risk index?are used to jointly complete the risk assessment of deep foundation pit.In the data prediction part of the monitoring project at this stage,The relative error between the predicted value of the monitoring data and the measured value of the monitoring data is less than 0.1%,which means there is a high prediction accuracy.Dynamic risk monitoring and risk warning jointly form a dynamic management environment for the risk of foundation pit.When the monitoring data are abnormal or the value reaches the alarm standard,the early warning and forecast system will be triggered and the risk response behavior will be triggered.(3)Taking the actual case deep foundation pit project of ZH international square in henan province as an example,the risk management system based on DS theory-neural network was used to complete the whole process of risk management implementation.The risk level of the deep foundation pit of the evaluation case is level1,which is consistent with the safety status of the site construction.The yellow alarm was triggered in the process of the risk management of the project construction in the case.According to the risk trigger response plan,the corresponding response measures were taken for the triggered alarm risk events to effectively reduce the probability of the occurrence of safety accidents.