Safety Monitoring and Early Warning Indexes of Urban Underground Space
In order to ensure the safety of the development and utilization of urban underground space and promote the sustainable development of cities,this work proposed five principles of urban underground space monitoring by means of literature investigation,field investigation and expert consultation.The monitoring objects were divided into three categories,i.e.,engineering structure,surrounding rock and soil mass,and surrounding environment.The monitoring indexes were summarized into four categories of deformation,mechanics,vibration and macrostate,among which the deformation indexes implemented double control requirements,and the other three indexes implemented single control requirements.The prediction of monitoring trends adopted the formula method,regression analysis method,time series analysis method,gray prediction method,neural network method and support vector machine method.The monitoring control values were basically the same throughout the country,but the early warning classification standards varied between different cities.The graded early warning of Beijing and Guangzhou cities provides significant reference for other areas.Seven deficiencies have been proposed for the safety monitoring of urban underground space,i.e.,imperfect standards of warning classification,low efficiency of manual monitoring,single monitoring parameters,lack of monitoring information sharing,poor measurement accuracy,emphasis on monitoring rather than prediction,and lack of data fusion and machine learning applications.In view of these problems,seven measures were recommended,i.e.,the establishment of reasonable early warning classification standards,the development of automation and intelligent monitoring,multi-parameter comprehensive monitoring,the application of remote monitoring and cloud platform,the development of high-precision measurement equipment,equal emphasis on monitoring and prediction,and data fusion and machine learning.
underground engineeringsafety monitoringmonitoring indicatorstrend predictionearly warning and classification
万方数据
维普
