Intelligent Real-time Monitoring And Early Warning System During Construction Phase Of Highway Tunnel

In order to ensure the smooth progress of highway tunnel construction projects,it is necessary to monitor and analyze the deformation of the tunnel,so as to make an accurate and objective judgment on the stability of the tunnel structure system,and take countermeasures when necessary to avoid danger.When a tunnel collapse accident occurs during construction,the precise location and vital signs of the construction personnel are the important basis for formulating a rescue plan.The existing monitoring systems at home and abroad are mostly for specific projects or specific equipment monitoring.The system has a small scope of application,insufficient data processing,and lack of safety monitoring of tunnel construction personnel.Based on this,this paper takes the monitoring of tunnel deformation and personnel safety during the construction of highway tunnels as the research object,and integrates various monitoring information including structure monitoring and personnel monitoring into the same monitoring system to realize non-contact real-time deformation Monitoring and personnel safety monitoring.The main research contents and results of this article are as follows:(1)Demand analysis of intelligent real-time tunnel monitoring and early warning system during construction: Combined with the characteristics of deformation during tunnel construction,the accuracy requirements of deformation monitoring are determined according to construction specifications and engineering experience.The accuracy of deformation monitoring is 0.5-1 mm.Combining vital sign acquisition means and indoor positioning means,a safety monitoring method for personnel is proposed.Monitor the pulse of the construction personnel to obtain the life status,and obtain the position information of the personnel through the UWB positioning system.The accuracy of the positioning system is less than 10 cm.The system requirements for response speed,standby time,abnormal identification and data cleaning are determined according to the characteristics of the construction environment and construction management requirements.The response time of the system does not exceed 1min,and the monitoring system can be used normally in the environment of excavation and explosion impact.(2)Construction of tunnel intelligent real-time monitoring and early warning system:According to the requirements of the tunnel construction intelligent monitoring subsystem,design the hardware equipment such as tunnel monitoring perception subsystem,communication subsystem and central control subsystem;design the software framework of the monitoring system,including the development platform,system instructions,database and system interface.Make each module of the system cooperate and coordinate with each other.The perception system obtains the construction site data under the command of the central control system,immediately calculates and judges whether there is an abnormality according to the early warning algorithm,and starts the on-site sound and light alarm device when an abnormality occurs.The original monitoring data and analysis and judgment results are shared and interacted through the communication system,cloud platform and related engineering personnel.The on-site equipment can work normally under the condition of-40 ℃ to 80 ℃ without condensation,the power consumption is not more than 10 watts,and the working time is not less than 24 hours under the condition of power off.(3)Real-time deformation monitoring of tunnel sections and intelligent identification technology of abnormalities: Through the analysis of tunnel deformation monitoring data and stability judgment,the anomaly recognition algorithm of the monitoring data is selected as the distance recognition.By comparing the data interpolation correction algorithms,the optimal interpolation method is determined as linear interpolation,nearest point interpolation and segmentation3 Sub-Hermit interpolation can be used according to the actual situation.Determine the data preprocessing method of the monitoring system.The long-term and shortterm memory neural network is used to predict the data of tunnel deformation monitoring,and the model results after parameter adjustment are compared.When the parameters of the model are the learning rate 0.01,the hidden layer 16,the number of iteration steps 400 and the unit input 20,the effect of predicting the tunnel deformation amount by the system is the best.Under the condition of this parameter,the average operation time of the model is 2.1968 min,the average number of steps from iteration to convergence is 270,the average convergence mean square error is 0.000026,and the prediction results fit well with the measured data.The conversion formula of inclination angle and displacement is established according to the layout scheme of the inclination sensor in the tunnel section.Based on the above conditions,an early warning algorithm for tunnel deformation monitoring is determined to realize tunnel deformation monitoring and early warning.(4)Research on vital signs real-time sensing and personnel positioning technology for tunnel construction based on portable equipment: After real-time collection of human pulse data and pre-processing of the data,K-means algorithm is used to judge the data to realize the abnormal recognition of construction personnel’s heart rate.There are three cases of tachycardia,bradycardia and normal heart rate.The algorithm recognition results of tachycardia and bradycardia were compared with clinical judgment results,and the accuracy of algorithm judgment reached 98.7879% and 100% respectively under clinical judgment.Analyze the samples with normal heart rate that are different in the two ways.Judging the content of normal center rate,tachycardia and bradycardia in the sample,the accuracy rate when using the algorithm to identify is 100%,81.3953%,84.1270%,indicating that the accuracy of the selected algorithm is higher than that based on a single data Clinical judgment method.The positioning base station deployed on the tunnel construction site,then uses ultra-wideband technology and UWB to obtain the location data collection of the construction personnel and performs algorithm inverse solution to achieve the location of the construction personnel in the tunnel.Based on the above conditions,an early warning algorithm for tunnel personnel safety monitoring is determined to realize personnel safety monitoring and early warning.