Research On The Deformation Monitoring And Analysis Of Multi-sensor Tailings Pond Based On Beidou

As a dangerous source of man-made debris flow with high potential energy,tailings ponds have attracted much attention for their safety.The National Mine Safety Administration has clearly pointed out that it is necessary to ensure that the dry beach length,safety ultra-high,flood control storage capacity,buried depth of infiltration line,reservoir water level and other parameters of tailings pond are always under control.It is necessary to establish a sound early warning mechanism for typhoon,rainstorm,continuous rainfall and other extreme weather.With the rapid development of Internet technology,more and more of the multi-source sensor data were used to characterize the operation of the tailings health status,how to make good use of heterogeneous data mining tailings health status and the development law,and based on the existing data to predict the future trend of the perceived data to take preventive measures is an urgent need for the current tailings pond safety monitoring system.According to the existing research data,rainfall and fluctuation of reservoir water level height are important factors for the instability of tailings pond,and numerous tailings pond accidents are inextricably related to both.At present,the existing safety monitoring system of tailings pond only simply monitors the basic parameters of tailings pond,ignoring the role and influence of rainfall factors on various state parameters,and atmospheric precipitable water,as an important characterization factor of rainfall formation,has not been used in the monitoring system of tailings pond.Therefore,it is of great significance to carry out safety monitoring data analysis and precipitation impact study of tailings pond,and predict future data trend according to existing monitoring data,so as to realize the safe operation of tailings pond under continuous rainfall weather and ensure the safety of downstream people's lives and property.Based on the tailings pond safety monitoring project,this paper mainly conducts in-depth studies in the following aspects and has achieved certain research results:(1)Based on the temporal and spatial variation law of the tailings pond perception data,the data collected by various equipment in the tailings pond safety monitoring system are analyzed.The results show that the perception data of the tailings pond are correlated with each other,and the surface displacement of the tailings pond is affected by the stack structure of the dam body,and the deformation degree of the downstream is smaller than that of the upstream.The number of subdams also affects the stability of the surface of the dam body.(2)The method of atmospheric precipitable water vapor inversion by ground-based GNSS is tested.The GAMIT software is used to calculate the total delay in the zenith direction of the tailings pond,the static zenith delay obtained by the tropospheric delay model,and then the zenith wet delay is separated,and finally the atmospheric air over the tailings pond is obtained by the conversion factor.Precipitation.Correlation analysis between PWV and actual precipitation in the tailings reservoir area shows that:The PWV value and actual precipitation have the same phased change rule in time distribution,and the correlation coefficient between the two can reach 0.459 under heavy rainfall.The faster the change of the atmospheric precipitable value,the higher the probability of rainfall.Therefore,the atmospheric precipitable value has a certain indicative effect on the occurrence of rainfall and can be used as an important indicator for the safety monitoring of tailings ponds.(3)Use Python to extract the sensing data of the tailings library(reservoir water level,dry beach length,surface displacement,deep displacement,burying depth of the infiltration line)and resample the experimental data through a third-party library in Python,Unify all perception data into the same time unit,and filter out abnormal values.Finally,the influence of rainfall on each sensing data is analyzed for each sensing data.The results show that,except for surface displacement and deep displacement,other sensing data show a relatively strong response to rainfall factors.The data is unified to the same time unit,and outliers are filtered out.(4)Use the support vector machine algorithm to predict and analyze the water level and settlement displacement of the tailings reservoir,and design a variety of predictive factor schemes for comparison experiments.Finally,it is concluded that the prediction of the height of the reservoir water level should use atmospheric precipitation,actual precipitation,dry beach length,reservoir water level,and internal displacement as predictive factors.The mean square error between the predicted result and the actual measured value is 0.045 m.The tailings reservoir subsidence displacement prediction model should use atmospheric precipitation,actual rainfall,dry beach length,reservoir water level,and internal displacement as predictive factors.The mean square error between the predicted result and the actual measured value is 0.616 mm.The support vector machine prediction model has achieved satisfactory results in the short-term numerical prediction,but it is not suitable for the prediction of long-term series.(5)Traditional flood prevention and drainage analysis of tailings ponds is often manual calculation on a regular basis,which has defects such as large engineering volume and low efficiency.Based on the data of real-time water level monitoring of tailings pond and dry beach length monitoring,this paper obtains the basic data of flood regulation calculation,studies the model of flood regulation calculation,realizes the real-time flood regulation calculation of tailing pond,and provides guarantee for the safe operation of the tailing pond.Figure[62]Table[13]Reference[76]...