Study On Pollutant Diffusion And Traceability Technology Of Sudden Air Pollution Accident On Typical Complex Underlying Surface

In recent years,with the rapid development of the economy,sudden air pollution accidents under the complex underlying surface environment occur frequently,which poses a great threat to people’s life and health,socio-economic development and ecological environment.Rapid traceability and accurate diffusion simulation of pollutants in complex underlying surface environments are important prerequisites for rapid emergency response to sudden air pollution accidents,reducing accident hazards and ensuring people’s lives and health.It is also a scientific and methodological problem that needs to be solved urgently.Based on the tracer gas field simulation experiment of sudden air pollution leakage accident carried out on the real complex underlying surface,this study explored the gas diffusion law under the complex underlying surface,focused on the research on the key technology of small-scale pollution diffusion simulation and estimation of accident source parameters,and established the"monitoring-traceability-simulation"feedback technology system.The research results are of great significance for improving the scientificity and effectiveness of the response to sudden air pollution accidents,and reducing human health hazards and economic losses.Five field leakage accident simulation experiments of tracer gas（SF₆）were carried out in the tank（the height H was 22m）zone of a typical chemical enterprise,including four leakage and diffusion experiments of single point source at different heights（0.90H,0.77H and 0.27H）and one double point sources leakage and diffusion experiment,making up for the lack of pollutant diffusion experimental data in the real complex underlying surface environment.Based on the experimental results,the effects of leakage height and storage tank on pollutant diffusion were analyzed.It was found that when the leakage height was lower than the tank roof,the maximum ground concentration decreased with the increase of leakage height（0.77H～0.90H）,but when the release height（1.05h）was slightly higher than the tank roof,the maximum ground concentration（338.3ppbv）of gas was significantly higher than that of tank roof（0.90H）（122.7ppbv）.The pollutant diffusion was obviously affected by the downwind storage tank.The accurate simulation technology of complex underlying surface was established based on CFD model（fluent）,and the effects of typical parameters/schemes（inlet boundary,meshing and turbulence model）on numerical simulation were deeply analyzed.An inlet boundary selection scheme suitable for accurate CFD simulation under complex underlying surface was proposed.Hybrid polyhedron-hexahedron mesh and Realizable k-εmodel were most suitable for the diffusion simulation of gas leakage accident on complex underlying surface.Based on the established CFD model,the influence of leakage source on pollutant diffusion was carried out.It was found that the differences in micrometeorological conditions between the tank roof and the tank wall were present,resulting in the maximum gorund concentration of gas released from the tank roof（1.05H）being higher than that from the tank wall（0.90H）.In addition,considering that the high time cost,the application of CFD model in emergency diffusion simulation was limited.This thesis attempted to use the optimized Gaussian diffusion model for diffusion simulation research,and put forward three different optimization schemes,namely weight optimization（improving the correlation between simulation and observation）,parameter optimization（improving the accuracy of simulation results）and weight+parameter optimization.Although CFD model is more accurate in pollutant diffusion simulation,its application in emergency diffusion simulation of sudden accidents is limited due to its high running time and cost.In order to solve this problem,this study also counducted the simulation based the optimized Gaussian diffusion model.Three different optimization schemes were proposed,namely weight optimization（improving the correlation between simulation and observation）,parameter optimization（improving the accuracy of simulation results）and weight+parameter coupled optimization.The results showed that the the weight+parameter coupling optimization scheme showed the best performance improvement compared with the default parameters:the error decreased by approximately 50%～82%,and the correlation increased by approximately 0.18～0.36.Compared with the CFD model,the accuracy of the optimized Gaussian model（error 73.2%）was lower than that of the CFD model（error 58.7%）,but within the acceptable range.This indicated that it can replace the CFD model and be applied in the emergency diffusion simulation of accidental pollutants.Based on the experimental data and the accurate diffusion simulation technology on the complex underlying surface,comparing with the actual accident scenarios,including the scenarios with unknown source strength and known location（inversion of Q only）,the scenarios with unknown source strength and location（inversion of Q,x,y,z）,three source parameters estimation technologies were established:（1）source strength inversion technology based on CFD-transfer coefficient method,（2）source strength inversion technology based on response surface-transfer coefficient method,and（3）source parameter inversion technology based on optimized Gaussian model-intelligent optimization method.The inversion results of source intensity based on CFD-transfer coefficient method showed that when the selected point was within 10～20°on both sides of the dominant wind direction and within 150～200m from the release source,the inversion error of source intensity was the smallest（～30%）.The estimation results of source intensity based on response surface-transfer coefficient method were underestimated,and the error was 25%.It showed that the transfer coefficient response relationship which was established based on response surface method could obtain more accurate estimation results of source intensity by using conveniently observed parameters,which could replace CFD-transfer coefficient method to estimate source intensity and reduce time cost.Compared with the first two methods,in addition to obtaining the source intensity inversely,the source parameters back calculation technology based on optimized Gaussian model intelligent optimization method could identify the location of emission sources,and the accuracy of inversion was high,the inversion errors of Q₀,x₀,y₀ and z₀ were 7%～16%,2.8m,9m and 12.9m.However,the inversion stability of source parameters based on optimized Gaussian model-intelligent optimization method was poor.Multiple inversions might be requirement to reduce the error caused by instability.A"monitoring-traceability-simulation"feedback technology system for emergency response to sudden accidents was established.A field simulation experiment that can reflect the dynamic accident evolution process with different sources intensity was carried out to simulate the evolution process of the accident.The simulation experiment simplified the accident evolution process into three stages,namely the initial stage（2kg/s）,the middle stage（10kg/s）and the end stage（0.02kg/s）.Combined with the experiment,the application and verification of the feedback technology system was carried out.The results showed that the"monitoring-traceability-simulation"technical system could realize dynamic,rapid source traceability and pollutant diffusion simulation of sudden air pollution accidents on complex underlying surfaces.In accident response with unknown source strength only,the traceability technology based on response surface transfer coefficient and the pollutant diffusion simulation technology based on optimized Gaussian model had the best performance.The source intensity inversion error was 2%～29%,the diffusion simulation error was 111%,which was 154%higher than the accuracy of unoptimized Gaussian model.The correlation between simulation results and observation results was 0.68,which was not significant difference with the unoptimized Gaussian model,and the response time was about 2s.In the case of unknown source intensity and location,it was suggested to give priority to the source tracing technology based on optimized Gaussian model intelligent optimization method and pollutant diffusion simulation technology based on optimized Gaussian model.The source intensity inversion error was 11%～77%,and the x₀,y₀and z₀ inversion errors were 2.2～31.1m,2.5～70.0m and 2.0～14.1m respectively.The simulation result error was 83%,which was 275%higher than the accuracy of the unoptimized Gaussian model.The correlation between the simulation result and the observation result was 0.55,which was similar with the unoptimized Gaussian model,and the average operation time was about 2.7s.Based on the above technologies and verification results,the process of simulation and traceability technology that can be popularized on another complex underlying surface was further proposed.