Inverse CFD Simulations Of Life Information Identification In Incompact Ruins Formed After The Earthquake

The searching and rescue of the trapped people after the earthquake has been attracting much attention.If the location of the trapped people can be determined in time after the earthquake,the rescue can be carried out quickly,which is of great significance to saving the life safety of the trapped people.In order to search and locate the trapped people under the loose ruins formed after the earthquake,this research takes the trapped people as the heat source and carbon dioxide release source in the same location,and carries out innovative work,such as numerical research on flow heat transfer and carbon dioxide diffusion transport in the simplified porous ruins model and inversion traceability.Firstly,the loose ruins were simplified as porous media and solid phase coupled physical model,based on the finite volume method of dimensionless Darcy-Brinkman momentum and energy equations of the discrete solution,this research analyzes effects of the Reynolds number（5×10²≤Re≤10⁴）,Darcy number(10^-3≤Da≤10^-2)and Grashof number（10⁵≤Gr≤10⁷）,on the structural transformations of fluid flow and heat transfer process,then explores the people trapped in exhaled carbon dioxide under different initial conditions of the spread of the law of transport.Secondly,an approximate inverse time inversion method based on the convective downwind scheme discretization and diffusion term constraint is proposed,and the corresponding basic theory,implementation logic and computer program are constructed to carry out the inverse time inversion research of the carbon dioxide migration history and the location of the release source.The influences of time step,external wind speed,permeability of porous ruins,thermal buoyancy,number of sources and location spacing on inverse time inversion are analyzed in detail.The results show that the new inversion method can effectively identify the true location of the source of carbon dioxide（trapped people）.Then,based on the monitoring information of a limited number of CO2 sensors,the standard reverse position probability of sensor alarmig monitors,the conditional position probability of sensor current concentration monitoring and the conditional position probability inversion program of historical concentration monitoring are constructed respectively.For people trapped in the different positions in the ruins,a source location identification was carried out with probability inversion program.The results show that the identification accuracy is low only using carbon dioxide sensor alarming information to calculate the standard location probability distribution and the joint probability distribution,which can only give a rough area of the source location.However,the probability distribution of the conditional position obtained after comprehensive processing of concentration monitored by single or multiple sensors is extremely concentrated and very close to the real source location,which significantly improves the accuracy of inverse identification.