| With the acceleration of urbanization in China,there are widespread problems of uneven water distribution and high water level in landfills with expanding volume,which have a serious impact on the stability of landfills.Therefore,it is of great practical significance to study the detection method of water distribution in landfills.Focusing on the problem of "seeking water" in landfills,this paper has carried out the design and development of landfills water distribution detection system,field test and data algorithm inversion.The work and results are as follows:1、Firstly,studied the knowledge related to resistivity method in electrical exploration,including the formation principle of human site electric field,basic theoretical formula of electrostatic field,common resistivity devices and sampling methods,numerical simulation method of resistivity forward and inverse problems.2、Designed and developed a landfill water distribution detection system,which was composed of signal output module,power amplifier module,electrode converter module,data acquisition module and upper computer control software.The detection system can output a stable excitation source to form a geoelectric field.With the electrode conversion module,electrical signals control relay switching to divide several electrodes into power supply electrode and sampling electrode.The sampling electrode can complete the collection of profile data of the landfill,and then master the water distribution area.3、Aiming at the data inversion imaging problem of landfill water distribution area,it constructed a forward simulation model,acquired a large number of simulation data,and studied the Back-Propagation Network(BP neural network for short),the inversion structure model and sample division,then it put forward the inversion method of landfill water distribution data based on BP neural network.4、Undertook the landfill landfill site in Harbin pile body water distribution data collection,verified the landfill water stability of the distributed detection system.The feasibility of using BP neural network to "seeking water" in landfill was verified by measured data.The results of field experiment showed that the BP neural network inversion algorithm has better practical effect on the measured data.With experimental analysis and field test,the detection system can detect the water distribution of landfill site stably in long term with certain application value.The inversion results of simulated data and measured data show that BP neural network has good inversion accuracy in the problem of "seeking water" in landfills,which can provide reference experience for other landfills exploration problems in the future. |