Font Size: a A A

Offshore Pollution Source Localization Based On WSN In Two-dimensional And Three-dimensional Space

Posted on:2019-02-23Degree:MasterType:Thesis
Country:ChinaCandidate:Z FengFull Text:PDF
GTID:2348330548451564Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
Frequent water pollution accidents have seriously affected human production and life,so real-time monitoring and pollution source localization in water environment are important.In this paper,localization algorithms of water pollution source are proposed to solve the pollution source localization in static horizontal water plane and three-dimensional underwater space.The effectiveness of the proposed algorithms are verified by simulation experiments.The main contents of this paper are as follows:Firstly,in order to solve the problem of the pollution source localization in the static water horizontal plane,a centroid localization algorithm based on the concentration field with equipotential lines is proposed.The algorithm can solve the pollution source localization problem without analytical diffusion model effectively.Experimental results show that the best positioning performance can be obtained in the positioning algorithm of the centroid location algorithm based on the concentration field with equipotential lines compared with the coarse positioning algorithm and the analytical algorithm based on diffusion models.Secondly,to solve the problem of pollution source localization in the nearshore three-dimensional static underwater space,an unscented Kalman filter algorithm based on spatio-temporal sampling data is proposed.The accuracy between the unscented Kalman filter algorithm based on spatio-temporal sampling data and the unscented Kalman filter method based on time-domain sampling data is compared in Matlab.Experimental results show that the unscented Kalman filter based on spatio-temporal sampling data has better positioning accuracy,the number of sensor nodes affects the positioning accuracy of the two methods,and the unscented Kalman filter method based on spatio-temporal sampling data has the better positioning performance and more positioning stability when the sensor nodes are deployed densely.Finally,based on hydrological simulation software MODFLOW,the diffusion process of two-dimensional planar water pollution source is simulated.In the simulation the concentration data related to water pollution source is generated,and concentration fields with circle equipotential line and symmetry equipotential line are analyzed.An important basis for the research about the source localization method which is independent of diffusion model is provided through the simulation.Furthermore,the concentration diffusion model of the pollution source in nearshore three-dimensional underwater space is analyzed.Based on the hydraulic modeling software platform DHI MIKE Zero,the diffusion data of the pollution source and the concentration field are simulated.The time-space sampling model is used to sample the simulated concentration data in time and space,and reduces sample size and decreases the redundant data information.
Keywords/Search Tags:pollution source positioning, concentration field with equipotential lines, three-dimensional underwater space, time-space sampling model
PDF Full Text Request
Related items