Study On The Method Of Tracing And Locating Pollutants In River Pollution Events And Its Application In Small Mobile Monitoring Platform

Along with the rapid industrialization in China,inland rivers are facing different pollution problems.However,in many cases,environment protection association might miss the best opportunity to treat the incident due to the incapability of finding the pollution source.Therefore,it is of great significance to develop a rapid and flexible method of tracking and locating the pollution source.This thesis focuses on developing key techniques for portable water quality monitoring device,such as path planning,pollution source tracking and portable prototype system implementation.The work can be implemented for monitoring emergent water pollution,tracking possible pollution source and benefiting the social environment protection program.The main work and innovations of this thesis are as follows:(1)In order to achieve automatic path generation for portable water quality monitoring device,we have developed a novel path planning algorithm for irregular boundary and smart obstacle avoidance.Classical algorithms might be easily trapped by the concave boundary of the watercourse.To solve this problem,a global static path planning algorithm has been proposed,based on genetic algorithm and artificial potential field method.In addition,D-star dynamic programming has been studied specially for the possible watercourse jam.When facing new obstacles,D-star algorithm can make adjustment on global optimal path immediately.(2)We studied a novel water-pollution tracking algorithm based on joint behavior planning and concentration gradient method,aiming at dealing with the discrete measured values and small concentration gradients.Through analyzing different tracking algorithms,we propose a technical framework for tracking pollution source in watercourses.First of all,Markov-Monte-Carlo fast sampling method is applied on one-dimensional diffusion model to deduce the most probable area of the source.Secondly,dynamic programming,combined with behavior planning and concentration gradient,is employed to direct the motion of the portable device.Finally,two-dimensional model is applied for the detailed calculation,tracking algorithm analysis and data cross-validation.(3)In order to analyze and validate the efficiency of our path planning and source tracking algorithm,a system simulation platform is developed.Certain performance indicators,such as searching path length,tracking speed and locating accuracy,have been studied on sodium chloride solution diffusion experiment by multiple algorithms.(4)We developed a practical portable water-quality monitoring platform based on the above results and algorithms.Currently,both global static path planning and source tracking algorithm have been implemented on our portable water-quality monitoring device.It also has full functions of single-point/multi-point autonomous navigation.fix-point sampling,pollution source tracking and concentration contour map plotting.Overall.this thesis shows a prototype of portable water-quality detection and daily early warning system.Hopefully,it will provide valuable technical support for urban water-supply system and environment protection association.