On Traceability Theories And Methods Of Sudden Water Pollution In Rivers And Canals

In the intense human activities, subject to equipment failure, operator error, vandalism, natural disasters and other factors, more and more sudden water pollution accidents happened, which brought a serious threat to people's lives and property safety, social stability and ecological health. To effectively prevent these accidents and reduce the loss caused by them, great attentions are paid to research these problems, such as monitoring, precautioning, tracing and emergency disposing to the sudden pollution accidents, in the developed countries. But so far there is still lack of a relatively complete and mature theories, methods and technology. The traceability theories and methods to sudden water pollution accidents in rivers and canals are researched based on present researchs, observations on tracer tests and some related models and approaches comprehensive integrating system engineering and environmental hydraulics. It makes great siginificance in theories and wide range of applications prospects to knowledge the statistical rule of the diffusion and mix between water flow and pollutants in rivers and canals, developing analytical models and methods to traceability system and improving the efficiency and technical level to sudden water pollution incidents. The main contents and results are as follows:(1) The current situation and the reasons of sudden water pollution accidents were described. The research status and development trends of traceability method to sudden water pollution accidents in rivers and canals were analyzed. And the main research contents and the general idea of this thesis were presented.(2)The mechanism of contaminant transformation and migration, including the insoluble, soluble and volatile contaminants, such as different types of transport and transformation processes in the rivers and canals was analyzed; the basic principles of traceability were analyzed based on definitions and characters of sudden water pollution accidents and the concept of traceability; the traceability techniques and methods to sudden pollution accidents were summarized.(3) One, two and three-dimensional simulation model coupling water flow and quality were constructed; the characteristics of hydrodynamic and water environment in rivers and canals were analyzed; the calculation process to these models was presented. And the reasonableness of the coupling simulation model was verified by taking some part of Yangtze River as a numerical example. (4) An optimization traceability model of sudden water pollution accidents in rivers and canals was presented based on the framework of to optimization methods. A new algorithm based on hybrid differential evolution-Nelder Mead Simplex was designed. Numerical results showed that, the new traceability method can effectively reduce the30%reduction in the number of iterations and0.51%of total average absolute error, ie the traceability method based on hybrid Differential Evolution-Nelder Mead Simplex has strong convergence and accuracy comparing to the traceability method based on the differential evolution algorithm.(5) The traceability techniques and methods based on Bayesian framework is proposed to deal with the uncertainty of sudden pollution accidents in rivers and canals. Numerical examples showed that the average relative errors of the results obtained by Bayesian inference under different scenarios and different observation error conditions are less than6%. The average relative error and the relative standard deviation based on the Bayeian inference with unsteady flow and the measurement error0.1are less2.81%and25.65%comparing to the simplex algorithm. Therefore, traceability method based on Bayesian inference has strong applicable, higher recognition accuracy and immunity to noise.(6)In order to improve the accuracy and speed of traceability in sudden water pollution accidents, a traceable method combing Bayesian inference and differential evolution algorithm is first designed. Numerical examples showed the new method can effectively reduce the3/4iterations and make an average relative error of the results less than4%. It is to say that the new traceability designed in this thesis has a faster convergence speed and higher precision comparing to Bayesian inference method.(7) The content and research results were summarized systematically, and some possible future research directions are extended.The innovations of this paper as follows:1)Based on the basic principle of traceability track, a new method based on hybrid Differential Evolution-Nelder Mead Simplex was proposed to research traceability problem of sudden water pollution accident in rivers and canals;2)With full consideration to the uncertainty of sudden water pollution incident in rivers and canals, a new tracking model was presented based on Bayesian inference;3)A new tracking method was proposed combing Bayesian inference and differential evolution algorithm. It can not only improve calculation accuracy and sampling efficiency, but also can deal with uncertainties.