Study On The Longitudinal Dispersion Coefficients In Natural Rivers

To understand the scalar transport in the solvent is of importance in chemical engineering,pollute control,water remediation.This thesis concentrate on the pollute transport in natural rivers,studying the longitudinal dispersion coefficient in the natural straight rivers.The first chapter will show the background and current study on the longitudinal dispersion.The second chapter gives a theoretical solution for the longitudinal dispersion coefficient in the natural rivers.The triple integral form for the longitudinal dispersion coefficient caused by the velocity gradient and the depth-averaged longitudinal velocity distribution in the width direction are used to obtain the longitudinal dispersion coefficient in the flume.By transforming the non-integral expression for the velocity distribution into a trigonometric function series via Fourier transformation and substituting this series into the triple integral formula,we obtain the longitudinal dispersion coefficient.We then obtain a dimensionless formula for longitudinal dispersion in the flume via regression analysis and the result is consistent with the experimental results given in the previous study and this thesis.By analyzing the field data of the longitudinal dispersion coefficient in the natural river and the corresponding values calculated from the formula for the smooth flume of the same hydraulic parameters,we derive a formula for the longitudinal dispersion coefficient in natural rivers.The formula proves a suitable measure for estimating pollutetransport in natural rivers.The third chapter gives a physically sound formula for the longitudinal dispersion coefficient in the natural river.Most of the previous studies on the longitudinal dispersion coefficient in natural rivers are based on a pre-fixed formula,and the formula was obtained through optimizing the parameter.But this pre-fixed the formula lacks physical explanation.In this chapter,a canonical form for longitudinal dispersion coefficientsis proposed,and this form reflects the physics of longitudinal dispersion and works for complex flow conditions in natural rivers.This general form is more concise than the previous predictor.By using a genetic programming without pre-specifying correlations among field data or pre-specifying the form of the formula,we obtain a predictor for longitudinal dispersion coefficients of natural rivers.This formula is physically sound(i.e.has the above-mentioned canonical form)and is commensurate to or better than the previous formulafor the longitudinal dispersion coefficient in natural rivers.Finally,a grey modelis also used to verify that the universal form is appropriate.In the fourth chapter we simulate the contaminate dispersion at the early stage based on a random walk particle tracking method.At the large time,the longitudinal distribution of the mean cross-sectional concentration was given,and there are many predictors for longitudinal dispersion in natural rivers.Studies on the early stage is an important case.At the early stage,the longitudinal distribution of the mean cross-sectional concentration is non-Gaussian and the concentration distribution is non-uniform.However,results in different literatures offer different timescales for the early stage,urging us to quantitatively study the timescales for the early stage.Using the random walk particle tracking method,we study the development of the longitudinal dispersion in the early stage for Pe=1,10,100.Finally,we obtain the timescales for the longitudinal dispersion coefficient,skewness andtransverse concentration uniformity to reach their asymptotic values.These timescales offer references for the study on the scalar transport at the early stage.In the fifth chapter,we simulate the longitudinal dispersion when the boundary is absorptive.We give the microscopic boundary condition from the macroscopic condition.In the sixth chapter,we give a general form for the Pick's law when the diffusion coefficient field is heterogeneous.In the seven chapter,we give a short summary of this thesis and the potential study on the scalar transport.