Endüstriyel kaynaklarda hüzme yükselmesi

Abstract

ÖZET Endüstriyel kaynaklar düşük ısıl kapasite, kısa bacalar, küçük de şarj hızları ve yerleşim bölgeleri içinde yer almaları gibi özellikle ri dolayısı ile bu bölgelerdeki hava kalitesinin korunmasında birinci önceliğe sahip kaynaklardır. Bu tür kaynaklar için EPA'nm geliştir diği hava kirliliği modellerinde Briggs'in hüzme yükselmesi eşitlikle ri kullanılmaktadır. Bu ifade büyük nokta kaynaklar için kalibre edildiği için endüstriyel kaynaklara uygulanmasında önemli hatalara sebebiyet vermektedir. Yaptığımız çalışmada bu husus incelenmiş ve endüstriyel kaynaklarda huzme yükselmesini veren genel ifadeler çıka rılmıştır. Birinci bölümde Dünya enerji tüketimindeki gelişmeler ve ülkemizde fosil yakıt kullanımından oluşan kirleticiler incelenmiş, çalışmanın amacı ve kapsamı verilmiştir. İkinci bölümde huzme tanımlanmış ve huzme yükselmesine etki eden meteorolojik faktörler ve kaynak özellikleri üzerinde durulmuştur. Üçüncü bölümde huzme yükselmesi teorileri incelenmiştir. Yoğunluk farklarının etken olduğu huzmeler için `2/3 Kanunu 'nun çıkarılışı ve rilmiştir. Daha sonra konuyla ilgili literatür incelemesi yer al maktadır. Bu incelemeler kaynaktan uzaklığa bağlı olarak huzme yö rüngesini veren ifadelerle nihai hüzme yükselmesini veren ifadeler olmak üzere iki bölümde yer almıştır. Dördüncü bölümde, deneysel çalışmanın planlanması, çalışmanın yürü tüldüğü tesisin tanıtımı, kullanılan ölçüm tekniği ve meteorolojik parametrelerin ölçümü üzerinde durulmuştur. Ölçüm sonuçları ve lite ratürden derlenen datalar geri yıkama potansiyelleri gözönüne alınarak üç ayrı kategoride incelenmiş ve regresyon doğruları bulunmuştur. Beşinci bölümde gözlem sonuçları çeşitli yönleri ile değerlendiril miş ve genel ifadeler çıkarılmıştır. Bulunan ifadeler ile Briggs eşit liği arasındaki farkın nedenleri üzerinde durulmuştur. Altıncı bölümde, bulunan neticeler toplu halde özetlenmiştir. xıv

SUMMARY PLUME RISE IN INDUSTRIAL SOURCES Air quality in any region is affected by air polluting sources, which are defined as point, line and area sources. Point sources play an important role on air quality due to their variety, polluting potential and continuity. Therefore, control of point sources is an important step in preventing air pollution. Point sources are divided into two groups according to the, _` buoyancy flux (F). These are;. thermal power plants (F>100 m.s ) and industrial sources (F<100 m~~.s ). Because of the fact that industrial sources are generally located within or close to residential areas, they are significant emission sources which affect the air quality in these areas. Mathematical models are important tools to develop the relation between the polluting source and air quality. Several models have been developed for this purpose. Gaussian dispersion models are accepted as the basis for modelling point sources due to their simplicity and dependability and are therefore used in models like CRESTER, RAM, CDM and VALLEY which are applied by Environmental Protection Agency of the U.S. The models can be used for estimating pollutant concentrations and the extent of source control required to attain a certain air quality standard. The selection of the stack heights of the point sources can be carried out through these models. However application of these models require the definition and determination of numerous parameters. The fundamental equation of the Gaussian model for point sources is given below. C(x,y,z) = Q exp(-^- ) {exp(Z~^ 1+ exp l^f^D (1) 27ruö o 2<r 2<r 2 Q y z y z z -3 Where C is the concentration of the pollutant (yg.m ),Q is the source emission rate (g.s ),u is the horizontal wind speed at stack height, 0 and °` are horizontal and vertical standard deviations (m), and H y 2 v ' ' xvis the effective stack height (m). The effective stack height can be expressed as; H = hs + Ah (2) where hs is the physical stack height (m), and Ah is the plume rise (m). Five parameters (wind speed, horizontal and vertical dispersion coefficients, emission rate and plume rise) are needed to estimate ground-level concentrations. Relative errors in peredicting or measuring wind speed, emission rate and horizontal dispersion cause the same order of relative error in concentration; whereas relative errors in plume rise or vertical dispersion estimates can result in relative errors as the square of the ratio of effective stack height to vertical dispersion. Therefore plume rise is an essential para meter for obtaining accurate estimates of ground-level concentrations. Several models have been developed for the estimation of plume rise (Ah). These models make use of the factors effecting the plume rise, and therefore they are mostly empirical or semiempirical nature. The dispersionmo.delsof Elft for plume rise have used `2/3 Law`. This ' law is expressed as; A., 3,1/3 `l/3 -1 2/3 f. Ah = ( 7j-).F.u.x (3) 2 3/- Where 3 is the entrainment coefficient and x is the downwind distance from the source (m). Buoyancy flux (F) is given below. VTA F = g.wQ.r (-V5-) W s s ıs Where g is the gravitational acceleration.(m.s ), w is the stack gas exit velocity (m.s 1), Ts is the stack gas exit temperature ( K), and T is the ambient air temperature (°K). Equation (3) may be derived from analytical solutions of conservation equations of fluid mechanics. This solution is obtained by introducing an entrainment hypothesis into the one-dimensional, or integral form of the equations describing conservation of mass, momentum, and heat for a plume element. Two additional simplifications are required, however, before an analytical solution can be derived; the Boussinesq approximation and bend-over plume assumption. The former approximation assumes that ambient and plume densities are equal except in density difference terms; the latter approximation assumes that the plume centerline velocity is approximately equal to the mean wind speed or in other words, that the plume takes on the average horizontal wind speed immediately upon entering the atmosphere. o 1/3 Briggs has proposed a value of 1.6 for the coefficient (.-=-?*) in 2R the 2/3 Law following several field studies for thermal power plants. However, the ground-level pollutant concentration in industrial sources are underestimated when the Briggs' s coefficient is used in the` 2/3 Law`. Therefore value of this coefficient must be determined for industrial sources. xvi.The purpose of this study is to determine the value of the coefficient ( 3 )l/3 more accurately for industrial point sources by 2jT2r conducting field experiments and by using data from literature. Thus a more realistic approach would be obtained in modelling of these sources. In the first chapter world energy consumption and resulting air pollution problems are delineated, and purpose and objectives of the study are given. In the second chapter, description of the plume is made, the effect of plume rise on the ground level concentration is investigated, and the factors.affecting the plume rise are given. These factors are studied under two categories as meteorologgical parameters, and factors reflecting characteristics of the source. In the third chapter, the theory of plume characteristics is illustrated. The theory of the bent over plumes in neutral atmospheric conditions is given, and the `2/3 low` is derived. Literature survey related to the plume rise is given. The equations defining the initial and the ultimate plume rises are derived separately. In the fourth chapter, the experimental methods employed in this study are explained. The plant, where the experimental work was made is described, and the topogrofical and meteorological characteristics of the region of the study are given. The measurement techniques for the plume rise are investigated and the method used is described in detail. The results of the measurements are grouped into three categories according to the downwash potential. The cases studied are w/u>1.5 (regligible downwash), 1.0<w/u<1.5 (medium downwash) and w/u<1.0 (significant downwash). These are also applied to the data related to plume rise of industrial sources obtained from the literature. In the fifth chapter, the results are evaluted. Both the plume rise data obtained from the field measurements and those obtained from the literature are analysed with respect to downwind distance, wind profile and buoyancy flux F. In the sixth chapter, the discussions of the results and conclusions are given. The results are drawn for n eutral and unstable atmospheric conditions can be summarized as follows. 1. `2/3 Law` overestimates the plume rise for all downwash categories. 2. As downwash becomes significant the scatter of the data increases, and correlation between theoretical predictions and experimental results becomes poor. The experimental results give negative correlation for w/u <1.0. This indicates a significant influence of downwash on the plume. 3. For w/u >1.5, the regression line gives 15-25% lower plume rise compared to the Briggs formula. This deviation becomes 25-30% for 1.0< w/u< 1.5, and approximately 50% for w/u < 1.0 xviiThe deviations from `2/3 Law` for the industrial sources can be explained as follows; 1. The wind speed used in the `2/3 Law` is the wind speed at stack height. Major point sources have tall stack heights (h,> 150m). In the upper atmospheric layers wind profile can be assumed constant, whereas it becomes significant in the layers 50-100m above the ground. Therefore in the industrial sources, wind speed at the effective stack height should be taken instead of at the physical stack height. 2. The entrainment coeficient,3, becomes greater than 0,6 which is taken in the `2/3 Law`. This is due to the effect of the wind velocity gradient. The entrainment of ambient air into the plume increases due to the effect of the wind shear, and as a result plume dilution increases, the difference in densities decreases and plume rise decreas. 3. In industrial sources, stack gas exit velocity is generally lower than 10 m.s--'-. The wakes formed by the stack and the nearby buildings, result in downwash due to low stack gas exit velocity. 4. Due to strong atmospheric turbulence as a results of the topografical and meteorological characteristic of the study area, the density differentials between the plume and the ambient air are weakened, and lower rises are obtained. The effective plume rise parameter in the Gaussian dispersion model affects the ground-level concentration significantly. Therefore in general the lowest estimation of the plume rises is assumed to ensure safe ground-level concentration. In industrial sources these estimation must be accurate enough to predict the actually conditions. Accordingly, taking into account of the observed data and the data from the literature, the following relations are found to express the plume rise more accurately - For w/u 1.5 Ah = 1.3 F1/3 u-1.x2/3 (5) - For 1.0 w/u 1.5 Ah = 1.15 F1/3. u`1. x2/3 (6) - For w/u 1.0 Ah = 0,8 F1/3. u`1. x2/3 (7) The following general expressions are found from analysis of the data of SEKA-Bolu and Bringfelt; - For the data of SEKA-Bolu Ah = 1,1 F1/3. u`1. x2/3 Corr= 0,77 (8) XV111- For the data from Bringfelt Ah = 1,2 F1/3. u`1. x2/3 Corr= 0.66 (9) The general expression in obtained for combined data groups Ah = 1.15 F1/3. u`1. x2/3 (10) In this work, the value of this coefficient is determined as 1.15 instead of 1.6, and maximum concentration values were calculated using the former value. It was shown that this value results in a better approximation of the plume rise expression for industrial sources xix