外文翻譯---旋轉(zhuǎn)回?zé)崞骼碚撆c熱力發(fā)電廠的一臺(tái)空氣預(yù)熱器實(shí)驗(yàn)結(jié)果的比較（譯文）

1、A comparison of rotary regenerator theory and experimental results for an air preheater for a thermal power plantTeodor Skiepko a,*, Ramesh K. Shah ba Bialystok Technical University, 15-351 Bialystok, Poland b Department

2、 of Mechanical Engineering, Rochester Institute of Technology, Rochester, NY 14623-5604, USAAbstractThe aim of the paper is to compare results obtained based on theoretical modeling with directly measured experimental da

3、ta on a full scale operating air preheater. First, the model of rotary regenerator energy transport involving longitudinal matrix heat con- duction is formulated in the paper. Then a solution of the model equation system

4、 is presented with reference to authors? former papers. The results representing temperature distributions of heat exchanging gases and continuously rotating matrix are illustrated by means of 3D charts. For the rotary a

5、ir preheater of 5.3 m diameter, the temperature distributions computed are compared with experimental data. Right trends and a fair agreement between theory and experiments are found. Finally, the computed and ex- perime

6、ntal regenerator heat transfer effectivenesses are compared and found to be within ?3% agreement at about 88% regenerator effectiveness. ? 2003 Published by Elsevier Inc.1. IntroductionAn important characteristic of a ro

7、tary heat ex- changer is to combine compactness with high perfor- mance. Such a heat exchanger has a round disk matrix core having a large amount of heat transfer surface. The disk rotates continuously with a constant fr

8、action of the core submerged into the hot fluid stream and the re- maining fraction into the cold fluid stream. In turn, when the hot gas penetrates into flow passages of the core, the energy is stored in the matrix. Con

9、versely, during flow of the cold gas through the same passages, the matrix core delivers the energy to the cold fluid. Because heat is not transferred across a wall separating the fluids but heat is stored and rejected a

10、lternately by each matrix element, the rotary heat exchangers are frequently referred to as rotary regenerators. Analysis of heat transfer in regenerators was initiated and developed in Germany during 1910–1930s. Prior t

11、o 1948, all the models were based on a set of classical idealizations as listed by Shah [1]. Mondt [2] has pro- vided additional references on developments in themodeling with longitudinal conduction effects included in

12、the analysis since 1948. Bahnke and Howard [3] analyzed numerically the rotary regenerator problem in order to determine the regenerator heat transfer effectiveness with longitudinal conduction effect included. They show

13、ed that the effec- tiveness depends upon NTU0; C?; ðaAÞ?; A? s; k and pro- vided the results for a wide range of parameters: 16NTU06100; 0:96C?61;0:256ðaAÞ? ¼A? s 61; 0:016 k60:32. Bahnke and How

14、ard?s results for a limited in- dustrially useful range were correlated by Shah [4] in an algebraic formula. In a series of papers, Skiepko [5–7] derived an analytical solution for the rotary regenerator problem includin

15、g the longitudinal conduction effect in the matrix wall. Based on the solution, he evaluated the effect of dimensionless regenerator model parameters on the gas and matrix temperature distributions and showed 3D temperat

16、ure distribution graphs. Romie [8] expressed the effect of longitudinal wall heat conduction on the regenerator effectiveness by means of two auxil- iary factors. More recently, Shah and Skiepko [9] pro- vided data on th

17、e regenerator effectiveness for cases in which ðaAÞ? 6¼A? s. Concluding, all the aforementioned results are based on theoretical modeling. No compari- sons of model results with experimental data are avail

18、- able in the open literature. The objective of the paper is* Corresponding author. E-mail address: tskiepko@cksr.ac.bialystok.pl (T. Skiepko).0894-1777/$ - see front matter ? 2003 Published by Elsevier Inc. doi:10.1016/

19、S0894-1777(03)00048-7Experimental Thermal and Fluid Science 28 (2004) 257–264www.elsevier.com/locate/etfs? gas temperaturesh1ð0 6 u1 6 1; z ¼ 0Þ ¼ 1 ð3Þh2ð0 6 u2 6 1; z ¼ 0Þ &

20、#188; 0 ð4Þ? coupling of matrix temperatures#1ðu1 ¼ 0; zÞ ¼ #2ðu2 ¼ 1; 1 ? zÞ ð5Þ#1ðu1 ¼ 1; zÞ ¼ #2ðu2 ¼ 0; 1 ? zÞ ð6Þ?

21、assumed adiabaticity of matrix surface faceso#j oz? ? ? ? uj;z¼0 ¼ 0 and o#j oz? ? ? ? uj; z¼1 ¼ 0; j ¼ 1; 2ð7ÞThe problem formulation above is identical to that investigated by Bahnke

22、and Howard [3]. They solved the problem by a numerical analysis.3. Method of solution and resultsEqs. (1)–(7) have been solved analytically by a method developed by Skiepko [5,6]. The final formulas determining gas and m

23、atrix temperature distributions within hot and cold gas flow zones are available in in- finite series forms [6] and are not presented here due to space limitations. The coefficients of the infinite series are determined

24、numerically by solving a set of linear equations obtained from truncating the infinite series to some finite number of terms and using the collocation method for boundary conditions of Eqs. (5) and (6).In Figs. 2 and 3,

25、gas and matrix temperature distri- butions are shown from the solution of Eqs. (1)–(7). Notice that the gas and matrix temperature fields are markedly nonlinear either in the direction of gas flows and in matrix rotation

26、. The effect of longitudinal matrix heat conduction is demonstrated using three values of kj parameter, i.e. k1 ¼ k2 ¼ 0 when no conduction effect is included, and k1 ¼ k2 ¼ 0:01 and 0.2. The trends i

27、n temperature variations as functions of coordinates show that heat conduction may affect essentially the temper- ature fields of both gases and matrix when k1 and k2 values are higher than about 0.01. For more detailed

28、analysis of the conduction effect on gas and matrix temperature distributions, see Skiepko [6,7]. The model described by Eqs. (1)–(7) is based on a set of idealizations that make the problem easily manage- able. However,

29、 the inaccuracy incurred by the idealiza- tions in dependent temperature distributions can be evaluated only by comparisons with experimental data.4. Experimental exchanger measurementsWe will now present a quantitative

30、estimate of the modeling inaccuracy by comparing results of the mod- eling with experimental data for regenerator outlet hot and cold gas temperature distributions and the regen- erator effectiveness. The details on the

31、experimental rotary regenerator have been presented by Skiepko [10], thus only a sum- mary is given here. Experiments described here were performed on a rotary heat exchanger (an air preheater) for a pulverizedFig. 2. Ga

32、s temperature distributions in a rotary heat exchanger de- termined based on model solution [6] for: NTUj ¼ 8; C? r;j ¼ 2, kj ¼ 0; 0:01 and 0.2, j ¼ 1; 2.Fig. 3. Matrix temperature distribution in a r

33、otary heat exchanger determined based on model solution [6] for: NTUj ¼ 8, C? r;j ¼ 2, kj ¼ 0; 0:01 and 0.2, j ¼ 1; 2.T. Skiepko, R.K. Shah / Experimental Thermal and Fluid Science 28 (2004) 257–264 2