SYNTHESIS OF MULTIPLE EFFECT EVAPORATOR SYSTEM

Abstract

Asimple mathematical model based on stream analysis utilizing the principles of Process Integration (PI) has been developed to analyze asystem of multiple effect evaporators (MEE) and to screen the best flow sequence which will offer best steam economy (SE) for the system. So far, the above work was purely under the domain of complex mathematical models. The present model has been applied to a number of MEE systems including a septuple effect flat falling film evaporator (SEFFFE) system under different operating configurations such as different flow sequences, steam splitting, vapor bleeding and feed, product and condensate flashing. This system has been employed in anearby Paper Mill to concentrate black liquor. The results of the present simplified model have been compared with the mathematical model developed by Bhargava (2004) for the above SEFFFE system. It has been observed that the results of the present model and that of Bhargava (2004) compare well under the range of operating variables generally encountered in Indian Paper Mills. Further, to test the effectiveness and reliability of the present model, its results have been compared with models of other investigators such as Kern (1950), Nishitani and Kunugita (1979), Holland (1975), Lambert et. al. (1987), Zain and Kumar (1996) andRaghu Ram and Banerjee (2003). The results of the present model also compare well with the above models. The present model has been deve.oped to serve three pnrpose, The firs, pnrpose is to screen out the best possib.e flow sequence ofaMEE system working under different operatmg configuration md under acondition when input parameters vary considerably. The second purpose is to use i, as a simulation tool to predict the input-output relationships. The third purpose is to use the model to improve the existing design of the SEFFFE system by improving its steam economy. As the present model is based on stream analysis it offers better insight into the problem by employing the concepts of PI and temperature path traced by each stream and suggests three different methods, which are based on steam economy, steam consumption and temperature path method, for the determination of optimum flow sequence. Based on the above three methods the optimum flow sequences predicted by the present model under different operating configurations match completely with that of the prediction of Bhargava (2004), Nishitani and Kunugita (1979) and Kern (1950). As a simulation tool for the prediction of input-output relationships the present model also compares well with that of the models of Bhargava (2004), Nishitani and Kunugita (1979), Lambert et. al. (1987), Kern (1950), Holland (1975), Zain and Kumar (1996) and Raghu Ram and Banerjee (2003). In fact, the models of above investigators have been used for different MEE systems in which number of effects vary from three to seven and have been employed to concentrate different fluids such as chemical solution, milk, caustic soda, black liquor and sugar. The input parameters in the above cited work also vary to a large extent. It is observed that the present model predictions for steam consumption and that of Bhargava (2004), Nishitani and Kunugita (1979), Lambert et. al. (1987), Kern (1950), Holland (1975), Zain and Kumar (1996) and Raghu Ram and Banerjee (2003) differ by 3.5 %, 2.2 %, 7.2 %, 5.9 %, 3.14 %, 9.7 % and 3.8 %, respectively. n When the present model is used as asimulation tool to improve the existing design of the SEFFFE system employed in anearby Paper Mill to concentrate weak black liquor many interesting results were obtained. In fact, the temperature path analysis used in the present model suggested that the use ofre-heaters will benefit the system. Adetailed analysis shows that inclusion offive re-heaters in backward flow sequence which happens to be the best flow sequence for the SEFFFE system can improve the steam economy by 4.7 %. Afurther analysis of the above system by the present model shows that for all flow sequences investigated, the average %contributions, in terms of energy, towards total evaporation in decreasing order are 74.6 %, 18.4 %, 4.3 %, 2.17%and 0.51 %for vapor generated in first to sixth effect, live steam, secondary condensate flash vapor, primary condensate flash vapor and vapor from product flash, respectively. An analysis shows that the energy contribution (towards total evaporation) of each flash tank is different, the maximum being 1.84 %whereas, the minimum being 0.42%which clearly shows that these do not contribute equally towards total evaporation.