A Tool for Optimal Synthesis of Industrial Refrigeration Systems : Application to an Olefins Plant.
暂无分享,去创建一个
Olefins Plants show an intrinsic cost of the energy demand which derives from Process and Plant design. Few innovative changes have been implemented in the past decades. Such changes were mainly directed to increasing the throughput of the plant and its flexibility rather than to provide reduction on energy costs per unit of product. Looking at Olefins Plants from a different sight, i.e. the innovation of processes configuration and technologies, they offer a wide temperature interval which ranges from 800-850 degrees Celsius at the coils outlet to minus 125-130 degrees in the cold-box; pressure may rise up to 30 bar; standard process productions quote 500.000 tons per year of ethylene. These figures put these plants in the ideal condition of being considered as an effective benchmark for all those research activities of approaching the energy efficiency problems by means of the adoption of new technologies or new process configurations. The process under study is a typical olefin plant that produces ethylene by thermal cracking of hydrocabons. This method proceeds by a free radical mechanism including initiation, propagation, and termination steps. The thermal cracking generates several valuable by-products including methane, propylene, butadiene, benzene,... and also fuel oil and tar. These products are usually separated by a set of distillations at low temperature (down to about 100 K) and high pressure in order to separate the products with an acceptable purity. For low pressure plants, the low temperature is achieved in a cold box using a complex refrigeration system with three cascaded refrigeration cycles using methane, ethylene and propylene. The methane refrigeration system works between 136 K and 180 K : it is a classical system with one evaporator and one condenser. The ethylene refrigeration system operates between 172 K and 243 K: it includes three pressure levels. The propylene cycle works between 233 K and 319 K and includes 4 pressure levels. These three systems are linked to each other through their condensers: methane condenser is cooled by ethylene shich in turn is cooled by propylene and finally, cooling water is used to condense propylene. The goal of the study was to develop a method to help in identifying the process modifications required to reduce the energy consumption of the cold box and to maximize the efficiency of the integrated refrigeration system. Energy integration techniques have been used to target the minimum energy requirement of the process and to characterize the refrigeration system required. If the heat cascade and the corresponding composite curves (e.g. figure 2) give an indication of the refrigeration requirement of the process including all the possible heat recovery exchangers to be considered in the system, it does not give a proper answer to the definition of refrigeration requirement target. This is because, the energy requirement is expressed in terms of mechanical power required in the refrigeration rather than in terms of thermal energy. There was therefore a need for a tool able to target as well the optimal integration of the refrigeration system. The retrofit and the optimal integration of refrigeration systems is a complex task due to a large number of degrees of freedom. To solve such problem, the combined use of process modelling, optimisation based energy integration techniques and graphical representations appear to be a very useful tool. The goal of this paper is to illustrate the application of such tools to retrofit the cold box of an olefin plant.
[1] Richard Turton,et al. Analysis, Synthesis and Design of Chemical Processes , 2002 .
[2] Bodo Linnhoff,et al. Heat and power networks in process design. Part I: Criteria for placement of heat engines and heat pumps in process networks , 1983 .
[3] Costas D. Maranas,et al. Optimal synthesis of refrigeration cycles and selection of refrigerants , 1999 .
[4] Ignacio E. Grossmann,et al. Optimal synthesis of integrated refrigeration systems—I: Mixed-integer programming model , 1986 .