Streams, intermediate

In some cases the chemical manufacturer purifies a portion of this intermediate stream to make a high purity product. In other cases, the chemical manufacturer sells a low purity product to a gas company and the gas company purifies it to make a high purity product. In both bases, purification is done on a continuous basis, rather than cylinder by cylinder. The purification processes tend to utilize standard methods. 

The typical SEA process uses a manganese catalyst with a potassium promoter (for solubilization) in a batch reactor. A manganese catalyst increases the relative rate of attack on carbonyl intermediates. Low conversions are followed by recovery and recycle of complex intermediate streams. Acid recovery and purification involve extraction with caustic and heat treatment to further decrease small amounts of impurities (particularly carbonyls). The fatty acids are recovered by freeing with sulfuric acid and, hence, sodium sulfate is a by-product. 

All reactor modes can sometimes be advantageously operated with recychng of part of the product or intermediate streams. Heated or cooled recycle streams serve to moderate undesirable temperature travels, and they can be processed for changes in composition before being returned. 

Review of process chemistiy, including reac tions, side reactions, heat of reaction, potential pressure buildup, and characteristics of intermediate streams... 

In treating stream mixing, the mixing device was defined as part of the designated refinery operation itself. This approach was also undertaken by Zhang and Zhu (2006). Therefore the only mixers considered are where the final blending takes place. This approach distinguishes the contribution of each feedstock to the final product. With this type of formulation, all variables and attributes of intermediate streams will depend on the crude type. 

The intermediate material balances within and across the refineries can be expressed as shown in constraint (3.2). The coefficient acr,dr,i,P can assume either a positive sign if it is an input to a unit or a negative sign if it is an output from a unit. The multirefinery integration matrix dr y accounts for all possible alternatives of connecting intermediate streams dr CIR of crude cr CR from refinery ie I to process p P in plant i i. Variable xiRef. , represents the... 

In constraint (3.4) we convert the mass flow rate to volumetric flow rate by dividing it by the specific gravity SG( r( tr of each crude type cr CR and intermediate stream cir CB. This is done as some quality attributes blend only by volume in the products blending pools. 

Constraint (3.9) sets an upper bound on intermediate streams flow rates between the different refineries. The integer variable y pipeR fi represents the decision of exchanging intermediate products between the refineries and takes on the value of one if the commodity is transferred from plant i I to plant i C I or zero otherwise,... 

Constraints (5.13) and (5.14) represent the material balance that governs the operation of the petrochemical system. The variable x 1 represents the annual level of production of process m Mpa where ttcpm is the input-output coefficient matrix of material cp in process m Mpel. The petrochemical network receives its feed from potentially three main sources. These are, (i) refinery intermediate streams of an intermediate product cir RPI, (ii) refinery final products Ff ri of a final product cfr RPF, and (iii) non-refinery streams Fn px of a chemical cp NRF. For a given subset of chemicals cp CP, the proposed model selects the feed types, quantity and network configuration based on the final chemical and petrochemical lower and upper product demand Dpet and DPet for each cp CFP, respectively. In constraint (5.15), defining a binary variable yproc et for each process m Mpet is required for the process selection requirement as yproc et will equal 1 only if process m is selected or zero otherwise. Furthermore, if only process m is selected, its production level must be at least equal to the process minimum economic capacity B m for each m Mpet, where Ku is a valid upper... 

Table 7.6 shows the solution of the refineries network using the SAA scheme with N = 2000 and N = 20000 where the proposed model required 790CPUs to converge to the optimal solution. In addition to the master production plan devised for each refinery, the solution proposed the amounts of each intermediate stream to be exchanged between the different processes in the refineries. The formulation considered the uncertainty in the imported crude oil prices, petroleum product prices and demand. The three refineries collaborate to satisfy a given local market demand where the model provides the production and blending level targets for the individual sites. The annual production cost across the facilities was found to be 6 650 868. 

Furthermore, for values of 0i and 02 exceeding 100, the model did not recommend exchange of intermediate streams between the refineries due to the high risk associated with such investment. However, the values of both 0j and 02 are left to the decision maker s preference. 

Because the performance of a particular piece of equipment depends on its input, recycling of streams in a process introduces temporarily unknown, intermediate streams whose amounts, compositions, and properties must be found by calculation. For a plant with dozens or hundreds of streams the resulting mathematical problem is formidable and has led to the development of many computer algorithms for its solution, some of them making quite rough approximations, others more nearly exact. Usually the problem is solved more easily if the performance of the equipment is specified in advance and its size is found after the balances are completed. If the equipment is existing or must be limited in size, the balancing process will require simultaneous evaluation of its performance and consequently is a much more involved operation, but one which can be handled by computer when necessary. 

If the HEN has more than the minimum number of exchangers (say nv more than the minimum) and nT variable target temperatures, then nv + nT of the intermediate stream temperatures and heater loads can be chosen as control variables. Stream split fractions are always available as control variables. These variables are adjusted to try to make the HEN feasible for the assumed, fixed values of the uncertain supply temperatures and flow rates. The HEN is feasible if and only if s 0. 

To allow algebraic equations to be used to locate ATm, assume that the heat capacities can be approximated by piecewise constant functions of temperature, with discontinuities at temperature breakpoints TBRj. Then for each exchanger, Arm can occur only at either end or at a breakpoint location inside the exchanger. However, a remaining difficulty is that since the intermediate stream temperatures are not known before the resilience test, the breakpoint locations are also not known a priori. 

In a minimum unit HEN the intermediate stream temperatures and heater loads, and thus the breakpoint locations, are uniquely determined by the energy balance and energy recovery constraints. Thus for given supply temperatures and flow rates, the Arm violations (and surpluses) and load violations (and surpluses) in each exchanger k are also uniquely determined. 

Note that the energy recovery constraint reduces to 0 = 0 for this network since there are no heaters. The energy balance constraints for the two exchangers can be solved for intermediate stream temperatures T5 and... 

---