Once the problem is formulated mathematically, its solution is carried out through implementation of an optimization algorithm. Economic potential is maximized or cost is minimized (see App. A) in a structural and parameter optimization. Should an integer variable be optimized to zero, the corresponding feature is deleted from the structure and the structure is reduced in complexity. In effect, the discrete decision-making aspects of process design are replaced by a discrete/continuous optimization. Thus the initial structure in Fig. 1.7 is optimized to reduce the structure to the final design shown in Fig. 1.8. In Fig. 1.8, the membrane separator on the hydrogen feed has been removed by optimization, as have the isothermal reactor and many other features of the initial structure shown in Fig. 1.7.  [c.11]

Single reactions. In a single reaction such as Eq. (2.2) which produces a byproduct, there can be no influence on the relative amount of product and byproduct formed. Thus, with single reactions such as Eqs. (2.1) to (2.3), the goal is to minimize the reactor capital cost (which usually means minimizing reactor volume) for a given reactor conversion. Increasing the reactor conversion increases size and hence cost of the reactor but, as we shall see later.  [c.25]

If k-2 increases faster than kx, operate at low temperature (but beware of capital cost, since low temperature, although increasing selectivity, also increases reactor size). Here there is an economic tradeoff between decreasing byproduct formation and increasing capital cost.  [c.42]

The rate at which the catalyst is lost or degrades has a major influence on the design. If degradation is rapid, the catalyst needs to be regenerated or replaced on a continuous basis. In addition to the cost implications, there are also environmental implications, since the lost or degraded catalyst represents waste. While it is often possible to recover useful materials from degraded catalyst and to recycle those materials in the manufacture of new catalyst, this still inevitably creates waste, since the recovery of material can never be complete.  [c.49]

An initial guess for the reactor conversion is very difficult to make. A high conversion increases the concentration of monoethanolamine and increases the rates of the secondary reactions. As we shall see later, a low conversion has the effect of decreasing the reactor capital cost but increasing the capital cost of many other items of equipment in the flowsheet. Thus an initial value of 50 percent conversion is probably as good as a guess as can be made at this stage.  [c.51]

Even though choices of separators must be made at this stage in the design, it must be borne in mind that the assessment of separation processes ideally should be done in the context of the total system. As is discussed later, separators which use an input of heat to carry out the separation often can be run at effectively zero energy cost if they are appropriately heat integrated with the rest of the process. This includes the three most common types of separators, i.e., distillation columns, evaporators, and dryers. Although they are energy intensive, they also can be energy efficient in terms of the overall process if they are properly heat integrated (see Chaps. 14 and 15).  [c.76]

Both vacuum operation and the use of refrigeration incur capital and operating cost penalties and increase the complexity of the design. They should be avoided if possible. For a first pass through  [c.76]

Another variable that needs to be set for distillation is refiux ratio. For a stand-alone distillation column, there is a capital-energy tradeoff, as illustrated in Fig. 3.7. As the refiux ratio is increased from its minimum, the capital cost decreases initially as the number of plates reduces from infinity, but the utility costs increase as more reboiling and condensation are required (see Fig. 3.7). If the capital  [c.77]

When separating azeotropic mixtures, if possible, changes in the azeotropic composition with pressure should be exploited rather than using an extraneous mass-separating agent. When using an extraneous mass-separating agent, there are inevitably losses from the process. Even if these losses are not significant in terms of the cost of the material, they create environmental problems somewhere later in the design. As discussed in detail in Chap. 10, the best way to solve effluent problems is to deal with them at the source. The best way to solve the effluent problems caused by loss of the extraneous mass-separating agent is to eliminate it from the design. However, clearly in many instances practical difficulties and excessive cost might force its use. Occasionally, a component that already exists in the process can be used as the entrainer or solvent, thus avoiding the introduction of extraneous materials for azeotropic and extractive distillation.  [c.83]

Given these degrees of freedom, how can an initialization be made for the design The most significant degree of freedom is the choice of number of stages. If the evaporator is operated using hot and cold utility, as the number of stages is increased, a tradeoff might be expected, as shown in Fig. 3.14. Here, starting with a single stage, it has a low capital cost but requires a large energy cost. Increasing the stages to two decreases the energy cost in return for a small increase in capital cost, and the total cost decreases. However, as the stages are increased, the increase in capital cost at some point no longer compensates for the corresponding decrease in energy cost, and the total cost increases. Hence there is an optimal number of stages. However, no attempt should be made to carry out this optimization at this point, since the design is almost certain to change significantly when heat integration is considered later (see Chap. 15).  [c.87]

Figure 3.14 Variation of total cost with number of stages indicates that three stages is the optimal number for a stand-alone system in this case. Figure 3.14 Variation of total cost with number of stages indicates that three stages is the optimal number for a stand-alone system in this case.
If the evaporator design is considered against a background process and heat integration with the background process is possible, then very difierent designs can emerge. When making an initial choice of separator, a simple, low-capital-cost design of evaporator should be chosen.  [c.89]

If the total heat consumed is from an external utility (e.g., mains steam), then a high efficiency is desirable, even perhaps at the expense of a high capital cost. However, if the heat consumed is by recovery from elsewhere in the process, as is discussed in Chap. 15, then comparison on the basis of dryer efficiency becomes less meaningful.  [c.91]

Distillation is by far the most commonly used method for the separation of homogeneous fluid mixtures. The cost of distillation varies with operating pressure, which, in turn, is mainly determined by the molecular weight of the materials being separated. Its widespread use can be attributed to its ability to  [c.92]

Also, instead of using two separators, a purge can be used (see Fig. 4.2c). Using a purge saves the cost of a separator but incurs raw materials losses and possibly waste treatment and disposal costs.  [c.96]

The fourth option, shown in Fig. 4.4[c.100]

EP = value of products - raw materials cost = [176 X S X 0.45 + 36 X (2 - S) X 0.35]  [c.105]

In the case of a liquid recycle, the cost of this pressure increase is usually small. Pumps usually have low capital and operating costs relative to other plant items. On the other hand, to increase the pressure of material in the vapor phase for recycle requires a compressor. Compressors tend to have a high capital cost and large power requirements giving higher operating costs.  [c.115]

Rather than relying on heuristics which can be ambiguous or in conflict, a parameter would be preferred that can measure quantitatively the relative performance of different sequences. The vapor flow rate up the column is a good measure of both capital and operating costs. There is clearly a relationship between the heat duty required to run the distillation and the vapor rate, since the latent heat relates these two parameters. However, there is also a link between vapor rate and capital cost, since a high vapor rate leads to a large-diameter column. The high vapor rate also requires large reboilers and condensers. Thus vapor rate is a good measure of both capital and operating costs on individual columns. Consequently, sequences with a lower total vapor load would be preferred to those with a high total vapor load. But how is the total vapor load predicted  [c.135]

However, use of total vapor rate is still only a guide and might not give the correct rank order in some cases. In fact, given some computational aids, it is a practical proposition to size and cost all the alternative sequences using a shortcut sizing calculation, such as the Fenske-Gilliland-Underwood approach, together with cost correlations. Even though practical problems might involve a large number of components, it is rare for them to have more than six products, which means 42 possible sequences from Table 5.1. In addition, process constraints often reduce this number.  [c.142]

This result is important, since in practice we should not focus exclusively on the single sequence with the lowest overall cost. Rather, because of the uncertainties in the calculations and the fact that other factors need to be considered in a more detailed evaluation, the best few sequences should be evaluated in more detail. Thus there is no need to solve the separation sequence and heat integration problems simultaneously. Rather, decouple the two problems and tackle them separately, simplifying considerably the overall task. It must be emphasized strongly that the decoupling depends on the absence of significant constraints limiting the heat integration potential within the distillation sequence. For example, there may be limitations on the pressure of some of the columns due to product decomposition, etc. that limit the heat integration potential. In these  [c.143]

Having established that there is apparently a mechanism whereby the problems of sequencing and heat integration can be decoupled for simple columns on the basis of energy costs, it is interesting to consider whether there is any conflict with capital cost. A column sequence that handles a large amount of heat must have a high capital cost for two reasons  [c.146]

It is thus unlikely that a distillation sequence has a small capital cost and a large operating cost, or vice versa. A nonintegrated sequence with a large heat load has high vapor rates, large heat transfer areas, and large column diameters. Even if the heat requirements of this sequence are substantially reduced by integration, the large heat transfer areas and large diameters must still be  [c.146]

Thus capital cost considerations reinforce the argument that the nonintegrated sequence with the lowest heat load is that with the lowest total cost.  [c.147]

Complex column arrangements, such as the Petlyuk column, offer large potential savings in energy compared with sequences of simple columns. The dividing-wall also offers large potential savings in capital cost. However, it is recommended that complex column arrangements only be considered on a second pass through the  [c.156]

Using targets rather than design for the outer layers allows many design alternatives to be screened quickly and conveniently. Screening many design alternatives by complete designs is usually simply not practical in terms of the time and effort required. Using targets to suggest design changes works inward to the center of ttie onion and evolves the design for the inner layers. First, consider Hie details of how to set energy targets. Capital cost targets are considered in the  [c.159]

Figure 6.6 illustrates what happens to the cost of the system as the relative position of the composite curves is changed over a range of values of AT ir,. When the curves just touch, there is no driving force for heat transfer at one point in the process, which would require an  [c.165]

There is a tradeoff between energy and capital cost i.e., there is an economic degree of energy recovery. Chapter 7 explains how this tradeoff can be carried out using energy and capital cost targets.  [c.166]

Cost of imported electricity = (7 - 3.96) X 19.2 X 10  [c.200]

The energy cost of the process can be set without having to design the heat exchanger network and utility system. These energy targets cam be calculated directly from the material and energy balance. Thus  [c.210]

Equation (5.8) tends to predict vapor loads slightly higher than those predicted by the full multicomponent form of the Underwood equation. The important thing, however, is not the absolute value but the relative values of the alternative sequences. Porter and Momoh have demonstrated that the rank order of total vapor load follows the rank order of total cost.  [c.137]

Whatever the method used to screen possible sequences, it is important not to give exclusive attention to the one that appears to have the lowest vapor load or lowest total cost. There is often little to choose in this respect between the best few sequences, particularly when the number of possible sequences is large. Other considerations such as heat integration, safety, and so on also might have an important bearing on the final decision. Thus the screening of sequences should focus on the best few sequences rather th2in exclusively on the single best sequence.  [c.142]

In the study carried out by Stephanopoulos, Linnhoff, and Sophos, in each of three examples these workers found that the optimal nonintegrated sequence turned out to be the optimal integrated sequence in terms of minimum total cost (capital and operating). In addition, it was observed that higher heat loads tended to occur in sequences that also had wider spans in temperature between reboiler and condenser. These studies suggest that the two problems of separation sequencing and heat integration can, for all practical purposes, be decoupled. The result is illustrated in Fig. 5.6. The best nonintegrated sequence turns out to be among the best few integrated sequences in terms of total cost (capital and operating).  [c.143]

When separating a three-component mixture using simple columns, there are only two possible sequences (see Fig. 5.1). Consider the first characteristic of simple columns. A single feed is split into two products. As a first alternative to two simple columns, the possibilities shown in Fig. 5.10 can be considered. Here, three products are taken from one column. The designs are in fact both feasible and cost-effective when compared with simple arrangements on a standalone basis (i.e., reboilers and condensers operating on utilities) for certain ranges of conditions. If the feed is dominated by the middle product (typically more than 50 percent of the feed) and the heaviest product is present in small quantities (typically less than 5 percent), then the arrangement shown in Fig. 5.10a can be an attractive option. The heavy product must find its way down the column past the sidestream. Unless the heavy product has a small flow and the middle product a high flow, a reasonably pure middle product cannot be achieved. In these circumstances, the sidestream is usually taken as a vapor product to obtain a reasonably pure sidestream.  [c.147]

The cost of shaftwork required to run a refrigeration system can be estimated approximately as a multiple of the shaftwork required for an ideal system. The performance of an ideal system is given by  [c.207]

See pages that mention the term Cassitetite : [c.26]    [c.75]    [c.78]    [c.88]    [c.88]    [c.143]    [c.154]    [c.166]    [c.172]    [c.172]    [c.198]    [c.200]    [c.200]    [c.200]    [c.200]    [c.201]   
Modern inorganic chemistry (1975) -- [ c.167 ]