Improving Heat Exchanger Network Design of a Revamped Chemical Plant

____________________ Keyword: Pinch analysis, Industrial system, Utility, Heat exchanger, Payback Period. Indonesian Journal of Science & Technology Journal homepage: http://ejournal.upi.edu/index.php/ijost/ Indonesian Journal of Science & Technology 2 (1) (2017) 87-96 88 | Indonesian Journal of Science & Technology, Volume 2, Issue 1, April 2017 Hal. 87-96 DOI: http://dx.doi.org/10.17509/ijost.v2i1 pISSN 2528-1410 eISSN 2527-8045


INTRODUCTION
Pinch analysis is a well-known tool to give insights on thermal energy saving.(Aspelund et al., 2007) Its diverse applications in various industries show its usefulness.(Varbanov, 2014) For further information, literatures have covered this subject in great details.(Kemp, 2005;Linnhoff, 1993;Feng & Zhu, 1997).
Previous study has been reported about early phase process evaluation, (Putra, 2016).Here, in this article, the pinch analysis has been applied to improve the design of a chemical plant.A revamp project was done on the plant to keep up to the increasing market demands.As background information, its current and revamped situations will be discussed.Next, a pinch analysis of the plant is applied and a straightforward solution is recommended.Due to confidentiality, all drawings and data have been changed from the original case, leaving only the similar problem and the recommended solution.

CURRENT SITUATION
The chemical production is a very simple process, where the feed is reacted in a reactor and the outlet stream is separated in two sequential distillation columns.(Douglas, 1985) Unreacted feed is recycled to a feed tank, while the product and the waste are sent to their corresponding destinations.(Adams & Seider, 2006) The plant has already had an existing heat integration feature, where the feed stream is preheated with several streams coming from reactor and distillation columns.(Stankiewicz & Moulijn, 2000) The current plant runs without using heat exchanger HE-16, upstream of D1 (see Figure 1).The heat exchanger is connected to the cooling water system to further reduce the temperature of the outlet reactor before entering D1.Unfortunately, it was broken.Nonetheless, the operators did not have any problem in running the plant.Hence, HE-16 was bypassed and the stream was not further cooled (See Figure 1).

REVAMPED SITUATION
In the revamped situation, due to the new increased plant capacity, many of the unit operations are just enlarged.(Wang et al., 2011) This includes many of the heat exchangers and pumps.A new identical reactor is installed, while the two distillation columns are not modified.They are apparently big enough for the new capacity.The bigger outlet stream from the reactor will now need to be cooled down to keep the current operating situation in D1 the same.Due to the fact that the HE-16 is already present and it just needs some reparations, it was decided that this heat exchanger would be re-operated.Hence, this was the slight difference from the current situation.
An energy consumption analysis was done during the project.It was shown that the revamped situation would require external heating utilities of about 5.22 MW.It was about 36% lower compared to an old case without the existing heat integration feature.Similarly, it was required to discharge 5.53 MW to the cooling water system.This was 35% lower compared to a design without heat integration feature.A pinch analysis was performed (see Figure 2).This figure shows relationship between yielded hot/cold composite curves and temperature.The curves were obtained by different temperature minimum (∆Tmin) of 30C.It was calculated that the revamped case would be about 29% higher and 27% higher compared to the heating and the cooling utility targets (4.05 and 4.36 MW, respectively).Hence, based on these numbers, it seems that the current heat exchanger network design does not far off of the target.Nonetheless, an intriguing question was raised; could we do better within this expansion project regarding the energy consumption?If it is possible, how much energy could be reduced and how would be done that?

ANALYSIS OF THE EXISTING HEAT EXCHANGER NETWORK DESIGN
To answer such questions, the composite curves were revisited and the heat exchanger network design was observed deeper.(Kazantzi, 2006) Squeezing more from the heat integration does not lead to significant saving.This is shown in Figure 3, where ∆Tmin is varied from 5 to 60 o C, whereas the current heat exchanger network design could be said to have an average of 50 o C.
From this figure, it has been found that even using a small temperature ∆Tmin, it would not reduce both utilities significantly.On the contrary, it would definitely increase the plant complexity.Another look at the current heat exchanger network design, presented in a grid diagram, is shown in  Based on this analysis, there are two violations to the golden rules in pinch analysis.To make clear the discussion, dotted red lines can be shown in detail.First, the existing operation of HE-16 with cooling water is violating the rule of not to use cooling utility above the pinch temperature.Hence, its cooling water duty of 0.92 MW also yields the same amount of penalty to the heating utility.
The second violation is a heat transferred across pinch in HE-01.The amount of heat is with the amount of 0.134 MW.This will also give the same amount of penalties to both heating and cooling utilities.

PROPOSED SOLUTION
The above observation clearly shows what needs to be done to improve the current heat exchanger network design.Due to its significant amount and simpler to implement, removing the first violation is relatively a straightforward solution.Hence, the feed stream downstream of HE-01 is split into two streams.One part goes to its current destination (the existing HE-02), and the other part is rerouted to the existing HE-16.With this modification, the heating utility in HE-03/04 is reduced to 2.6 MW, from previously 3.5 MW (Figures 1 and 4).This is shown in Figures 5 and 6, in the grid and the block diagram, respectively.The dotted lines in both figures show the proposed modification.Compared to the current utilities of the revamped case, the proposed modification has a saving opportunity of about 0.92 MW.This translates into 165 k€/yr of saving, which is significant.Based on the estimated investment, the payback period is about 3 months.

Figure 4 .
The grid diagram is constructed with the temperature ∆Tmin of 30 o C and pinch temperature of 52 o C (see Figure2).

Figure 2 .
Figure 2. Composite curves of the plant

Figure 3 .
Figure 3. ∆Tmin variation of the composite curves The utility consumptions of the proposed solution are shown under the column of "proposed solution".The proposed solution requires each heating and cooling utility of only

Figure 4 .
Figure 4. Grid diagram of the heat exchanger network for the current design

Figure 5 .
Figure 5. Proposed modification shown in the grid diagram

Table 1 .
This table describes about comparison for heating utility and cooling utility.The table compares without heat integration, current utilities, target at temperature ∆Tmin = 30C, and proposed solution.