- Email: [email protected]
A novel approach to hot oil system design for energy conservation
Applied Thermal Engineering 66 (2014) 423e434
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
A novel approach to hot oil system design for energy conservation Abtin Ataei a, d, Nassim Tahouni b, *, Seyed Masoud Haji Seyedi a, Seyed Majid Hashemian c, ChangKyoo Yoo d, M. Hassan Panjeshahi b, e a
Graduate School of the Environment and Energy, Science and Research Branch, Islamic Azad University, Tehran, Iran School of Chemical Engineering, College of Engineering, University of Tehran, Tehran, Iran c Department of Mechanical Engineering, Shahrood University of Technology, Shahrood, Iran d Department of Environmental Science & Engineering, Green Energy Center/Center for Environmental Studies, Kyung Hee University, Yongin, South Korea e Department of Chemical and Petroleum Engineering, University of Calgary, Calgary, Alberta, Canada b
h i g h l i g h t s A new design method was developed for hot oil system by changing arrangement of HEN. Hot oil is supplied to a network of heater with a parallel/series configuration. Better hot oil generator performance and increased heating capacity was achieved. Debottlenecking procedures for the design of hot oil systems was developed.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 October 2013 Accepted 22 January 2014 Available online 31 January 2014
In this paper, a new systematic design methodology was developed for hot oil system by changing arrangement of heat exchanger network from parallel to mixed series/parallel. In re-circulating hot oil systems, hot oil from the hot oil generator is supplied to a network of heaters that usually has a parallel configuration. However, re-use of hot oil between different heating duties enables hot oil networks to be designed with series arrangements. This allows better hot oil generator performance and increased heating capacity, both in the context of new design and retrofit. First, the hot oil generator and the hot oil network were examined separately, in order to discuss the nature of hot oil system design. A model of hot oil systems was then developed to examine the performance of the hot oil generator to recirculation flow rate and return temperature, as well as to predict heating efficiency. In second step, the design of the overall hot oil system was developed by investigating the interactions between the hot oil network design and the hot oil generator performance. Debottlenecking procedures for the design of hot oil systems was also developed. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Hot oil system Hot oil generator Hot oil network Heat exchanger network Debottlenecking
1. Introduction Hot water heaters, steam heaters, and recirculating hot oil systems are all used to supply the required heat to a process [1]. Of these methods, recirculating hot oil systems are by far the most common today, as they do not have a drum to supply latent heat, and they are able not only to conserve process water and steam, but also to reduce energy and capital cost, compared to hot water and steam heater systems [2e4].
* Corresponding author. Tel.: þ98 21 6695 7788; fax: þ98 21 6695 7784. E-mail addresses: [email protected] (N. Tahouni), [email protected] (M.H. Panjeshahi). http://dx.doi.org/10.1016/j.applthermaleng.2014.01.044 1359-4311/Ó 2014 Elsevier Ltd. All rights reserved.
Hot oil heating is a type of indirect heating in which a liquid phase heat transfer medium is heated and circulated to one or more heat energy users within a closed loop system. Mineral oil, pressure-less synthetic oil and pressurized synthetic oil are common heat transfer mediums [4]. Main suppliers of hot oil packages and plants such as Sigma Thermal [4], Gaumer Process [5], Chromalox [6], Thermal Fluid Systems [7] and GTS Energy [8] offer those systems at up to 420 C to be used to heat revolving rolls, platens, molds, jacketed tanks and autoclaves in petrochemical, chemical and pulp-and-paper plants. The general specifications of hot oil systems offered by ChromaloxÒ are given in Table 1 [6]. However, no attention has been placed on the interactions between hot oil generator and heat exchanger networks [9,10], even though changes to operating conditions of hot oil systems
424
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Table 1 General specifications of hot oil systems offered by ChromaloxÒ [6]. Model
System type
Application
Operating temperature ( C)
Mbha
CMXO
Heat transfer non-pressurized Heat transfer non-pressurized Heat transfer non-pressurized Heat transfer pressurized Heat transfer pressurized Heat transfer multiple zone Heat transfer vaporizer
Synthetic oil
10e290
20.4e81.9
Synthetic oil
10e330
30.7e1365
Synthetic oil
10e315
30.7e2047
Synthetic oil
10e400
30.7e2047
SylthermÒ 800
40e400
30.7e2047
Synthetic oil
10e400
30.7e4094
30 to 400
51.2e1024
COS PFC CLD CLS OMHTS CHTV
Ò
Dowtherm TherminolÒ
a Mbh is the ASME & ANSI standard for one thousand British Thermal Units per hour.
frequently occur in industrial sites. Design and operating problems of hot oil generators have been the focus of attention on the part of manufacturers and process engineers. Mitra [10] represented the design considerations of hot oil systems and sizing of equipment. Mukherjee [11] addressed some techniques for improving the performance of the individual components in hot oil systems. Arnold [12] developed a differential equation model to do a detailed thermal analysis for a hot oil system in different temperature ranges. Ennis [13] discussed some of the key safety design and operational aspects of hot oil systems. Probert [14] presented a method for designing and rating of hot oil storage tanks. Halttunen [15] addressed a method to analysis energy costs of hot oil pumping systems. Wallace et al. [16] designed a new method for hot oil integration with a heat recovery system generator. Policastro [17] suggested an advanced, flexible control system for hot oil plants. Colaco and Floyd [18] developed a computerized method to do control and analysis of conditions of a hot oil system on a dynamic basis. Gu and Liu [19] presented an analysis on the flow process of hot oil in the organic heat transfer material heater based on finite time thermodynamics. Hlozek and Bardov [20] designed a hot oil system to do waste heat recovery from reciprocating gas engines. Nasir et al. [21] designed a hot oil system to utilize the turbine waste heat to generate electric power in a Neptune plant. Mostajeran Goortani et al. [22] introduced a hot oil system to recover the heat from stack gases and distribute it to the appropriate cold streams in a Kraft mill plant. Ohm et al. [23] described a drying technique for the upgrading of crushed low-rank coal utilizing a hot oil system. Singhmaneeskulchai et al. [24] studied on dynamic data reconciliation of a utility heat exchanger using hot oil from a waste heat recovery unit as a hot stream to heat up ethane product as a cold process stream from a natural gas separation plant. However, to date, research on heating systems have focused on the individual components [10], and not on the system as a whole. Because of the interactions between hot oil networks and hot oil generator performance, all of the heating system components should be considered when designing and operating these systems. Process integration method can be applied to address the interactions between the components of hot oil systems. Process Integration is a family of methodologies for combining several parts of processes or whole processes to reduce the consumption of resources [25]. This methodology examines the potential for improving and optimizing the heat exchange between heat sources and sinks in order to conserve energy, reduce costs and emissions [25,26]. Klemes et al. [25] represented the recent developments in process integration.
Various process integration approaches have been developed over the past two decades to study energy systems with the goal of identifying energy recovery opportunities in process industries. Thermal pinch analysis, which was developed for the analysis and optimization of heat exchanger networks, seems to be the most commonly used heat integration method [27]. Thermal pinch analysis was introduced by Linnhoff and Flower [28]. This analysis was expanded and widely publicized by Linnhoff and Hindmarsh [29], Smith [30,31], Kemp [32] and Gundersen [33].The mass pinch analysis was introduced by El-Halwagi and Manousiouthakis [34]. This analysis was applied in the area of wastewater minimization by Wang and Smith [35,36]. This method was widely expanded by Mann and Liu [37], Prakash and Shenoy [38] and Wan Alwi and Manan [39]. The mass and thermal pinch analyses were combined together by Savulescu et al. [40], Panjeshahi et al. [41,42], Ahmetovíc and Kravanja [43,44] to identify opportunities to simultaneously reduce the water and energy consumption in industrial facilities. Consider some of the possible changes to an existing hot oil system. A new heat exchanger might be introduced into the heat exchanger network, or the heat duty of heaters changed, or process changes might change the operating conditions. These process changes influence the conditions of the hot oil return and, consequently, affect the performance of the hot oil generator. In such situations, it is often not clear how the system will be affected by the new conditions or how the hot oil network design will affect the heating system. As such, a combined thermal and mass pinch analysis should be used to investigate the interactions for the overall system. In this paper, we present a systematic method for the design of hot oil systems that accounts for these interactions and process constraints, based on a combination of thermal and mass pinch analyses. First, the hot oil generator and the hot oil network will be examined separately, in order to discuss the nature of hot oil system design. A model of hot oil systems will be developed to examine the performance of the hot oil generator to recirculation flow rate and return temperature, as well as to predict heating efficiency. A methodology for hot oil network design will then be developed, assuming fixed inlet and outlet conditions for the hot oil. Finally, the design of the overall hot oil system will be developed by investigating the interactions between the hot oil network design and the hot oil generator performance [45]. Debottlenecking procedures for the design of hot oil systems will also be developed. In summary, moving from parallel to series arrangements for hot oil networks: Increases the efficiency of the hot oil system, Decreases the temperature differences in the hot oil heat exchangers.
2. Design of hot oil networks The current practice for hot oil network design most often uses parallel configurations [4,10,46], whereby hot oil is supplied directly to individual heat exchangers. After the hot oil has been used in each heat exchanger, the cold oil returns to the hot oil generator (Fig. 1a). The minimum hot oil demand is determined by minimizing the flow rate to the individual heat exchangers (Fig. 1b). Under a parallel arrangement [47], return hot oil flow rate and temperature become maximized, leading to poor generator performance [45]. Thus far, no systematic methods have been suggested for dealing with the design of hot oil networks, and the traditional parallel design is not flexible when dealing with various process
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
425
Fig. 1. Parallel configuration of hot oil networks.
restrictions. Therefore, a new hot oil network design methodology will now be developed. Not all heating duties require hot oil at the hot oil supply temperature, which allows us, if appropriate, to change the hot oil network from a parallel to a series design. A series arrangement, in which hot oil is reused in the network, returns the hot oil at a lower temperature and flow rate. From the predictions of the model, the heat supplied by this hot oil generator can be expected to increase under these conditions. In other words, if the design configuration is converted from parallel to series arrangements [48], the hot oil generator can service a higher heat load for the heaters. 2.1. New design methodology for hot oil networks A simple problem (Example 1) will be used to develop the design methodology for hot oil networks. The hot oil system in Example 1 has four heat exchangers that use hot oil as the heating medium for cold process streams. The temperature, flow rate, and heating duty of the cold process streams are shown in Table 2. The operating data was given from an actual plant. The data for cold process streams are represented as CP values, which is the product of specific heat capacity and flow rate. It is assumed that the heat capacity of hot oil is constant throughout the temperature range. The parallel configuration hot oil network in Example 1 has inlet and outlet CPs of 28.69 kW/ C and outlet temperature of 181.46 C. The non-pressurized synthetic oil supplied by ChromaloxÒ [6] is selected for this plant. According to Table 1 and Fig. 2, the model of that type of hot oil is COS [6]. The maximum operating oil temperature for this type of hot oil is 330 C. Therefore, the maximum hot oil inlet temperature is 320 C, taking into account the necessity of a 10 C safety margin. To develop a systematic method for the design of such systems, some clues can be taken from water pinch analysis [49] and developed for hot oil network design. In hot oil network analysis, it is assumed that any hot oil-using operation can be represented as a counter-current heat exchange operation with a minimum temperature difference (Fig. 3). As seen in Fig. 3, the hot oil inlet temperature to each heat exchanger should be between the maximum Table 2 Cold process stream data of hot oil networks: Example 1. Heat exchanger
Inlet stream temperature ( C)
Outlet stream temperature ( C)
CP (kW/ C)
Total load (kW)
1 2 3 4
230 230 95 95
310 250 250 170
10 25 10 15
800 500 1550 1125
hot oil supply temperature (320 C) and the temperature of the outlet cold process stream from that heat exchanger plus DTmin. The hot oil outlet temperature from each heat exchanger must be between the temperature of the inlet cold process stream to each heat exchanger plus DTmin and the minimum hot oil return temperature. The concept of the limiting hot oil profile is taken from water pinch analysis and shown in Fig. 4; it is defined here as the minimum inlet and outlet temperatures for the hot oil stream. As seen in Fig. 4, the hot oil supply line touches the composite curve at 240 C where the pinch point is defined. It means the hot oil supplied by generator should be directly sent to the heat exchangers which need the oil hotter than 240 C (above the pinch) and the outlet oil streams may be reused in other heat exchangers (below the pinch). As seen, these allowable temperatures are limited by the minimum temperature difference (DTmin). In a new design, this could be the practical minimum temperature difference for a given type of heat exchanger. In retrofitting, the temperature difference could be chosen to comply with the performance limitations of an existing heat exchanger under revised operating conditions of reduced temperature differences and increased flow rate. In addition, the limiting hot oil profile might be determined by other process constraints, such as corrosion, fouling, maximum allowable oil temperature, etc. Any hot oil line at or above this profile is considered a feasible design. The limiting profile is used to define a boundary between feasible and infeasible regions. The limiting hot oil profile allows the individual streams of the hot oil network to be represented on a common basis, as hot oil and energy characteristics are represented simultaneously. It should be emphasized that the final design will not necessarily feature the minimum temperature difference incorporated in the limiting data. It simply represents a boundary between feasible and infeasible conditions. Most heaters in the final network design will feature temperaturedriving forces greater than those used for the specification of limiting conditions. This study focuses primarily on retrofit design and, therefore, restricts consideration to deal only with cold streams to be heated by hot oil. A better design for hot oil networks will be exploited under a fixed heat exchanger network configuration. For a grassroot design, the design of hot oil networks and heat exchanger networks should be addressed simultaneously. In this paper, the topology of the heat exchanger network is assumed to be fixed. The duties of the hot and cold streams in the heat exchanger network are, thus, assumed to be unrelated to the heating system. In other words, the streams heated by the hot oil do not affect other streams in the heat exchanger network. As the inlet temperature of hot oil to heaters is decreased, the driving force for the heat exchangers is decreased, and additional heat exchanger area might be required. At the same time, however,
426
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 2. Hot oil and steam range.
Fig. 3. Representation of heat exchangers using hot oil.
the flow rate is increased. Therefore, the reduction in driving force from decreasing temperature difference is compensated by increased hot oil flow rate. The limiting hot oil data shown in Example 1 have been extracted from the cold process stream data and provided in Table 3. A “hot oil composite curve” can be constructed by combining all individual profiles into a single curve within temperature intervals (Fig. 5a). The design of the hot oil network is based on the hot oil composite curve, which represents overall limiting conditions of the whole network. The hot oil supply line is a straight line matched against the hot oil composite curve to represent the overall hot oil flow rate and conditions. Minimizing the outlet temperature of the hot oil supply line minimizes the flow rate of the hot oil by maximizing hot oil reuse (Fig. 5b). Each point
where the supply line touches the composite curve creates a pinch in the design. It is important to note that the interpretation of the pinch does not imply a zero driving force of heat transfer, but a minimum driving force. Only the parts of the design where the supply line touches the composite curve will feature minimum driving forces; all other parts will feature temperature differences above minimum. The water main method [41] for the design of water reuse networks can be extended to the design of hot oil networks. The original method identified water reuse opportunities for problems wherein reuse was constrained by concentration limits. To achieve the minimum hot oil flow rate target in design, the problem to be decomposed into design regions, as illustrated in Fig. 6. This shows the hot oil composite curve from Fig. 5 decomposed into two regions: above and below the pinch. Above the pinch, from 320 to 240 C, the system requires the full flow rate of 18.92 kg/s. However, below the pinch, a lower flow rate will solve the problem. Fig. 6 shows a steeper line drawn against the part of
Table 3 Limiting hot oil data: Example 1.a Heat Inlet stream Outlet stream CP (kW/ C) Total load (kW) exchanger temperature ( C) temperature ( C) 1 2 3 4 Fig. 4. Strategy for minimizing the flow rate of heating oil.
a
320 260 260 180
240 240 105 105
10 25 10 15
DTmin ¼ 10 C; hot oil inlet temperature ¼ 320 C.
800 500 1550 1125
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
427
Fig. 5. Hot oil composite curve and targeting for maximum reuse.
the problem below the pinch. Of the CP of 18.92 kW/ C from the hot oil, after use up to the pinch condition of 240 C, only a CP of 18.33 kW/ C is used, and the balance (CP ¼ 0.59 kW/ C) is returned directly to the hot oil furnace. In this way, each part of the problem only uses the minimum amount of hot oil. However, it should again be noted that, in minimizing the amount of hot oil used, the temperature differences in the heater are also minimized, leading to tradeoffs between the designs of the hot oil system and the heat exchangers. As shown in Fig. 6, three hot oil mains are conceptualized: one at the supply temperature of 320 C, a second at the pinch temperature of 240 C, and a third at 105 C, the minimum temperature allowable in the system. The flow rate of hot oil required in each of the mains is shown at the top of the main. The flow rate to be returned to the hot oil furnace from each of the mains is shown at the bottom of the main. The heating streams are superimposed onto the hot oil mains at their appropriate temperatures. In Fig. 6, Stream 1, which starts at 320 C and terminates at 240 C, is shown between the 320 C and 240 C hot oil mains. Stream 2 starts at 260 C, between the first two hot oil mains. Stream 3 starts at 260 C, again between the first two hot oil mains. However, in this case, Stream 3 terminates at the minimum temperature of 105 C. Initially, therefore, it is broken into two parts, each within the appropriate design region. Finally, Stream 4 starts at 180 C below the pinch oil mains and finishes at the105 C hot oil main. The streams are then connected to the appropriate hot oil main, to satisfy the individual heating requirement. This provides an initial design for the hot oil network to meet the target requirements. However, the design has not yet been completed, as there is a fundamental difficulty with the arrangement shown in Fig. 6. Stream 3 requires a change in flow rate at the pinch hot oil main
Fig. 6. Completed hot oil design grid.
(240 C), which means that the heating duty corresponding with Stream 3 would have to be broken down into two heat exchangers, each being supplied with different flow rates. The change in flow rate for Stream 3 is, in fact, easily removed. Consider Fig. 7, which shows the temperature vs. heat duty for an operation with a change in flow rate, as in the case of Stream 3 in Fig. 6. A heat balance around Part 1 in Fig. 7 gives (assuming the specific heat capacity of the hot oil to be constant):
F1 ðTPinch T1 Þ ¼ F2 TPinch Tin;Max
(1)
Moving the mixing junction to the inlet of the operation, as shown in Fig. 7, and carrying out a heat balance at the new mixing junction, gives [42]:
Tin ¼
ððF2 F1 Þ*TPinch Þ þ ðF1 *T1 Þ T ðF1 *ðTPinch T1 ÞÞ ¼ Pinch F2 F2 (2)
Substituting Equation (1) into Equation (2) gives:
Tin ¼
ðF2 *TPinch Þ ðF2 *ðTPinch Tin:Max ÞÞ ¼ Tin;min F2
(3)
In other words, if the mixing junction is moved from the middle to the beginning of the operation, there is a constant flow rate throughout the operation that corresponds with the minimum inlet temperature after mixing. Fig. 8 shows the corresponding grid diagram to correct the change in flow rate. The change in flow rate for Stream 3 that previously occurred at the pinch temperature mains is now added from the pinch temperature mains to the inlet of Stream 3. This provides a design for the hot oil network that achieves the target minimum flow rate of CP ¼ 18.92 kW/ C. The arrangement shown in Fig. 8 involves reuse of hot oil from Streams 1 and 2 into Streams 3 and 4 via a hot oil main at pinch temperature 240 C. An alternative way to arrange the design is to make the connection directly, rather than through intermediate hot oil main. If the intermediate hot oil main is removed, then there are basically two sources of hot oil from Streams 1 and 2 at 240 C and two sinks for hot oil in Streams 3 and 4 at 240 C. These sources and sinks can be connected together in different ways. If the intermediate hot oil main at 240 C is removed from the design, then the streams can be connected directly together. Fig. 9 shows a flow sheet for one possible arrangement. In Fig. 9, there is reuse from Heat Exchanger 1 to Heat Exchanger 3 and from Heat Exchanger 2 to Heat Exchanger 4. Some of the hot oil from the furnace goes through Heat Exchanger 1, and some bypasses it, to be mixed in before entering Heat Exchanger 3. This step is necessary to comply with the inlet constraints for Heat Exchanger 3. The flow sheet in Fig. 9, then, needs to be assessed for its practicality and operability; the
428
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 7. Changing the mixing arrangement can avoid a change in hot oil flow rate.
design can then be evolved. For example, the bypass around Heat Exchanger 1 in Fig. 9 can be eliminated, and all of the CP ¼ 12.67 kW/ C can be put through Heat Exchanger 1 before entering Heat Exchanger 3. In the design, the essential feature for achieving the target is the reuse of hot oil between exchangers. Different arrangements than the one shown in Fig. 10 are possible for achieving the target by connecting the sources and sinks at 240 C differently. Parallel hot oil use that is minimized to achieve DTmin at the outlet of each heat exchanger leads to a flow rate for the network of 9.66 kg/s, with a hot oil return temperature of 181.3 C. Maximizing the reuse reduces this flow rate to 6.39 kg/s, and the hot oil return temperature decreases to 110 C (Table 4). A number of complications are likely to be encountered, which need to be addressed:
hot oil composite curve, as illustrated in Fig. 10. If more design regions exist, additional hot oil mains are required. However, the procedure is, basically, the same as in the example described above, taking into account the additional mains. b) Furthermore, if the hot oil supply line does not correspond with the minimum flow rate because of system interactions or temperature constraints for dew point 130 C (assuming DT ¼ 30 C for flue gas and hot oil, exhaust temperature of flue gas will be 160 C), then a pinch point is not created with the limiting hot oil composite curve, no pinch means that the temperature driving force is negative and is not acceptable (Fig. 11a). The setting could be between minimum flow rate (maximum reuse) and no reuse (parallel arrangement), as shown in Fig. 11a. The oil
a) One complication that can occur is that the number of design regions can be greater than two. Fig. 10 shows a design problem involving just two design regions: below the pinch and above the pinch. The hot oil composite curve in Fig. 10 has three design regions, rather than two. The design regions are identified by drawing straight lines between the extreme convex points of the Fig. 9. Flow sheet of hot oil network with maximum reuse of hot oil.
Fig. 8. Final grid design after changing the mixing arrangement to avoid flow rate changes.
Fig. 10. More complex problems might require a greater number of hot oil mains.
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
429
Table 4 Comparison of exit conditions of hot oil networks. Method
Flow rate (kg/s)
CP (kW/ C)
THot
Parallel Max. reuse %
10.58 6.98 34
28.68 18.92 34
181.5 110 39
oil; out
( C)
main method is based on the concept of the pinch point and cannot be applied to problems without a pinch. The new design methodology should provide for hot oil networks without a pinch. The limiting profile represents the boundary between feasible and infeasible operation. In other words, any composite curve above the original one is feasible. Thus, the hot oil composite curve can be modified in the feasible region without creating feasibility
Fig. 12. Finding a new pinch point: Example 1.
problems (Fig. 11b). If the hot oil composite curve could be modified to make a pinch point with the desired hot oil supply line in the feasible region, the hot oil network problem would be changed into a problem with a pinch. Pinch migration is introduced here to
Fig. 11. Hot oil composite curve development.
430
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 13. Pinch migration and temperature shift: Example 1.
convert problems without a pinch into those with a pinch with the desired supply line (Fig. 11b). Two approaches to pinch migration can be adopted (Fig. 11c). The first is to shift heat load, in which the hot oil composite curve moves along the heat load axis. The second is temperature shift, in which the hot oil composite curve moves along the temperature axis. In the two approaches, temperature shift is adopted, because heat load shift would result in an energy penalty. The next problem is how to find the new pinch and how to modify the composite curve with a temperature shift. Let us introduce a target temperature of 130 C for the hot oil for Example 1. A new pinch is created between the modified composite curve and the new supply line, which is calculated from a simple heat balance (Fig. 12). The new calculated pinch of 248.3 C is migrated from the original pinch of 240 C. It is necessary for individual duties to apply a temperature shift for modification of the composite curve. Hot oil Streams 1, 2, and 3 take part in creating the original pinch, which means Streams 1, 2, and 3 are the candidates for temperature shift. The limiting hot oil modifications occur in two stages. The first stage is to shift the temperature of the limiting hot oil profiles according to the value of the temperature shift (8.3 C in this example). Modified profiles might cross the supply line, thus requiring another step. The second stage is to increase the flow rate of the limiting hot oil profile when the shifted profile is restricted by temperature limitations. The limiting hot oil profile is modified to satisfy temperature limitations by increasing the hot oil flow rate CP. In Example 1, Streams 2 and 3 can be modified to obtain new limiting hot oil data simply by shifting temperatures. However, for Stream 1, it is necessary to increase flow rate, because the 20 C hot oil supply temperature restricts the temperature shift of the
limiting data. The heat balance equations determine the increased flow rate and the new limiting exit temperature [31].
CPnew ¼ CPold
new THO;in
old
¼ CP
old T old TPinch HO;in
! (4)
new T new TPinch HO;in
old old THO;out THO;in
!
CPnew
After modification of the conditions for each individual heat exchanger, the modified hot oil profiles are as shown in Fig. 13, and the new limiting hot oil data are as shown in Table 5. For Stream 1, the CP is increased from 10 to 11.18 kW/ C, as a result of the second stage modification. The new composite curve is constructed by combining all modified limiting profiles. The modified hot oil composite now creates a pinch point with the desired hot oil supply line. The hot oil network design can now be carried out using the hot oil mains method. The resulting design is shown in Fig. 14. The hot oil generator return temperature and flow rate agree with the target. The pinch migration and temperature shift method enables creation of a design with any target temperature. The hot oil network will have different configurations with different target temperatures, which can be seen by comparing the maximum reuse design (Fig. 9) with the design with a temperature constraint (Fig. 14). The pumping load and operating cost of the base and new design method are given in Table 6.
Table 5 Temperature-shifted limiting hot oil data. Heat Inlet stream Outlet stream CP (kW/ C) Total load (kW) exchanger temperature ( C) temperature ( C) 1a 2a 3a 4 a
320 268.3 268.3 180
Modified data.
248.3 248.3 113.3 105
11.15 25 10 15
(5)
800 500 1550 1125 Fig. 14. Hot oil network design without a pinch.
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434 Table 6 eThe grassroot project results.
431
Table 9 Hot oil outlet conditions of parallel design.
Parameters
Base
New
%
Parameters
Base
New
%
Pump load (kW) Gas exhaust CP require (kW/ C) Operation cost ($/yr)
4999 13.72 1,094,965
4036 11.96 863,863
19 13 21
Outlet stream temperature ( C) Outlet CP (kW/ C) Heat load (kW)
209.13 34.50 3825
201.24 38.94 4625
3.92 12.88 17.29
Table 7 Limiting hot oil data of base case.a Heat exchanger
Inlet stream temperature ( C)
Outlet stream temperature ( C)
CP (kW/ C)
Total load (kW)
1 2 3
320 245 180
245 200 120
15 40 15
1125 1800 900
a
DTmin ¼ 10 C; hot oil inlet temperature ¼ 320 C.
3. Debottlenecking of hot oil systems When hot oil networks need to increase the heat load of individual heaters or a new heat exchanger is introduced into an existing system, hot oil systems can become bottlenecked [31]. As the increase in heating load influences the hot oil generator performance, and there are interactions between hot oil networks and the hot oil generator, the best solution is often obtained by modifying the network. 3.1. General considerations for debottlenecking From the previous results of hot oil generator modeling and hot oil network design, a general guideline for the design of hot oil systems can be suggested. The heat supplied by the generator can be increased by changing inlet hot oil conditions from high flow rate and temperature to low flow rate and temperature. In general, doing so will require changing the hot oil network design from parallel to series or mixed parallel/series arrangements, with reuse of hot oil, decreasing the flow rate of hot oil and the return temperature. By changing from parallel arrangements to hot oil reuse designs, the heat supplied by the generator can be increased without any energy penalty and without investment in a new hot oil generator. A debottlenecking procedure [31,35] for hot oil systems will now be developed, using Example 2. The base case for Example 2 has three existing heat exchangers. The limiting hot oil data are provided in Table 7. In this example, a new heat exchanger (Table 8) is introduced into the base case, which makes the hot oil system bottlenecked. New outlet conditions of the hot oil when the parallel arrangement is retained with the new heat exchanger are shown in Table 9. The flow rate and heat load of the return oil to the generator increase, although the temperature decreases, thereby influencing the performance of the hot oil generator. Fig. 15 shows the performance of the parallel arrangement. First, the hot oil inlet temperature (Tin ¼ 310 C) to the network is colder than the desired
inlet temperature (320 C), which means that additional heating equipment needs to be installed to heat the hot oil to the minimum permissible inlet temperature (320 C). Second, the heat load of the network (4.625 MW) is bigger than the heat supply of the hot oil generator (4.235 MW), which also means that another 0.389 MW of heat load needs to be dissipated by additional heating. When the traditional parallel arrangements with new operating conditions are applied, the hot oil flow rate and the heat load of the generator are consequently increased. If there are no other design options than parallel arrangements, an additional hot oil generator (or supplemental firing) is needed to satisfy the new bottlenecked conditions.
3.2. Debottlenecking design procedures of hot oil systems A design procedure for debottlenecking hot oil systems will now be developed. The hot oil composite curve can first be constructed from the limiting hot oil data. The hot oil network performance can be changed within a feasible region that is bounded by the maximum reuse supply line and the parallel design supply line (Fig. 16a). In Fig. 16b, the feasible hot oil supply line (line AB) represents the attainable outlet conditions from the hot oil generator model by changing design configurations. As the inlet conditions to the generator affect its performance, it is necessary to know how the inlet conditions affect the hot oil generator system [50]. The hot oil supply line has the same heat load (4.625 MW) from the viewpoint of the hot oil network (Fig. 16a). However, the heat supplied by the hot oil system changes as inlet conditions to the hot oil generator change (Table 10). The heat supplied by the generator systems increases as the design configuration changes from parallel to maximum reuse (A to B in Fig. 16b). For our example, the following conditions should be satisfied for the new hot oil network design:
Table 8 Limiting hot oil data for new heat exchanger. Heat exchanger
Inlet stream temperature ( C)
Outlet stream temperature ( C)
CP (kW/ C)
Total load (kW)
1 2 3 4
320 245 180 220
245 200 120 140
15 40 15 10
1125 1800 900 800
Fig. 15. Changes in hot oil systems with parallel arrangements.
432
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 16. Hot oil supply line targeting.
(1) The inlet temperature to the hot oil network should be 320 C. (2) The heat supplied by the hot oil generator system should be equal to the heat load of the hot oil network. In this example, it is not necessary to achieve a temperature higher than 320 C. As can be seen in Table 8, the target conditions lie somewhere along the feasible hot oil supply line. The next stage is to find the target supply conditions for the hot oil generator. The feasible hot oil supply line can move from BN to BM in Fig. 16b. The target conditions, which satisfy the desired temperature to hot oil network (320 C), are found by changing the hot oil supply conditions from BN to BM. The heat removal of the heating system is the same as the heat load of the hot oil network at the target conditions (B*), where the inlet temperature to the hot oil network is satisfied. Target conditions are given by the intersection
Table 10 -Effects of hot oil inlet conditions. Case
Heat removal case of hot oil system (MW)
Parallel (A) Maximum reuse (B) Target
4.235 4.685 4.625
between the feasible hot oil supply line and the isothermal line of the heating system outlet temperature. The target conditions for debottlenecking have been found by using the heating system model. The target conditions are a CP of 28.3 kW/ C and a temperature of 156 C. At target conditions, the heating demands of the networks are satisfied without additional heating capacity. Below the target temperature, the current heating systems cannot operate up to heating demand. The next stage is to design the hot oil network with target conditions. As the new hot oil supply line has no pinch with the limiting composite curve, the temperature shift and pinch migration method is applied, as explained in the previous section. The new pinch point is calculated, and the limiting hot oil profile is modified (Fig. 17). The final design for the debottlenecked hot oil system is shown in Fig. 18. The suggested method for hot oil network design is based on a conceptual design methodology and, therefore, other design configurations can evolve. The design complexity can be reduced for the sake of simplicity, but this would likely result in a heating system performance penalty. The design of hot oil systems involves tradeoffs, including hot oil generator costs, pressure drops, piping costs, design complexity, etc. An optimization method is required to make the tradeoffs in a structured way; this issue will be the objective of future work. The proposed debottlenecking procedure enables the hot oil generator to manage the increased heat load by changing the
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
433
Fig. 17. Pinch migration and temperature shift: Example 2.
However, the CP value of the common line for the parallel design is greater than that of the reuse design (Table 11). Therefore, those modifications should carefully be taken into account in practical debottlenecking projects. 4. Conclusions Fig. 18. Final design of debottlenecked hot oil systems.
network design from parallel to series arrangements. The design method targets the hot oil generator conditions and then designs the hot oil network for the new target conditions. The design procedure for debottlenecking hot oil systems can be summarized as follows: (1) Define the feasible hot oil supply line from the composite curve and parallel supply line. (2) Target the hot oil generator supply conditions from the heating systems model and the feasible hot oil supply line. (3) Design the hot oil network for the target conditions with pinch migration and temperature shifting. The flow rate to the individual heat exchangers is likely to change; therefore, the design of individual heat exchangers needs to be checked to ensure that it is feasible. In addition, the procedure changes the conditions of the return hot oil and the recirculation hot oil flow rate. As a result, the pressure drop in the equipment and piping will also change. Therefore, the performance of the hot oil pumps needs to be checked, because the pumping head, efficiency, and required power depend on the flow rate carried by the pump. When the design (Fig. 18) is compared with parallel arrangements, the CP value of the individual heat exchangers in the parallel design is the same as or less than that of the reuse design (Table 11). Table 11 CP vs. hot oil network design. Heat exchanger
CP (kW/ C) Parallel (no reuse)
Target (reuse)
1 2 3 4 Total Common linea
15 15 4.5 4.44 38.9 38.9
17.09 23.76 10.00 10.08 60.95 28.20
%
13.98 58.42 122.33 127.01 56.51 27.58
a Common line refers to the hot oil pipeline between the hot oil generator and the hot oil network.
A new methodology for the design of hot oil networks has been developed, in order to satisfy any supply conditions for the hot oil generator. The design can be carried out with any target temperature by introducing the concepts of pinch migration and temperature shift. From the interactions between the performance of the hot oil generator and the design of the heaters, the proposed debottlenecking procedures allow increased capacity without investment in new hot oil generator equipment when generator capacity is limited. The heat load distribution of heating systems has also been considered when a hot oil system is bottlenecked beyond generator capacity or when a temperature constraint limits the return temperature. A number of design options for debottlenecking heating systems have been discussed, in order to improve hot oil generator performance and to distribute heat load between the generator and other design options. Acknowledgements The financial support of the Iran Energy Efficiency Organization (IEEO), (SABA) is gratefully acknowledged. Nomenclature B* targeted condition CP heat capacity multiplied by flow rate, (kW/ C) DTshift amount of temperature shift, ( C) DTmin minimum temperature approach, ( C) F flow rate F1 outlet hot oil flow rate of exhaust heat exchanger, (t/h) F2 inlet hot oil flow rate of exhaust heat exchanger, (t/h) F3 hot oil flow rate flowing into the hot oil network, (t/h) H enthalpy, (kJ/kg) Q heat load, (kW) T temperature, ( C) T* migrated pinch temperature, ( C) T1 outlet hot oil temperature of exhaust heat exchanger, ( C) T2 inlet hot oil temperature of exhaust heat exchanger, ( C) T3 hot oil temperature flowing into the hot oil network, ( C)
434
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Subscripts Cold cold process stream in heat exchanger in inlet conditions Max maximum Min minimum out outlet conditions P pinch point Superscripts HE process/hot oil heat exchanger HEN heat exchanger network HO hot oil New migrated pinch point Old original pinch point References
[22]
[23]
[24]
[25] [26] [27]
[28] [29]
[1] A. Ataei, K.S. Lee, J.J. Lim, M.J. Kim, H.B. Liu, O.Y. Kang, T.S. Oh, C.K. Yoo, A review on environmental process engineering, Int. J. Environ. Res. 5 (4) (2011) 875e890. [2] A. Ataei, Application of combined pinch and exergy analysis in retrofit of an olefin plant for energy conservation, Sci. Res. Essays 6 (12) (2011) 2437e2446. [3] A. Ataei, C.K. Yoo, Combined pinch and exergy analysis for energy efficiency optimization in a steam power plant, Int. J. Phys. Sci. 5 (7) (2010) 1110e1123. [4] Sigma Thermal, Thermal Oil Heating Systems, 2013. www.sigmathermal.com (accessed 12.12.13). [5] Gaumer Process, Gaumer Hot Oil Systems, 2013. www.gaumer.com (accessed 12.12.13). [6] Chromalox, Hot Oil Heaters e Heat Transfer Systems, 2013. www.chromalox. com (accessed 12.12.13). [7] Thermal Fluid Systems, Hot Oil Heating Systems, 2013. www.tfsheat.com (accessed 12.12.13). [8] GTS Energy, Thermal Fluid Heaters, 2013. www.gtsenergy.com (accessed 12.12.13). [9] A. Ataei, C.K. Yoo, Application of pinch technology for waste heat recovery in the linear-alkyl-benzene sulfonation process, Asian J. Chem. 23 (4) (2011) 1539e1547. [10] S. Mitra, Design Considerations of Hot Oil System, 2013. www.academia.edu/ 1514808/design_consideration_of_hot_oil_system (accessed 07.06.13). [11] R. Mukherjee, Role of optimal heat exchange in chemical plant energy conservation, Chem. Eng. World 19 (10) (1984) 47e49. [12] F.G. Arnold, Thermal analysis of hot oiling, Am. Soc. Mech. Eng. (1990) 4. PET1, Energy-Sources Technology Conference and Exhibition, New Orleans, LA, USA, 14 January 1990 through 18 January 1990. [13] T. Ennis, Safety in design of thermal fluid heat transfer systems, Inst. Chem. E 155 (2009) 162e169, 21st Institution of Chemical Engineers Symposium on Hazards 2009 e Hazards XXI: Process Safety and Environmental Protection, Manchester, United Kingdom, 10 November 2009 through 12 November 2009. [14] D. Probert, Design and performance of hot-oil storage tanks, Appl. Energy 1 (4) (1975) 247e278. [15] O.H. Halttunen, Analyzing energy costs of hot oil pumping systems e comparison of centrifugal, canned-motor, and seal-free magnetic drive pumps, Plant Eng. 32 (10) (1978) 155e161. [16] P.S. Wallace, K.A. Johnson, J.L. Kasbaum, Hot Oil Integration with Heat Recovery System Generator and Method of Operation, US Patent 6,606,862, 2003. [17] E.F. Policastro, Hot oil, cool control, InTech 55 (4) (2008) 24e25. [18] R.J. Colaco, M.C. Floyd, Supervisory operation and data system controls hot oil pipeline, Oil Gas J. 74 (9) (1976) 87e90. [19] W. Gu, Y. Liu, Analysis on the flow process of hot oil in the organic heat transfer material heater based on finite time thermodynamics, Adv. Mater. Res. 250e253 (2011) 2979e2983, 1st International Conference on Civil Engineering, Architecture and Building Materials, CEABM 2011, Haikou, China, 18 June 2011 through 20 June 2011. [20] R.J. Hlozek, J.B. Bardov, Waste heat recovery from reciprocating gas engines: glycol regeneration, in: Proceedings, Annual Convention e Gas Processors Association, 1980, pp. 100e104. Proc Annu Conv Gas Process Assoc Tech Pap 59th, Houston, TX, USA, 17 March 1980 through 19 March 1980. [21] P. Nasir, S. Jones, D. Schochet, D. Posner, Utilization of turbine waste heat to generate electric power at Neptune plant, in: GPA Annual Convention Proceedings, 2004, p. 9. Gas Processors Association e 83rd Annual Convention
[30] [31] [32]
[33]
[34] [35] [36] [37] [38]
[39]
[40] [41]
[42] [43] [44]
[45] [46]
[47]
[48] [49] [50]
Proceedings 2004: Linking Energy Partners for the Future, New Orleans, LA, United States, 14 March 2004 through 17 March 2004. B. Mostajeran Goortani, E. Mateos-Espejel, M. Moshkelani, J. Paris, Energy efficiency improvement of a Kraft process through practical stack gases heat recovery, Appl. Therm. Eng. 31 (2011) 4091e4096. T.I. Ohm, J.S. Chae, J.H. Lim, S.H. Moon, Evaluation of a hot oil immersion drying method for the upgrading of crushed low-rank coal, J. Mech. Sci. Technol. 26 (4) (2012) 1299e1303. P. Singhmaneeskulchai, N. Angsutorn, K. Siemanond, Dynamic data reconciliation in a hot-oil heat exchanger for validating energy consumption, Chem. Eng. Trans. 35 (2013) 493e498, http://dx.doi.org/10.3303/CET1335082. J.J. Klemes, P.S. Varbanov, Z. Kravanja, Recent developments in process integration, Chem. Eng. Res. Des. 91 (2013) 2037e2053. B. Bakhtiari, S. Bedard, Retrofitting heat exchanger networks using a modified network pinch approach, Appl. Therm. Eng. 51 (2013) 973e979. R.F. Dunn, M.M. El-Halwagi, Review process integration technology review: background and applications in the chemical process industry, J. Chem. Technol. Biotechnol. 78 (2003) 1011e1021, http://dx.doi.org/10.1002/ jctb.738. B. Linnhoff, J.R. Flower, Synthesis of heat exchanger networks: I. Systematic generation of energy optimal networks, AIChE J. 24 (4) (1978) 633e642. B. Linnhoff, E. Hindmarsh, The pinch design method for heat exchanger networks, Chem. Eng. Sci. 38 (1983) 745e763. R. Smith, Chemical Process Design, McGraw-Hill, New York, USA, 1995. R. Smith, Chemical Process Design and Integration, John Wiley & Sons, UK, 2005. I.C. Kemp, Pinch Analysis and Process Integration, a User Guide on Process Integration for Efficient Use of Energy, Elsevier, Amsterdam, The Netherlands, 2007. T. Gundersen, Heat Integration: Targets and Heat Exchanger Network Design (Chapter 4), in: J.J. Klemes (Ed.), Handbook of Process Integration (PI): Minimisation of Energy and Water Use, Waste Emissions, Woodhead Publishing, Cambridge, UK, 2013. M.M. El-Halwagi, V. Manousiouthakis, Synthesis of mass-exchange networks, AIChE J. 35 (1989) 1233e1244. http://dx.doi.org/10.1002/aic.690350802. Y.P. Wang, R. Smith, Wastewater minimisation, Chem. Eng. Sci. 49 (7) (1994) 981e1006. Y.P. Wang, R. Smith, Waste-water minimisation with flow-rate constraints, Chem. Eng. Res. Des. 73 (8) (1995) 889e904. J.G. Mann, Y.A. Liu, Industrial Water Reuse and Wastewater Minimization, McGraw Hill, New York, USA, 1999, ISBN 0071348557. R. Prakash, U.V. Shenoy, Targeting and design of water networks for fixed flowrate and fixed contaminant load operations, Chem. Eng. Sci. 60 (1) (2005) 255e268. S.R. Wan Alwi, Z.A. Manan, Water pinch analysis for water management and minimisation: an introduction, in: Klemes (Ed.), Process Integration Handbook, Woodhead Publishing, Cambridge, UK, 2013, ISBN 978 0 85709 593 0. http://dx.doi.org/10.1533/9780857097255.3.353. L.E. Savulescu, M. Sorin, R. Smith, Direct and indirect heat transfer in water network systems, Appl. Therm. Eng. 22 (8) (2002) 981e988. M.H. Panjeshahi, A. Ataei, M. Gharaie, R. Parand, Optimum design of cooling water systems for energy and water conservation, Chem. Eng. Res. Des. 87 (2) (2009) 200e209. M.H. Panjeshahi, M. Gharaie, A. Ataei, Debottlenecking procedure of effluent thermal treatment system, Energy 35 (2010) 5202e5208. E. Ahmetovíc, Z. Kravanja, Simultaneous synthesis of process water and heat exchanger networks, Energy 57 (2013) 236e250. E. Ahmetovíc, Z. Kravanja, Simultaneous optimization of heat-integrated water networks involving process-to-process streams for heat integration, Appl. Therm. Eng. 62 (1) (2014) 302e317. http://dx.doi.org/10.1016/j. applthermaleng.2013.06.010. B. Galan, I.E. Grossman, Optimal design of distributed wastewater treatment networks, Ind. Eng. Chem. Res. 37 (10) (1998) 4036e4048. A. Azhdari, H. Ghadamian, A. Ataei, C.K. Yoo, A new approach for optimization of combined heat and power generation in edible oil plants, J. Appl. Sci. 9 (21) (2009) 3813e3820. M.M. El-Halwagi, A.M. El-Halwagi, V. Manousiouthakis, Optimal design of dephenolization networks for petroleum-refinery wastes, Trans. IChemE (Part B) 70 (1992) 131e139. J.K. Kim, R. Smith, Cooling water system design, Chem. Eng. Sci. 56 (12) (2001) 3641e3658. W.J. Kuo, R. Smith, Effluent treatment system design, Chem. Eng. Sci. 52 (23) (1997) 4273e4290. C.K. Yoo, A. Ataei, Y.S. Kim, M.J. Kim, H.B. Liu, J.J. Lim, Environmental systems engineering: a state of the art review, Sci. Res. Essays 5 (17) (2010) 2341e 2357.
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
A novel approach to hot oil system design for energy conservation Abtin Ataei a, d, Nassim Tahouni b, *, Seyed Masoud Haji Seyedi a, Seyed Majid Hashemian c, ChangKyoo Yoo d, M. Hassan Panjeshahi b, e a
Graduate School of the Environment and Energy, Science and Research Branch, Islamic Azad University, Tehran, Iran School of Chemical Engineering, College of Engineering, University of Tehran, Tehran, Iran c Department of Mechanical Engineering, Shahrood University of Technology, Shahrood, Iran d Department of Environmental Science & Engineering, Green Energy Center/Center for Environmental Studies, Kyung Hee University, Yongin, South Korea e Department of Chemical and Petroleum Engineering, University of Calgary, Calgary, Alberta, Canada b
h i g h l i g h t s A new design method was developed for hot oil system by changing arrangement of HEN. Hot oil is supplied to a network of heater with a parallel/series configuration. Better hot oil generator performance and increased heating capacity was achieved. Debottlenecking procedures for the design of hot oil systems was developed.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 October 2013 Accepted 22 January 2014 Available online 31 January 2014
In this paper, a new systematic design methodology was developed for hot oil system by changing arrangement of heat exchanger network from parallel to mixed series/parallel. In re-circulating hot oil systems, hot oil from the hot oil generator is supplied to a network of heaters that usually has a parallel configuration. However, re-use of hot oil between different heating duties enables hot oil networks to be designed with series arrangements. This allows better hot oil generator performance and increased heating capacity, both in the context of new design and retrofit. First, the hot oil generator and the hot oil network were examined separately, in order to discuss the nature of hot oil system design. A model of hot oil systems was then developed to examine the performance of the hot oil generator to recirculation flow rate and return temperature, as well as to predict heating efficiency. In second step, the design of the overall hot oil system was developed by investigating the interactions between the hot oil network design and the hot oil generator performance. Debottlenecking procedures for the design of hot oil systems was also developed. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Hot oil system Hot oil generator Hot oil network Heat exchanger network Debottlenecking
1. Introduction Hot water heaters, steam heaters, and recirculating hot oil systems are all used to supply the required heat to a process [1]. Of these methods, recirculating hot oil systems are by far the most common today, as they do not have a drum to supply latent heat, and they are able not only to conserve process water and steam, but also to reduce energy and capital cost, compared to hot water and steam heater systems [2e4].
* Corresponding author. Tel.: þ98 21 6695 7788; fax: þ98 21 6695 7784. E-mail addresses: [email protected] (N. Tahouni), [email protected] (M.H. Panjeshahi). http://dx.doi.org/10.1016/j.applthermaleng.2014.01.044 1359-4311/Ó 2014 Elsevier Ltd. All rights reserved.
Hot oil heating is a type of indirect heating in which a liquid phase heat transfer medium is heated and circulated to one or more heat energy users within a closed loop system. Mineral oil, pressure-less synthetic oil and pressurized synthetic oil are common heat transfer mediums [4]. Main suppliers of hot oil packages and plants such as Sigma Thermal [4], Gaumer Process [5], Chromalox [6], Thermal Fluid Systems [7] and GTS Energy [8] offer those systems at up to 420 C to be used to heat revolving rolls, platens, molds, jacketed tanks and autoclaves in petrochemical, chemical and pulp-and-paper plants. The general specifications of hot oil systems offered by ChromaloxÒ are given in Table 1 [6]. However, no attention has been placed on the interactions between hot oil generator and heat exchanger networks [9,10], even though changes to operating conditions of hot oil systems
424
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Table 1 General specifications of hot oil systems offered by ChromaloxÒ [6]. Model
System type
Application
Operating temperature ( C)
Mbha
CMXO
Heat transfer non-pressurized Heat transfer non-pressurized Heat transfer non-pressurized Heat transfer pressurized Heat transfer pressurized Heat transfer multiple zone Heat transfer vaporizer
Synthetic oil
10e290
20.4e81.9
Synthetic oil
10e330
30.7e1365
Synthetic oil
10e315
30.7e2047
Synthetic oil
10e400
30.7e2047
SylthermÒ 800
40e400
30.7e2047
Synthetic oil
10e400
30.7e4094
30 to 400
51.2e1024
COS PFC CLD CLS OMHTS CHTV
Ò
Dowtherm TherminolÒ
a Mbh is the ASME & ANSI standard for one thousand British Thermal Units per hour.
frequently occur in industrial sites. Design and operating problems of hot oil generators have been the focus of attention on the part of manufacturers and process engineers. Mitra [10] represented the design considerations of hot oil systems and sizing of equipment. Mukherjee [11] addressed some techniques for improving the performance of the individual components in hot oil systems. Arnold [12] developed a differential equation model to do a detailed thermal analysis for a hot oil system in different temperature ranges. Ennis [13] discussed some of the key safety design and operational aspects of hot oil systems. Probert [14] presented a method for designing and rating of hot oil storage tanks. Halttunen [15] addressed a method to analysis energy costs of hot oil pumping systems. Wallace et al. [16] designed a new method for hot oil integration with a heat recovery system generator. Policastro [17] suggested an advanced, flexible control system for hot oil plants. Colaco and Floyd [18] developed a computerized method to do control and analysis of conditions of a hot oil system on a dynamic basis. Gu and Liu [19] presented an analysis on the flow process of hot oil in the organic heat transfer material heater based on finite time thermodynamics. Hlozek and Bardov [20] designed a hot oil system to do waste heat recovery from reciprocating gas engines. Nasir et al. [21] designed a hot oil system to utilize the turbine waste heat to generate electric power in a Neptune plant. Mostajeran Goortani et al. [22] introduced a hot oil system to recover the heat from stack gases and distribute it to the appropriate cold streams in a Kraft mill plant. Ohm et al. [23] described a drying technique for the upgrading of crushed low-rank coal utilizing a hot oil system. Singhmaneeskulchai et al. [24] studied on dynamic data reconciliation of a utility heat exchanger using hot oil from a waste heat recovery unit as a hot stream to heat up ethane product as a cold process stream from a natural gas separation plant. However, to date, research on heating systems have focused on the individual components [10], and not on the system as a whole. Because of the interactions between hot oil networks and hot oil generator performance, all of the heating system components should be considered when designing and operating these systems. Process integration method can be applied to address the interactions between the components of hot oil systems. Process Integration is a family of methodologies for combining several parts of processes or whole processes to reduce the consumption of resources [25]. This methodology examines the potential for improving and optimizing the heat exchange between heat sources and sinks in order to conserve energy, reduce costs and emissions [25,26]. Klemes et al. [25] represented the recent developments in process integration.
Various process integration approaches have been developed over the past two decades to study energy systems with the goal of identifying energy recovery opportunities in process industries. Thermal pinch analysis, which was developed for the analysis and optimization of heat exchanger networks, seems to be the most commonly used heat integration method [27]. Thermal pinch analysis was introduced by Linnhoff and Flower [28]. This analysis was expanded and widely publicized by Linnhoff and Hindmarsh [29], Smith [30,31], Kemp [32] and Gundersen [33].The mass pinch analysis was introduced by El-Halwagi and Manousiouthakis [34]. This analysis was applied in the area of wastewater minimization by Wang and Smith [35,36]. This method was widely expanded by Mann and Liu [37], Prakash and Shenoy [38] and Wan Alwi and Manan [39]. The mass and thermal pinch analyses were combined together by Savulescu et al. [40], Panjeshahi et al. [41,42], Ahmetovíc and Kravanja [43,44] to identify opportunities to simultaneously reduce the water and energy consumption in industrial facilities. Consider some of the possible changes to an existing hot oil system. A new heat exchanger might be introduced into the heat exchanger network, or the heat duty of heaters changed, or process changes might change the operating conditions. These process changes influence the conditions of the hot oil return and, consequently, affect the performance of the hot oil generator. In such situations, it is often not clear how the system will be affected by the new conditions or how the hot oil network design will affect the heating system. As such, a combined thermal and mass pinch analysis should be used to investigate the interactions for the overall system. In this paper, we present a systematic method for the design of hot oil systems that accounts for these interactions and process constraints, based on a combination of thermal and mass pinch analyses. First, the hot oil generator and the hot oil network will be examined separately, in order to discuss the nature of hot oil system design. A model of hot oil systems will be developed to examine the performance of the hot oil generator to recirculation flow rate and return temperature, as well as to predict heating efficiency. A methodology for hot oil network design will then be developed, assuming fixed inlet and outlet conditions for the hot oil. Finally, the design of the overall hot oil system will be developed by investigating the interactions between the hot oil network design and the hot oil generator performance [45]. Debottlenecking procedures for the design of hot oil systems will also be developed. In summary, moving from parallel to series arrangements for hot oil networks: Increases the efficiency of the hot oil system, Decreases the temperature differences in the hot oil heat exchangers.
2. Design of hot oil networks The current practice for hot oil network design most often uses parallel configurations [4,10,46], whereby hot oil is supplied directly to individual heat exchangers. After the hot oil has been used in each heat exchanger, the cold oil returns to the hot oil generator (Fig. 1a). The minimum hot oil demand is determined by minimizing the flow rate to the individual heat exchangers (Fig. 1b). Under a parallel arrangement [47], return hot oil flow rate and temperature become maximized, leading to poor generator performance [45]. Thus far, no systematic methods have been suggested for dealing with the design of hot oil networks, and the traditional parallel design is not flexible when dealing with various process
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
425
Fig. 1. Parallel configuration of hot oil networks.
restrictions. Therefore, a new hot oil network design methodology will now be developed. Not all heating duties require hot oil at the hot oil supply temperature, which allows us, if appropriate, to change the hot oil network from a parallel to a series design. A series arrangement, in which hot oil is reused in the network, returns the hot oil at a lower temperature and flow rate. From the predictions of the model, the heat supplied by this hot oil generator can be expected to increase under these conditions. In other words, if the design configuration is converted from parallel to series arrangements [48], the hot oil generator can service a higher heat load for the heaters. 2.1. New design methodology for hot oil networks A simple problem (Example 1) will be used to develop the design methodology for hot oil networks. The hot oil system in Example 1 has four heat exchangers that use hot oil as the heating medium for cold process streams. The temperature, flow rate, and heating duty of the cold process streams are shown in Table 2. The operating data was given from an actual plant. The data for cold process streams are represented as CP values, which is the product of specific heat capacity and flow rate. It is assumed that the heat capacity of hot oil is constant throughout the temperature range. The parallel configuration hot oil network in Example 1 has inlet and outlet CPs of 28.69 kW/ C and outlet temperature of 181.46 C. The non-pressurized synthetic oil supplied by ChromaloxÒ [6] is selected for this plant. According to Table 1 and Fig. 2, the model of that type of hot oil is COS [6]. The maximum operating oil temperature for this type of hot oil is 330 C. Therefore, the maximum hot oil inlet temperature is 320 C, taking into account the necessity of a 10 C safety margin. To develop a systematic method for the design of such systems, some clues can be taken from water pinch analysis [49] and developed for hot oil network design. In hot oil network analysis, it is assumed that any hot oil-using operation can be represented as a counter-current heat exchange operation with a minimum temperature difference (Fig. 3). As seen in Fig. 3, the hot oil inlet temperature to each heat exchanger should be between the maximum Table 2 Cold process stream data of hot oil networks: Example 1. Heat exchanger
Inlet stream temperature ( C)
Outlet stream temperature ( C)
CP (kW/ C)
Total load (kW)
1 2 3 4
230 230 95 95
310 250 250 170
10 25 10 15
800 500 1550 1125
hot oil supply temperature (320 C) and the temperature of the outlet cold process stream from that heat exchanger plus DTmin. The hot oil outlet temperature from each heat exchanger must be between the temperature of the inlet cold process stream to each heat exchanger plus DTmin and the minimum hot oil return temperature. The concept of the limiting hot oil profile is taken from water pinch analysis and shown in Fig. 4; it is defined here as the minimum inlet and outlet temperatures for the hot oil stream. As seen in Fig. 4, the hot oil supply line touches the composite curve at 240 C where the pinch point is defined. It means the hot oil supplied by generator should be directly sent to the heat exchangers which need the oil hotter than 240 C (above the pinch) and the outlet oil streams may be reused in other heat exchangers (below the pinch). As seen, these allowable temperatures are limited by the minimum temperature difference (DTmin). In a new design, this could be the practical minimum temperature difference for a given type of heat exchanger. In retrofitting, the temperature difference could be chosen to comply with the performance limitations of an existing heat exchanger under revised operating conditions of reduced temperature differences and increased flow rate. In addition, the limiting hot oil profile might be determined by other process constraints, such as corrosion, fouling, maximum allowable oil temperature, etc. Any hot oil line at or above this profile is considered a feasible design. The limiting profile is used to define a boundary between feasible and infeasible regions. The limiting hot oil profile allows the individual streams of the hot oil network to be represented on a common basis, as hot oil and energy characteristics are represented simultaneously. It should be emphasized that the final design will not necessarily feature the minimum temperature difference incorporated in the limiting data. It simply represents a boundary between feasible and infeasible conditions. Most heaters in the final network design will feature temperaturedriving forces greater than those used for the specification of limiting conditions. This study focuses primarily on retrofit design and, therefore, restricts consideration to deal only with cold streams to be heated by hot oil. A better design for hot oil networks will be exploited under a fixed heat exchanger network configuration. For a grassroot design, the design of hot oil networks and heat exchanger networks should be addressed simultaneously. In this paper, the topology of the heat exchanger network is assumed to be fixed. The duties of the hot and cold streams in the heat exchanger network are, thus, assumed to be unrelated to the heating system. In other words, the streams heated by the hot oil do not affect other streams in the heat exchanger network. As the inlet temperature of hot oil to heaters is decreased, the driving force for the heat exchangers is decreased, and additional heat exchanger area might be required. At the same time, however,
426
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 2. Hot oil and steam range.
Fig. 3. Representation of heat exchangers using hot oil.
the flow rate is increased. Therefore, the reduction in driving force from decreasing temperature difference is compensated by increased hot oil flow rate. The limiting hot oil data shown in Example 1 have been extracted from the cold process stream data and provided in Table 3. A “hot oil composite curve” can be constructed by combining all individual profiles into a single curve within temperature intervals (Fig. 5a). The design of the hot oil network is based on the hot oil composite curve, which represents overall limiting conditions of the whole network. The hot oil supply line is a straight line matched against the hot oil composite curve to represent the overall hot oil flow rate and conditions. Minimizing the outlet temperature of the hot oil supply line minimizes the flow rate of the hot oil by maximizing hot oil reuse (Fig. 5b). Each point
where the supply line touches the composite curve creates a pinch in the design. It is important to note that the interpretation of the pinch does not imply a zero driving force of heat transfer, but a minimum driving force. Only the parts of the design where the supply line touches the composite curve will feature minimum driving forces; all other parts will feature temperature differences above minimum. The water main method [41] for the design of water reuse networks can be extended to the design of hot oil networks. The original method identified water reuse opportunities for problems wherein reuse was constrained by concentration limits. To achieve the minimum hot oil flow rate target in design, the problem to be decomposed into design regions, as illustrated in Fig. 6. This shows the hot oil composite curve from Fig. 5 decomposed into two regions: above and below the pinch. Above the pinch, from 320 to 240 C, the system requires the full flow rate of 18.92 kg/s. However, below the pinch, a lower flow rate will solve the problem. Fig. 6 shows a steeper line drawn against the part of
Table 3 Limiting hot oil data: Example 1.a Heat Inlet stream Outlet stream CP (kW/ C) Total load (kW) exchanger temperature ( C) temperature ( C) 1 2 3 4 Fig. 4. Strategy for minimizing the flow rate of heating oil.
a
320 260 260 180
240 240 105 105
10 25 10 15
DTmin ¼ 10 C; hot oil inlet temperature ¼ 320 C.
800 500 1550 1125
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
427
Fig. 5. Hot oil composite curve and targeting for maximum reuse.
the problem below the pinch. Of the CP of 18.92 kW/ C from the hot oil, after use up to the pinch condition of 240 C, only a CP of 18.33 kW/ C is used, and the balance (CP ¼ 0.59 kW/ C) is returned directly to the hot oil furnace. In this way, each part of the problem only uses the minimum amount of hot oil. However, it should again be noted that, in minimizing the amount of hot oil used, the temperature differences in the heater are also minimized, leading to tradeoffs between the designs of the hot oil system and the heat exchangers. As shown in Fig. 6, three hot oil mains are conceptualized: one at the supply temperature of 320 C, a second at the pinch temperature of 240 C, and a third at 105 C, the minimum temperature allowable in the system. The flow rate of hot oil required in each of the mains is shown at the top of the main. The flow rate to be returned to the hot oil furnace from each of the mains is shown at the bottom of the main. The heating streams are superimposed onto the hot oil mains at their appropriate temperatures. In Fig. 6, Stream 1, which starts at 320 C and terminates at 240 C, is shown between the 320 C and 240 C hot oil mains. Stream 2 starts at 260 C, between the first two hot oil mains. Stream 3 starts at 260 C, again between the first two hot oil mains. However, in this case, Stream 3 terminates at the minimum temperature of 105 C. Initially, therefore, it is broken into two parts, each within the appropriate design region. Finally, Stream 4 starts at 180 C below the pinch oil mains and finishes at the105 C hot oil main. The streams are then connected to the appropriate hot oil main, to satisfy the individual heating requirement. This provides an initial design for the hot oil network to meet the target requirements. However, the design has not yet been completed, as there is a fundamental difficulty with the arrangement shown in Fig. 6. Stream 3 requires a change in flow rate at the pinch hot oil main
Fig. 6. Completed hot oil design grid.
(240 C), which means that the heating duty corresponding with Stream 3 would have to be broken down into two heat exchangers, each being supplied with different flow rates. The change in flow rate for Stream 3 is, in fact, easily removed. Consider Fig. 7, which shows the temperature vs. heat duty for an operation with a change in flow rate, as in the case of Stream 3 in Fig. 6. A heat balance around Part 1 in Fig. 7 gives (assuming the specific heat capacity of the hot oil to be constant):
F1 ðTPinch T1 Þ ¼ F2 TPinch Tin;Max
(1)
Moving the mixing junction to the inlet of the operation, as shown in Fig. 7, and carrying out a heat balance at the new mixing junction, gives [42]:
Tin ¼
ððF2 F1 Þ*TPinch Þ þ ðF1 *T1 Þ T ðF1 *ðTPinch T1 ÞÞ ¼ Pinch F2 F2 (2)
Substituting Equation (1) into Equation (2) gives:
Tin ¼
ðF2 *TPinch Þ ðF2 *ðTPinch Tin:Max ÞÞ ¼ Tin;min F2
(3)
In other words, if the mixing junction is moved from the middle to the beginning of the operation, there is a constant flow rate throughout the operation that corresponds with the minimum inlet temperature after mixing. Fig. 8 shows the corresponding grid diagram to correct the change in flow rate. The change in flow rate for Stream 3 that previously occurred at the pinch temperature mains is now added from the pinch temperature mains to the inlet of Stream 3. This provides a design for the hot oil network that achieves the target minimum flow rate of CP ¼ 18.92 kW/ C. The arrangement shown in Fig. 8 involves reuse of hot oil from Streams 1 and 2 into Streams 3 and 4 via a hot oil main at pinch temperature 240 C. An alternative way to arrange the design is to make the connection directly, rather than through intermediate hot oil main. If the intermediate hot oil main is removed, then there are basically two sources of hot oil from Streams 1 and 2 at 240 C and two sinks for hot oil in Streams 3 and 4 at 240 C. These sources and sinks can be connected together in different ways. If the intermediate hot oil main at 240 C is removed from the design, then the streams can be connected directly together. Fig. 9 shows a flow sheet for one possible arrangement. In Fig. 9, there is reuse from Heat Exchanger 1 to Heat Exchanger 3 and from Heat Exchanger 2 to Heat Exchanger 4. Some of the hot oil from the furnace goes through Heat Exchanger 1, and some bypasses it, to be mixed in before entering Heat Exchanger 3. This step is necessary to comply with the inlet constraints for Heat Exchanger 3. The flow sheet in Fig. 9, then, needs to be assessed for its practicality and operability; the
428
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 7. Changing the mixing arrangement can avoid a change in hot oil flow rate.
design can then be evolved. For example, the bypass around Heat Exchanger 1 in Fig. 9 can be eliminated, and all of the CP ¼ 12.67 kW/ C can be put through Heat Exchanger 1 before entering Heat Exchanger 3. In the design, the essential feature for achieving the target is the reuse of hot oil between exchangers. Different arrangements than the one shown in Fig. 10 are possible for achieving the target by connecting the sources and sinks at 240 C differently. Parallel hot oil use that is minimized to achieve DTmin at the outlet of each heat exchanger leads to a flow rate for the network of 9.66 kg/s, with a hot oil return temperature of 181.3 C. Maximizing the reuse reduces this flow rate to 6.39 kg/s, and the hot oil return temperature decreases to 110 C (Table 4). A number of complications are likely to be encountered, which need to be addressed:
hot oil composite curve, as illustrated in Fig. 10. If more design regions exist, additional hot oil mains are required. However, the procedure is, basically, the same as in the example described above, taking into account the additional mains. b) Furthermore, if the hot oil supply line does not correspond with the minimum flow rate because of system interactions or temperature constraints for dew point 130 C (assuming DT ¼ 30 C for flue gas and hot oil, exhaust temperature of flue gas will be 160 C), then a pinch point is not created with the limiting hot oil composite curve, no pinch means that the temperature driving force is negative and is not acceptable (Fig. 11a). The setting could be between minimum flow rate (maximum reuse) and no reuse (parallel arrangement), as shown in Fig. 11a. The oil
a) One complication that can occur is that the number of design regions can be greater than two. Fig. 10 shows a design problem involving just two design regions: below the pinch and above the pinch. The hot oil composite curve in Fig. 10 has three design regions, rather than two. The design regions are identified by drawing straight lines between the extreme convex points of the Fig. 9. Flow sheet of hot oil network with maximum reuse of hot oil.
Fig. 8. Final grid design after changing the mixing arrangement to avoid flow rate changes.
Fig. 10. More complex problems might require a greater number of hot oil mains.
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
429
Table 4 Comparison of exit conditions of hot oil networks. Method
Flow rate (kg/s)
CP (kW/ C)
THot
Parallel Max. reuse %
10.58 6.98 34
28.68 18.92 34
181.5 110 39
oil; out
( C)
main method is based on the concept of the pinch point and cannot be applied to problems without a pinch. The new design methodology should provide for hot oil networks without a pinch. The limiting profile represents the boundary between feasible and infeasible operation. In other words, any composite curve above the original one is feasible. Thus, the hot oil composite curve can be modified in the feasible region without creating feasibility
Fig. 12. Finding a new pinch point: Example 1.
problems (Fig. 11b). If the hot oil composite curve could be modified to make a pinch point with the desired hot oil supply line in the feasible region, the hot oil network problem would be changed into a problem with a pinch. Pinch migration is introduced here to
Fig. 11. Hot oil composite curve development.
430
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 13. Pinch migration and temperature shift: Example 1.
convert problems without a pinch into those with a pinch with the desired supply line (Fig. 11b). Two approaches to pinch migration can be adopted (Fig. 11c). The first is to shift heat load, in which the hot oil composite curve moves along the heat load axis. The second is temperature shift, in which the hot oil composite curve moves along the temperature axis. In the two approaches, temperature shift is adopted, because heat load shift would result in an energy penalty. The next problem is how to find the new pinch and how to modify the composite curve with a temperature shift. Let us introduce a target temperature of 130 C for the hot oil for Example 1. A new pinch is created between the modified composite curve and the new supply line, which is calculated from a simple heat balance (Fig. 12). The new calculated pinch of 248.3 C is migrated from the original pinch of 240 C. It is necessary for individual duties to apply a temperature shift for modification of the composite curve. Hot oil Streams 1, 2, and 3 take part in creating the original pinch, which means Streams 1, 2, and 3 are the candidates for temperature shift. The limiting hot oil modifications occur in two stages. The first stage is to shift the temperature of the limiting hot oil profiles according to the value of the temperature shift (8.3 C in this example). Modified profiles might cross the supply line, thus requiring another step. The second stage is to increase the flow rate of the limiting hot oil profile when the shifted profile is restricted by temperature limitations. The limiting hot oil profile is modified to satisfy temperature limitations by increasing the hot oil flow rate CP. In Example 1, Streams 2 and 3 can be modified to obtain new limiting hot oil data simply by shifting temperatures. However, for Stream 1, it is necessary to increase flow rate, because the 20 C hot oil supply temperature restricts the temperature shift of the
limiting data. The heat balance equations determine the increased flow rate and the new limiting exit temperature [31].
CPnew ¼ CPold
new THO;in
old
¼ CP
old T old TPinch HO;in
! (4)
new T new TPinch HO;in
old old THO;out THO;in
!
CPnew
After modification of the conditions for each individual heat exchanger, the modified hot oil profiles are as shown in Fig. 13, and the new limiting hot oil data are as shown in Table 5. For Stream 1, the CP is increased from 10 to 11.18 kW/ C, as a result of the second stage modification. The new composite curve is constructed by combining all modified limiting profiles. The modified hot oil composite now creates a pinch point with the desired hot oil supply line. The hot oil network design can now be carried out using the hot oil mains method. The resulting design is shown in Fig. 14. The hot oil generator return temperature and flow rate agree with the target. The pinch migration and temperature shift method enables creation of a design with any target temperature. The hot oil network will have different configurations with different target temperatures, which can be seen by comparing the maximum reuse design (Fig. 9) with the design with a temperature constraint (Fig. 14). The pumping load and operating cost of the base and new design method are given in Table 6.
Table 5 Temperature-shifted limiting hot oil data. Heat Inlet stream Outlet stream CP (kW/ C) Total load (kW) exchanger temperature ( C) temperature ( C) 1a 2a 3a 4 a
320 268.3 268.3 180
Modified data.
248.3 248.3 113.3 105
11.15 25 10 15
(5)
800 500 1550 1125 Fig. 14. Hot oil network design without a pinch.
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434 Table 6 eThe grassroot project results.
431
Table 9 Hot oil outlet conditions of parallel design.
Parameters
Base
New
%
Parameters
Base
New
%
Pump load (kW) Gas exhaust CP require (kW/ C) Operation cost ($/yr)
4999 13.72 1,094,965
4036 11.96 863,863
19 13 21
Outlet stream temperature ( C) Outlet CP (kW/ C) Heat load (kW)
209.13 34.50 3825
201.24 38.94 4625
3.92 12.88 17.29
Table 7 Limiting hot oil data of base case.a Heat exchanger
Inlet stream temperature ( C)
Outlet stream temperature ( C)
CP (kW/ C)
Total load (kW)
1 2 3
320 245 180
245 200 120
15 40 15
1125 1800 900
a
DTmin ¼ 10 C; hot oil inlet temperature ¼ 320 C.
3. Debottlenecking of hot oil systems When hot oil networks need to increase the heat load of individual heaters or a new heat exchanger is introduced into an existing system, hot oil systems can become bottlenecked [31]. As the increase in heating load influences the hot oil generator performance, and there are interactions between hot oil networks and the hot oil generator, the best solution is often obtained by modifying the network. 3.1. General considerations for debottlenecking From the previous results of hot oil generator modeling and hot oil network design, a general guideline for the design of hot oil systems can be suggested. The heat supplied by the generator can be increased by changing inlet hot oil conditions from high flow rate and temperature to low flow rate and temperature. In general, doing so will require changing the hot oil network design from parallel to series or mixed parallel/series arrangements, with reuse of hot oil, decreasing the flow rate of hot oil and the return temperature. By changing from parallel arrangements to hot oil reuse designs, the heat supplied by the generator can be increased without any energy penalty and without investment in a new hot oil generator. A debottlenecking procedure [31,35] for hot oil systems will now be developed, using Example 2. The base case for Example 2 has three existing heat exchangers. The limiting hot oil data are provided in Table 7. In this example, a new heat exchanger (Table 8) is introduced into the base case, which makes the hot oil system bottlenecked. New outlet conditions of the hot oil when the parallel arrangement is retained with the new heat exchanger are shown in Table 9. The flow rate and heat load of the return oil to the generator increase, although the temperature decreases, thereby influencing the performance of the hot oil generator. Fig. 15 shows the performance of the parallel arrangement. First, the hot oil inlet temperature (Tin ¼ 310 C) to the network is colder than the desired
inlet temperature (320 C), which means that additional heating equipment needs to be installed to heat the hot oil to the minimum permissible inlet temperature (320 C). Second, the heat load of the network (4.625 MW) is bigger than the heat supply of the hot oil generator (4.235 MW), which also means that another 0.389 MW of heat load needs to be dissipated by additional heating. When the traditional parallel arrangements with new operating conditions are applied, the hot oil flow rate and the heat load of the generator are consequently increased. If there are no other design options than parallel arrangements, an additional hot oil generator (or supplemental firing) is needed to satisfy the new bottlenecked conditions.
3.2. Debottlenecking design procedures of hot oil systems A design procedure for debottlenecking hot oil systems will now be developed. The hot oil composite curve can first be constructed from the limiting hot oil data. The hot oil network performance can be changed within a feasible region that is bounded by the maximum reuse supply line and the parallel design supply line (Fig. 16a). In Fig. 16b, the feasible hot oil supply line (line AB) represents the attainable outlet conditions from the hot oil generator model by changing design configurations. As the inlet conditions to the generator affect its performance, it is necessary to know how the inlet conditions affect the hot oil generator system [50]. The hot oil supply line has the same heat load (4.625 MW) from the viewpoint of the hot oil network (Fig. 16a). However, the heat supplied by the hot oil system changes as inlet conditions to the hot oil generator change (Table 10). The heat supplied by the generator systems increases as the design configuration changes from parallel to maximum reuse (A to B in Fig. 16b). For our example, the following conditions should be satisfied for the new hot oil network design:
Table 8 Limiting hot oil data for new heat exchanger. Heat exchanger
Inlet stream temperature ( C)
Outlet stream temperature ( C)
CP (kW/ C)
Total load (kW)
1 2 3 4
320 245 180 220
245 200 120 140
15 40 15 10
1125 1800 900 800
Fig. 15. Changes in hot oil systems with parallel arrangements.
432
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Fig. 16. Hot oil supply line targeting.
(1) The inlet temperature to the hot oil network should be 320 C. (2) The heat supplied by the hot oil generator system should be equal to the heat load of the hot oil network. In this example, it is not necessary to achieve a temperature higher than 320 C. As can be seen in Table 8, the target conditions lie somewhere along the feasible hot oil supply line. The next stage is to find the target supply conditions for the hot oil generator. The feasible hot oil supply line can move from BN to BM in Fig. 16b. The target conditions, which satisfy the desired temperature to hot oil network (320 C), are found by changing the hot oil supply conditions from BN to BM. The heat removal of the heating system is the same as the heat load of the hot oil network at the target conditions (B*), where the inlet temperature to the hot oil network is satisfied. Target conditions are given by the intersection
Table 10 -Effects of hot oil inlet conditions. Case
Heat removal case of hot oil system (MW)
Parallel (A) Maximum reuse (B) Target
4.235 4.685 4.625
between the feasible hot oil supply line and the isothermal line of the heating system outlet temperature. The target conditions for debottlenecking have been found by using the heating system model. The target conditions are a CP of 28.3 kW/ C and a temperature of 156 C. At target conditions, the heating demands of the networks are satisfied without additional heating capacity. Below the target temperature, the current heating systems cannot operate up to heating demand. The next stage is to design the hot oil network with target conditions. As the new hot oil supply line has no pinch with the limiting composite curve, the temperature shift and pinch migration method is applied, as explained in the previous section. The new pinch point is calculated, and the limiting hot oil profile is modified (Fig. 17). The final design for the debottlenecked hot oil system is shown in Fig. 18. The suggested method for hot oil network design is based on a conceptual design methodology and, therefore, other design configurations can evolve. The design complexity can be reduced for the sake of simplicity, but this would likely result in a heating system performance penalty. The design of hot oil systems involves tradeoffs, including hot oil generator costs, pressure drops, piping costs, design complexity, etc. An optimization method is required to make the tradeoffs in a structured way; this issue will be the objective of future work. The proposed debottlenecking procedure enables the hot oil generator to manage the increased heat load by changing the
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
433
Fig. 17. Pinch migration and temperature shift: Example 2.
However, the CP value of the common line for the parallel design is greater than that of the reuse design (Table 11). Therefore, those modifications should carefully be taken into account in practical debottlenecking projects. 4. Conclusions Fig. 18. Final design of debottlenecked hot oil systems.
network design from parallel to series arrangements. The design method targets the hot oil generator conditions and then designs the hot oil network for the new target conditions. The design procedure for debottlenecking hot oil systems can be summarized as follows: (1) Define the feasible hot oil supply line from the composite curve and parallel supply line. (2) Target the hot oil generator supply conditions from the heating systems model and the feasible hot oil supply line. (3) Design the hot oil network for the target conditions with pinch migration and temperature shifting. The flow rate to the individual heat exchangers is likely to change; therefore, the design of individual heat exchangers needs to be checked to ensure that it is feasible. In addition, the procedure changes the conditions of the return hot oil and the recirculation hot oil flow rate. As a result, the pressure drop in the equipment and piping will also change. Therefore, the performance of the hot oil pumps needs to be checked, because the pumping head, efficiency, and required power depend on the flow rate carried by the pump. When the design (Fig. 18) is compared with parallel arrangements, the CP value of the individual heat exchangers in the parallel design is the same as or less than that of the reuse design (Table 11). Table 11 CP vs. hot oil network design. Heat exchanger
CP (kW/ C) Parallel (no reuse)
Target (reuse)
1 2 3 4 Total Common linea
15 15 4.5 4.44 38.9 38.9
17.09 23.76 10.00 10.08 60.95 28.20
%
13.98 58.42 122.33 127.01 56.51 27.58
a Common line refers to the hot oil pipeline between the hot oil generator and the hot oil network.
A new methodology for the design of hot oil networks has been developed, in order to satisfy any supply conditions for the hot oil generator. The design can be carried out with any target temperature by introducing the concepts of pinch migration and temperature shift. From the interactions between the performance of the hot oil generator and the design of the heaters, the proposed debottlenecking procedures allow increased capacity without investment in new hot oil generator equipment when generator capacity is limited. The heat load distribution of heating systems has also been considered when a hot oil system is bottlenecked beyond generator capacity or when a temperature constraint limits the return temperature. A number of design options for debottlenecking heating systems have been discussed, in order to improve hot oil generator performance and to distribute heat load between the generator and other design options. Acknowledgements The financial support of the Iran Energy Efficiency Organization (IEEO), (SABA) is gratefully acknowledged. Nomenclature B* targeted condition CP heat capacity multiplied by flow rate, (kW/ C) DTshift amount of temperature shift, ( C) DTmin minimum temperature approach, ( C) F flow rate F1 outlet hot oil flow rate of exhaust heat exchanger, (t/h) F2 inlet hot oil flow rate of exhaust heat exchanger, (t/h) F3 hot oil flow rate flowing into the hot oil network, (t/h) H enthalpy, (kJ/kg) Q heat load, (kW) T temperature, ( C) T* migrated pinch temperature, ( C) T1 outlet hot oil temperature of exhaust heat exchanger, ( C) T2 inlet hot oil temperature of exhaust heat exchanger, ( C) T3 hot oil temperature flowing into the hot oil network, ( C)
434
A. Ataei et al. / Applied Thermal Engineering 66 (2014) 423e434
Subscripts Cold cold process stream in heat exchanger in inlet conditions Max maximum Min minimum out outlet conditions P pinch point Superscripts HE process/hot oil heat exchanger HEN heat exchanger network HO hot oil New migrated pinch point Old original pinch point References
[22]
[23]
[24]
[25] [26] [27]
[28] [29]
[1] A. Ataei, K.S. Lee, J.J. Lim, M.J. Kim, H.B. Liu, O.Y. Kang, T.S. Oh, C.K. Yoo, A review on environmental process engineering, Int. J. Environ. Res. 5 (4) (2011) 875e890. [2] A. Ataei, Application of combined pinch and exergy analysis in retrofit of an olefin plant for energy conservation, Sci. Res. Essays 6 (12) (2011) 2437e2446. [3] A. Ataei, C.K. Yoo, Combined pinch and exergy analysis for energy efficiency optimization in a steam power plant, Int. J. Phys. Sci. 5 (7) (2010) 1110e1123. [4] Sigma Thermal, Thermal Oil Heating Systems, 2013. www.sigmathermal.com (accessed 12.12.13). [5] Gaumer Process, Gaumer Hot Oil Systems, 2013. www.gaumer.com (accessed 12.12.13). [6] Chromalox, Hot Oil Heaters e Heat Transfer Systems, 2013. www.chromalox. com (accessed 12.12.13). [7] Thermal Fluid Systems, Hot Oil Heating Systems, 2013. www.tfsheat.com (accessed 12.12.13). [8] GTS Energy, Thermal Fluid Heaters, 2013. www.gtsenergy.com (accessed 12.12.13). [9] A. Ataei, C.K. Yoo, Application of pinch technology for waste heat recovery in the linear-alkyl-benzene sulfonation process, Asian J. Chem. 23 (4) (2011) 1539e1547. [10] S. Mitra, Design Considerations of Hot Oil System, 2013. www.academia.edu/ 1514808/design_consideration_of_hot_oil_system (accessed 07.06.13). [11] R. Mukherjee, Role of optimal heat exchange in chemical plant energy conservation, Chem. Eng. World 19 (10) (1984) 47e49. [12] F.G. Arnold, Thermal analysis of hot oiling, Am. Soc. Mech. Eng. (1990) 4. PET1, Energy-Sources Technology Conference and Exhibition, New Orleans, LA, USA, 14 January 1990 through 18 January 1990. [13] T. Ennis, Safety in design of thermal fluid heat transfer systems, Inst. Chem. E 155 (2009) 162e169, 21st Institution of Chemical Engineers Symposium on Hazards 2009 e Hazards XXI: Process Safety and Environmental Protection, Manchester, United Kingdom, 10 November 2009 through 12 November 2009. [14] D. Probert, Design and performance of hot-oil storage tanks, Appl. Energy 1 (4) (1975) 247e278. [15] O.H. Halttunen, Analyzing energy costs of hot oil pumping systems e comparison of centrifugal, canned-motor, and seal-free magnetic drive pumps, Plant Eng. 32 (10) (1978) 155e161. [16] P.S. Wallace, K.A. Johnson, J.L. Kasbaum, Hot Oil Integration with Heat Recovery System Generator and Method of Operation, US Patent 6,606,862, 2003. [17] E.F. Policastro, Hot oil, cool control, InTech 55 (4) (2008) 24e25. [18] R.J. Colaco, M.C. Floyd, Supervisory operation and data system controls hot oil pipeline, Oil Gas J. 74 (9) (1976) 87e90. [19] W. Gu, Y. Liu, Analysis on the flow process of hot oil in the organic heat transfer material heater based on finite time thermodynamics, Adv. Mater. Res. 250e253 (2011) 2979e2983, 1st International Conference on Civil Engineering, Architecture and Building Materials, CEABM 2011, Haikou, China, 18 June 2011 through 20 June 2011. [20] R.J. Hlozek, J.B. Bardov, Waste heat recovery from reciprocating gas engines: glycol regeneration, in: Proceedings, Annual Convention e Gas Processors Association, 1980, pp. 100e104. Proc Annu Conv Gas Process Assoc Tech Pap 59th, Houston, TX, USA, 17 March 1980 through 19 March 1980. [21] P. Nasir, S. Jones, D. Schochet, D. Posner, Utilization of turbine waste heat to generate electric power at Neptune plant, in: GPA Annual Convention Proceedings, 2004, p. 9. Gas Processors Association e 83rd Annual Convention
[30] [31] [32]
[33]
[34] [35] [36] [37] [38]
[39]
[40] [41]
[42] [43] [44]
[45] [46]
[47]
[48] [49] [50]
Proceedings 2004: Linking Energy Partners for the Future, New Orleans, LA, United States, 14 March 2004 through 17 March 2004. B. Mostajeran Goortani, E. Mateos-Espejel, M. Moshkelani, J. Paris, Energy efficiency improvement of a Kraft process through practical stack gases heat recovery, Appl. Therm. Eng. 31 (2011) 4091e4096. T.I. Ohm, J.S. Chae, J.H. Lim, S.H. Moon, Evaluation of a hot oil immersion drying method for the upgrading of crushed low-rank coal, J. Mech. Sci. Technol. 26 (4) (2012) 1299e1303. P. Singhmaneeskulchai, N. Angsutorn, K. Siemanond, Dynamic data reconciliation in a hot-oil heat exchanger for validating energy consumption, Chem. Eng. Trans. 35 (2013) 493e498, http://dx.doi.org/10.3303/CET1335082. J.J. Klemes, P.S. Varbanov, Z. Kravanja, Recent developments in process integration, Chem. Eng. Res. Des. 91 (2013) 2037e2053. B. Bakhtiari, S. Bedard, Retrofitting heat exchanger networks using a modified network pinch approach, Appl. Therm. Eng. 51 (2013) 973e979. R.F. Dunn, M.M. El-Halwagi, Review process integration technology review: background and applications in the chemical process industry, J. Chem. Technol. Biotechnol. 78 (2003) 1011e1021, http://dx.doi.org/10.1002/ jctb.738. B. Linnhoff, J.R. Flower, Synthesis of heat exchanger networks: I. Systematic generation of energy optimal networks, AIChE J. 24 (4) (1978) 633e642. B. Linnhoff, E. Hindmarsh, The pinch design method for heat exchanger networks, Chem. Eng. Sci. 38 (1983) 745e763. R. Smith, Chemical Process Design, McGraw-Hill, New York, USA, 1995. R. Smith, Chemical Process Design and Integration, John Wiley & Sons, UK, 2005. I.C. Kemp, Pinch Analysis and Process Integration, a User Guide on Process Integration for Efficient Use of Energy, Elsevier, Amsterdam, The Netherlands, 2007. T. Gundersen, Heat Integration: Targets and Heat Exchanger Network Design (Chapter 4), in: J.J. Klemes (Ed.), Handbook of Process Integration (PI): Minimisation of Energy and Water Use, Waste Emissions, Woodhead Publishing, Cambridge, UK, 2013. M.M. El-Halwagi, V. Manousiouthakis, Synthesis of mass-exchange networks, AIChE J. 35 (1989) 1233e1244. http://dx.doi.org/10.1002/aic.690350802. Y.P. Wang, R. Smith, Wastewater minimisation, Chem. Eng. Sci. 49 (7) (1994) 981e1006. Y.P. Wang, R. Smith, Waste-water minimisation with flow-rate constraints, Chem. Eng. Res. Des. 73 (8) (1995) 889e904. J.G. Mann, Y.A. Liu, Industrial Water Reuse and Wastewater Minimization, McGraw Hill, New York, USA, 1999, ISBN 0071348557. R. Prakash, U.V. Shenoy, Targeting and design of water networks for fixed flowrate and fixed contaminant load operations, Chem. Eng. Sci. 60 (1) (2005) 255e268. S.R. Wan Alwi, Z.A. Manan, Water pinch analysis for water management and minimisation: an introduction, in: Klemes (Ed.), Process Integration Handbook, Woodhead Publishing, Cambridge, UK, 2013, ISBN 978 0 85709 593 0. http://dx.doi.org/10.1533/9780857097255.3.353. L.E. Savulescu, M. Sorin, R. Smith, Direct and indirect heat transfer in water network systems, Appl. Therm. Eng. 22 (8) (2002) 981e988. M.H. Panjeshahi, A. Ataei, M. Gharaie, R. Parand, Optimum design of cooling water systems for energy and water conservation, Chem. Eng. Res. Des. 87 (2) (2009) 200e209. M.H. Panjeshahi, M. Gharaie, A. Ataei, Debottlenecking procedure of effluent thermal treatment system, Energy 35 (2010) 5202e5208. E. Ahmetovíc, Z. Kravanja, Simultaneous synthesis of process water and heat exchanger networks, Energy 57 (2013) 236e250. E. Ahmetovíc, Z. Kravanja, Simultaneous optimization of heat-integrated water networks involving process-to-process streams for heat integration, Appl. Therm. Eng. 62 (1) (2014) 302e317. http://dx.doi.org/10.1016/j. applthermaleng.2013.06.010. B. Galan, I.E. Grossman, Optimal design of distributed wastewater treatment networks, Ind. Eng. Chem. Res. 37 (10) (1998) 4036e4048. A. Azhdari, H. Ghadamian, A. Ataei, C.K. Yoo, A new approach for optimization of combined heat and power generation in edible oil plants, J. Appl. Sci. 9 (21) (2009) 3813e3820. M.M. El-Halwagi, A.M. El-Halwagi, V. Manousiouthakis, Optimal design of dephenolization networks for petroleum-refinery wastes, Trans. IChemE (Part B) 70 (1992) 131e139. J.K. Kim, R. Smith, Cooling water system design, Chem. Eng. Sci. 56 (12) (2001) 3641e3658. W.J. Kuo, R. Smith, Effluent treatment system design, Chem. Eng. Sci. 52 (23) (1997) 4273e4290. C.K. Yoo, A. Ataei, Y.S. Kim, M.J. Kim, H.B. Liu, J.J. Lim, Environmental systems engineering: a state of the art review, Sci. Res. Essays 5 (17) (2010) 2341e 2357.