- Email: [email protected]
A new targeting method for estimation of cogeneration potential and total annualized cost in process industries
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Contents lists available at SciVerse ScienceDirect
Chemical Engineering Research and Design journal homepage: www.elsevier.com/locate/cherd
A new targeting method for estimation of cogeneration potential and total annualized cost in process industries M.H. Khoshgoftar Manesh a , S. Khamis Abadi b , M. Amidpour a,∗ , M.H. Hamedi a a
Energy Integration Research Center, Energy Systems Engineering Department, Mechanical Engineering Faculty, K.N. Toosi University of Technology, Tehran, Iran b Department of Energy Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
a b s t r a c t One of the important tasks for optimal design and analyses of site utility systems is targeting total annualized cost and cogeneration potential. This paper introduced a new cogeneration and total annualized cost (TAC) targeting models that were developed to estimate the cogeneration potential of site utility systems and total annualized cost. The procedure which was proposed here provided a consistent, general procedure for determining mass flowrates and efficiencies of the applied turbines. This algorithm utilized the relationship of the entropy with the enthalpy and the isentropic efficiency. It is considered superior to previous works in that it was accurate, did not require any cumbersome simulation for initiation and could be easily traced,which enhance its programmability, considering full lifetime of the utility prices and it can estimate TAC and TASP with diffrent cost functions. Also, the developed model based on the trends of historical prices has been considered for estimation of utility costs. In addition, developed graphical representations based on site utility grand composite curve (SUGCC) were introduced: (1) TAC-SUGCC, to illustrate on the same diagram heat recovery through the steam mains, shaft work production, TAC, temperature of steam mains and steam flow rate simultaneously; and (2) TASP-SUGCC, to demonstrate TASP on the extended SUGCC diagram. Finally, two case studies were used to illustrate the usefulness of the new method. © 2012 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. Keywords: Targeting; Total annualized cost; Cogeneration; Total site
1.
Introduction
Chemical processes usually require steam at different pressure and temperature values for heating and non-heating purposes. In order to provide steam in the required condition, the designer has to decide whether to provide steam in the extreme condition and then let it down to different levels or produce steams separately in different boilers. Many industrial processes operate within total sites (Raissi, 1994), where they are serviced and linked through a common central utility system. This utility system meets the demands for heat and power of individual process units by their indirect heat integration. However, greater benefits in terms of energy and capital cost can be obtained by looking at the entire site.
Details explanation of the total site analysis, grand composite curve and the total site profiles can be seen in Klemeˇs et al. (2010). Total site integration addresses the task of optimizing each process and utility system in the context of the overall site (Sorin and Hammache, 2005). A number of models have been proposed for the early estimation of cogeneration for utility systems using steam turbines. Dhole and Linnhoff (1993) proposed an exergetic model based on the site source–sink profiles. Raissi (1994) presented the T-H model based on the Salisbury approximation, assuming that power is linearly proportional to the difference between the inlet and outlet saturation temperatures. Mavromatis and Kokossis (1998) introduced the non-linear model of THM (turbine hardware model) based on the principle of the Willans’ line in order to incorporate variation of efficiency with turbine size and
∗
Corresponding author. Tel.: +98 21 88677272; fax: +98 21 88677272. E-mail addresses: [email protected] (M.H. Khoshgoftar Manesh), sajad [email protected] (S.K. [email protected] (M. Amidpour), [email protected] (M.H. Hamedi). Received 7 July 2012; Received in revised form 16 November 2012; Accepted 7 December 2012 0263-8762/$ – see front matter © 2012 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cherd.2012.12.002
Abadi),
1040
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Nomenclature h ˙ m P Q˙ s T TAC W C t op
Ctot op Ctot ft SUGCC GCC TASP
specific enthalpy (kJ/kg) mass flow rate (kg/s) pressure (bar) heat load (MW) specific entropy (kJ/kg K) temperature (◦ C) Total annualized cost Shaft work (MW) Costs index corresponding to the number operating periods overall operating plant cost specific costs of utility in time period fraction of time horizon corresponding to each period site utility grand composite curve grand composite curve total annualized sale of products
Greek symbol isentropic efficiency Subscripts is isentropic saturated conditions Sat tot total Superscripts DEM process steam demand GEN process steam generation i steam main net heat or mass load net Op operating costs
operating load. An advanced approach to these concepts, known as top-level analysis, is one that allows for “scoping”, i.e. selecting site processes to target for heat integration improvements proposed by Varbanov et al. (2004). Harell (2004) introduced a graphic technique for estimating the cogeneration potential which utilized the concept of extractable power and header efficiency to establish cogeneration potential. Varbanov et al. (2004) developed the improved turbine hardware model. Sorin and Hammache (2005) also developed an exergetic model based on thermodynamic insights for the Rankine cycle and showed that power was not linear in saturation temperature differences. Mohan and ElHalwagi (2007) presented a linear algebraic approach based on the concept of extractable power and steam main efficiency. Perry et al. (2008) extended the total site concept to a broader spectrum of processes in addition to the industrial process. A potential for the integration of renewable energy sources was introduced to reduce the carbon footprint of a locally integrated. ˜ (2010) proposed a modified Medina-Flores and Picón-Núnez thermodynamic model by keeping the advantages of the THM. Bandyopadhyay et al. (2010) developed a linear model based on the Salisbury approximation and energy balance at steam mains. A new shaft work targeting model, termed the iterative bottom-to-top model (IBTM), was presented by Ghannadzadeh
et al. (2011). Kapil et al. (2012) introduced a new method for estimating cogeneration potential of site utility systems by a combination of bottom-up and top-down procedures. Varbanov and Klemes (2011) proposed an extension of the total site methodology covering industrial, residential, service, business and agricultural customers and the incorporation of renewable energy sources (solar, wind, biomass, and some types of waste), accounting for the often substantial variability on the supply and demand sides and for the use of non-isothermal utilities. Matsuda et al. (2012) studied the heat recovery potential for a large steel plant using total site profile analysis. This is followed by an assessment of the potential benefit of reducing steam demands at various levels by successively optimizing the system in steps of steam demand reduction. This results in a set of curves for steam marginal prices for the system under consideration. Liew et al. (2012) introduced four new contributions: (1) total site sensitivity table (TSST), a tool for exploring the effects of plant shutdown or production changes on heat distribution and utility generation systems over a total site; (2) a new numerical tool for TSHI, the total site problem table algorithm (TS-PTA), which extended the well-established problem table algorithm (PTA) to total site analysis; (3) a simple new method for calculating multiple utility levels in both the PTA and TSPTA; and (4) the total site utility distribution (TSUD) table, which can be used to design a total site utility distribution network. Varbanov et al. (2012) proposed the total site heat recovery targeting using multiple Tmin specifications for the site processes and process-utility interfaces. It is also possible to define and use the Tmin contributions of individual process streams in a process (Kravanja and Glavic, 1997). Fodor et al. (2012) focused on extending traditional total site integration methodology to produce more meaningful utility and heat recovery targets for the process design. This methodology was a further development of a recently extended traditional pinch methodology. The previous extension was on the introduction of using an individual minimum temperature difference (Tmin ) for different processes so that the Tmin is more representative of the specific process. Further, it deals with stream specific Tmin inside each process by setting different T contribution (Tcont ) and also using different Tcont between the process streams and the utility systems. Mohammad Rozali et al. (2012) extended the pinch analysis concept used in Process Integration to determine the minimum electricity targets for systems comprising hybrid renewable energy sources. Power pinch analysis (PoPA) tools described graphical techniques to determine the minimum target for outsourced electricity and the amount of excess electricity for storage during start up and normal operations. Nemet et al. (2012a) proposed a deterministic and stochastic multi-period mixed-integer nonlinear programming (MINLP) models for heat exchanger network (HEN) synthesis to account for future price projections, where the utility cost coefficients are forecasted for the lifetime of the process. Hackl and Harvey (2012) demonstrated how heat integration tools such as total site analysis and exergy analysis can be applied to target for shaft work and hot utility savings for processes and utility systems operating below ambient temperature.
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Mohammad Rozali et al. (2012) extended the graphical and numerical power pinch analysis (PoPA) method by considering the energy losses that occur in the power system conversion, transfer and storage. These losses on the minimum outsourced electrical energy targets and storage capacity are evaluated and the storage cascade table (SCT) of PoPA is developed to include the effect of energy losses in the system design. Nemet et al. (2012a) introduced the capital cost targeting method for total site heat recovery to determine the lower bound on the heat transfer area for meeting the targeted heat recovery on the total site. In this paper, new method for estimation of cogeneration potential and TAC has been proposed. The pervious works cited have not high accuracy for estimation of cogeneration potentials. In this paper, accurate cogeneration targeting and cost estimation models have been proposed. The new method applied for targeting shaftwork, steam production and used with high accuracy, considering degree of superheat in, and estimate TAC and TASP. Also, different models for estimation of costs have been considered. In addition, new graphical representations have been developed to demonstrate the TAC and TASP based on modification on original SUGCC. The utility prices can fluctuate rather quickly and the operating costs may be very different from a year to year. In this regard, the full lifetime of the utility prices is considered in this work. So, the developed model based on the trends of historical prices has been considered for estimation of operating costs.
2.
Methodology
2.1.
Algorithm of proposed procedure
In this section, the new model was presented in detail to target the cogeneration potential for site utility systems. The method uses the site utility grand composite curve (SUGCC), which represents another form of the site composite curves (Klemes et al., 1997). The SUGCC was obtained from the site composite curves by being represented on temperature–enthalpy axes of each steam main by its saturation temperature and steam generation and usage loads, from the source and sinks profiles of the site composites. The differences between steam generation and steam usage set the VHP demand or the supply heat available at each main. The new model calculates the minimum required flow rate from a steam generation unit and the levels of superheat in each steam main based on the heat loads specified by SUGCC. Fig. 1 shows a schematic of the utility system layout on SUGCC diagram. The L given steam mains are indexed by i from the highest pressure steam main, meaning that i is equal to 1, 2, 3 and 4 for very high pressure (VHP), high pressure (HP), medium pressure (MP) and low pressure (LP) steam mains. There is an expansion zone between two steam mains. Each zone are indexed by Z starting from top, i.e. Z = 1 is for VHP-HP, and one single steam turbine is placed in each zone. Fig. 2 shows the thermodynamic expansion of steam at two different pressure levels on a temperature-entropy diagram. The step S1 − S2 shows an isentropic expansion. An isentropic process is an ideal case in which there is no kind of irreversibility, such as mechanical friction and heat losses. Step S1 − S2 is a better representation of what happens in reality. The outlet
1041
Fig. 1 – Schematic of the utility system layout on SUGCC. of the turbine is shifted to the right side which indicates the increase of entropy (state of disorder) caused by losses (ALAzri, 2008). The isentropic efficiency is basically the ratio of the enthalpy difference of step S1 − S2 to that of step S1 –S2 . The isentropic efficiency is a function of the load and, for fixed values of flow rates, it would be better to consider the highest efficiency, assuming to use turbines, for which the calculated flow rate will be the full load (AL-Azri, 2008). In the study by Varbanov et al. (2004), thermodynamic model was used to estimate the isentropic efficiency as follows: is =
Wmax Wis,max
Wis,max =
Wis,max − A B
(1)
where A and B are constants that depend on the turbine and are functions of the saturation temperature. A and B are calculated by Eqs. (2) and (3): A = b0 + b1 Tsat
(2)
B = b2 + b3 Tsat
(3)
The values of these constants are given in Table 1. At the boiler exit, for a given pressure and steam temperature, the enthalpy can be obtained with the aid of steam tables.
Fig. 2 – Thermodynamic expansion of steam at two different pressure levels on a temperature-entropy diagram.
1042
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Table 1 – The regression coefficients used in the isentropic efficiency equation. Back pressure turbines
Condensing turbines
Wmax ≤ 2000 kW
Wmax > 2000 kW
Wmax ≤ 2000 kW
Wmax > 2000 kW
0 1.08 1.097 0.00172
0 4.23 1.155 0.000538
0 0.662 1.191 0.000759
−463 3.53 1.22 0.000148
b0 (kW) b1 (kW/◦ C) b2 b3 (◦ C−1 )
The actual input enthalpy of steam mains are usually provided from the calculations of the previous steam mains. The input isentropic enthalpy of steam main can be obtained in the superheated region. Then, the efficiency is calculated. The actual enthalpy which will serve as the input enthalpy for the next zone is then calculated using the isentropic enthalpies and efficiency by Eq. (4). hi,
actual
= hi−l,
isentropic
− (hi−l,
isentropic
− hi,
isentropic )
(4)
In this study, the calculation of superheat temperature at each steam level was done using the iterative procedure based on a certain desirable amount of superheat in the LP steam main. This superheat was required to be set at 10–20 ◦ C (Smith, 2005). If the degree of superheat in the resulting LP steam main was less than the required level, then, operating conditions of VHP would be updated and iterated until the acceptable superheated conditions would be met for the LP steam main. The mass flow rate of steam expanding through the Z-th ˙ z ) can be calculated by the mass balance for i-th by turbine (m Eq. (5), as shown in Fig. 3: ˙z= m
˙ z−1 + m ˙ DEM m i
˙ GEN −m i
Step 1. Preparation of a model in SUGCC Step 2. Initial estimates of boiler superheat temperature Step 3. Finding initial estimates of mass flow rates passing by each zone, assuming isentropic expansions throughout the levels by Eq. (8)
mz,
initial
=
Q˙ net,i hi − hf,i
(8)
Q˙ net,i = Q˙ iDEM − Q˙ iGEN where Q˙ net,i is the net load at the given level, hi the steam main isentropic enthalpy at the given level, hf is the saturated liquid enthalpy at the given level Step 4. Correcting efficiency by Eq. (1) ˙ NET Step 5. Correcting hi and m for the given steam level i Step 6. Repeating the steps from the second iteration through convergence in a manner until they meet the stopping criteria (Eq. (9))
Z i=1
2
(mZ − mz,new ) ≤ ε
(9)
(5)
˙ GEN where m is the flow rate of steam generated by the process i DEM ˙i is the flow rate of steam demanded by the process, and m which can be calculated by Eqs. (6) and (7): ˙ DEM m = i
Q˙ iDEM Actual hi − hf,i
(6)
˙ GEN = m i
Q˙ iGEN Actual hi − hf,i
(7)
where hf,i is the enthalpy of the saturated liquid enthalpy at the pressure of i-th steam main. The procedure of cogeneration targeting for a given site utility system (Fig. 4) is presented as follows:
Step 7. Checking the superheat temperature when the first loop of algorithm terminates (if it falls below the allowed minimum, the superheat temperature of the boiler is increased and repeats the steps until meeting the desirable amount of superheat in the LP steam main) Step 8. Finding, steam temperatures and mass flows of boiler and each steam mains and efficiency Step 9. Finding the shaft work in each zone Step 10. Generation of final SUGCC Step 11. Finding operating parameters of site utility Step 12. Finding parameters of main components such as boiler and steam turbines Step 13. Calculation of total annualized operating cost in life time of site utility (Section 2.2.1) Step 14. Calculation of total annualized capital cost (Section 2.2.2) Step 15. Calculation of TAC (Section 2.2) Step 16. Generation of TAC-SUGCC and TASP-SUGCC representations Step 17. Targeting cogeneration potential, TAC and TASP
2.2.
Total annualized cost targeting
The total annualized cost is given by the sum of the operating costs over the relevant devices plus the sum of the annualized capital costs (Eq. (10)). TAC = Operation cost + Capital cost Fig. 3 – Mass load balance for i-th steam main.
where TAC is the total annualized cost ($/year)
(10)
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
1043
Fig. 4 – Algorithm of proposed cogeneration and TAC targeting method.
2.2.1.
Modeling of operating costs
The utility prices can fluctuate rather quickly and the estimation of operating costs may be very different from a year to year. It is highly unrealistic to expect constant prices for utilities in the future. Nemet et al. (2012b) considered the full lifetime of the HEN and future utility prices. In this paper, the developed model based on the trends of historical prices has been considered for estimation of operating costs, where the annual utility cost with varying prices is calculated for every period during the lifetime as:
op
Ctot =
op
t
tot (Ct · ft ) · top
(11)
Price projections are based on past prices as described by Ulrich and Vasudevan (2006). Also, other further data needed can be found in INDEX MUNDI (2012) and Chemical Engineering Plant Cost Index – CEPCI (2012).
2.2.2.
Eq. (12) represents a linear formulation to relate the purchase cost of a boiler to its design steam flow rate and pressure. PCbol = A × Mstm-D + B A = 0.249 × Pbol + 47.19
(12)
B = 3.29 × Pbol + 624.6 where PCbol is the boiler purchase cost (1000$) (2002), A and B are regression coefficients for the purchase function (1000$/(kg/s), 1000$), Pbol is the boiler steam design pressure (bar), Mstm-D the boiler design steam flow (kg/s). In this study two functions used to estimate the purchase cost of a steam turbines. First function is based on Aguilar cost functions that represents an equation to estimate the purchase cost of a steam turbine to its design steam flow rate and volumetric flow passing through the unit (Eq. (13)): PCst = 474.5 × Vol + 1111.7 √ Vol = Mst × vin × vout
(13)
Modeling of capital costs
The basis for calculating the overall capital expenditure of the plant is the purchase cost of the equipment. The most significant cost is assumed contributed by the turbine and the boiler. The purchase cost of a boiler mainly depends on its design steam flow rate and pressure, as these parameters define the amount and specifications for the materials of construction.
where PCst is the steam turbine purchase cost (without including electric generator) (1000$) (2002), Vol the average steam volumetric flow rate passing through turbine (m3 /s), Mst the design steam turbine mass flow rate (kg/s), vin is the specific volume of the steam entering the turbine (m3 /kg), vout the specific volume of the steam discharged by the turbine (m3 /kg).
1044
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Second function is based on AL-Azri (2008) equipment cost functions that represents an equation to estimate the purchase cost of a steam turbine to its design shaft power output (Eq. (14)). PCst = 475 × (W × 2.93E − 04)
0.45
(14)
where W is the turbine shaft power output (kW). Eq. (15) represents an equation to estimate the purchase cost of an electric generator to its design output. PCgen = 0.0414 × Wgen + 190.49
(15)
where PCgen is the purchase cost for electric generators (1000$), Wgen is the design output for electric generators (kW). Eq. (16) represents an equation to estimate the purchase cost of a deaerator (Aguilar, 2005). PCdea = 0.958 × Mdea + 30.0
(16)
where PCdea is the deaerator purchase cost (1000$) (2002), Mdea is the design water mass flow from the deaerator (kg/s). In addition, it is possible to update these functions employing the chemical engineering plant cost index (CEPCI) (2012) factor corresponding to the ratio of the actual index over the base year one. The capital costs of the component is converted to annualized cost by using the capital recovery factor CRF (i,n) (Eq. (17))
CRF =
i(1 + i)
Table 2 – Data parameters for case study 1.
Pressure (bar(a)) Saturation temperature (◦ C) Net heat load (MW)
4.
VHP
HP
MP
LP
120 324 0
50 264 50
14 195 40
3 133 85
The case study
To show the applicability of the new method in the total site analysis, two case studies were considered.
n
n
(1 + i) − 1.
(17)
where i is the interest rate, n is the specified plant life.
3.
Fig. 5 – Schematic of the utility system layout on modified SUGCC.
New graphical representation
Due to importance of estimation of total annualized cost and cogeneration production ahead of design, two new graphical representations based on site utility grand composite curve (SUGCC) were introduced: (1) TAC-SUGCC, to illustrate on the same diagram heat recovery through the steam mains, shaft work production, TAC, temperature of steam mains and steam flow rate simultaneously; and (2) TASP-SUGCC, to demonstrate on the same diagram heat recovery through the steam mains, shaft work production, TASP, temperature of steam mains and steam flow rate. As shown in Fig. 5, the left side of horizontal axis shows the TAC for utility and the right side of horizontal axis shows the heat demand of the steam mains. Furthermore, it is helpful to display the system information graphically to visualize the TAC and shaft work production, steam generation and used in each steam main, heat demands and fuel requirement. In the TASP-SUGCC representation, the left side of horizontal axis demonstrates the TASP of site utility and the other side is same as original SUGCC. The left side illustrated the gross income obtained by sale of steam and power in each zone. TASP-SUGCC can visualize the TASP, steam generation and used, steam heat demands, fuel requirement and shaft work production in site utility, in the one diagram.
4.1.
Case study 1
In the first case study, the four considered steam levels were very high pressure (VHP), high pressure (HP), medium pressure (MP) and low pressure (LP) at 120, 50, 14 and 3 bar(a). The heat demand at HP, MP and LP steam levels was 50, 40 and 85 MW. The data parameters related to steam levels were specified as given in Table 2. Fig. 6 shows the necessary input data for cogeneration targeting of the site utility system related to case study 1 (Ghannadzadeh et al., 2011). In this case, the water supplied to the boiler was assumed to be at the temperature of 105 ◦ C and the degree of superheat in LP was assumed to be 40 ◦ C.
Fig. 6 – Total site utility system for case study 1.
1045
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
350
Saturation Temperature ( C)
i=1(VHP, 120 bar) 300
i=2(HP, 50 bar) 250
50 MW
200
i=3(MP, 14 bar) 40 MW
150
i=4(LP, 3 bar) 85 MW
100
Fig. 7 – Total site utility system for case study 2.
0
50
100
150
200
Heat Load(MW)
Fig. 8 – The SUGCC of case study 1.
Fig. 9 – TAC and cogeneration potential based on Aguilar (2005) steam turbine cost function obtained from the new method (case study 1).
Fig. 10 – TAC and cogeneration potential based on Al-Azri (2008) steam turbine cost function obtained from the new method (case study 1).
1046
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Fig. 11 – TAC and cogeneration potential based on THERMOFLOW obtained from the new method (case study 1).
Fig. 12 – TASP-SUGCC based on based on Al-Azri (2008) cost function (case study 1).
4.2.
Case study 2
In the second case study, the four considered steam levels were very high pressure (VHP), high pressure (HP), medium pressure (MP) and low pressure (LP) at 90, 46, 15.5 and 2.7 bar. The heat demand at MP and LP steam levels was 6.88 and 16.25 MW, respectively, and, in this case, the process steam generation at HP level was higher than the process steam demand and heat surplus at HP steam levels was 10.63 MW. The data parameters for each level are specified as given in Table 3. Fig. 7 shows the necessary input data for cogeneration targeting of the site utility system related to case study 2 (Ghannadzadeh et al., 2011). It was assumed that the water supplied to the boiler was at the temperature of 105 ◦ C and the degree of superheat in LP was assumed to be 20 ◦ C.
5.
Results
5.1.
Case study 1
The SUGCC of the first case study is shown in Fig. 8. It presents a schematic of the shaft power target in the second case study obtained from the new method. Table 4 indicates the shaft power targeting results from the main shaft work targeting models and the new method.
Table 3 – Data parameters for case study 2. VHP Pressure (bar(a)) Saturation temperature (◦ C) Net heat load (MW)
90 303 0
HP 46 259 −10.63
MP
LP
15.5 200 6.88
2.7 130 16.25
1047
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Table 4 – Cogeneration targeting results for case study 1. Methodology
Error %
IBTM THM – Model in STAR SHM New method STAR simulation Thermoflow simulation
−10.50 −63.00 +8.99 −0.352 −0.344 –
Total (MW) 34.100 14.100 41.430 37.970 37.973 38.104
VHP-HP (MW)
HP-MP (MW)
MP-LP (MW)
13.490 9.400 18.200 14.740 14.740 14.797
12.28 4.700 14.460 13.520 13.520 13.558
8.330 0 8.770 9.710 9.710 9.749
Table 5 – Cogeneration targeting results for case study 2. Error %
Total (MW)
IBTM THM [Model in STAR] SHM New method STAR simulation Thermoflow simulation
−2.91 −7.33 19.15 −0.26 −0.22 –
4.520 4.400 4.200 5.400 4.521 4.524
As shown in Table 4, the total power target of 34.100 MW from IBTM methodology was significantly different from the detailed design procedure of 38.104 MW with an error of 10.50%. The shaft work target obtained from the THM model of 14.100 MW was 63.00% different from the shaft work obtained from the detailed design procedure. Similarly, SHM model target was 8.73% different from the actual shaft work from the detailed design procedure. These discrepancies in the shaft work targets were due to the assumptions used in these models. The shaft work target obtained from the new method of 37.970 MW was only 0.35% different from the detailed design procedure in THERMOFLOW (Thermofloex 18, 2008) and was only 0.0079% different from the STAR Software. The levelized cost of equipment were computed considering an interest rate i = 5% and a plant expected life n = 20 year with annual operating time of 8000 h/year. Three cost functions were considered for the case study 1. The first one was related to Aguilar (2005). The second one is based on AL-Azri (2008) steam turbine cost function and the third one is based on THERMOFLOW software (Thermofloex 18, 2008). With considering steam turbine cost function based on Aguilar (2005) relations, TAC and cogeneration potential obtained from the new method was demonstrated in Fig. 9. Moreover, based on cost functions of AL-Azri (2008) TAC and cogeneration potential obtained from the new method were illustrated in Fig. 10. In addition, the schematic illustration of shaft power target and TAC in first case study based on THERMOFLOW steam turbine cost function was shown in Fig. 11. Finally, TASP-SUGCC based on AL-Azri (2008) cost function was demonstrated in Fig. 12. The right side of this presentation shows the gross income related to sale of each product generated in site utility such as steam and power in each zone.
5.2.
Case study 2
The SUGCC of the second case study is shown in Fig. 13. A schematic of the shaft power target in the second case study obtained from the new method is demonstrated in Fig. 13. In addition, Table 5 presents the shaft power targeting results from the main shaft work targeting models and the new method.
VHP-HP (MW) 0.80 0.50 1.80 0.57 0.57 0.571
HP-MP (MW)
MP-LP (MW)
1.9 1.9 1.9 2.0 2.0 2.001
1.70 1.80 1.70 1.95 1.951 1.952
As shown in Table 5, the total power target of 4.4 MW from the IBTM methodology was significantly different from the detailed design procedure of 4.524 MW with an error of 2.74%. The shaft work target obtained from the THM model of 4.200 MW was 7.16% different from the shaft work obtained from the detailed design procedure. Similarly, SHM model target was 19.36% different from the actual shaft work from the detailed design procedure. These discrepancies in the shaft work targets were due to the assumptions used in these models. The shaft work target obtained from the new method of 4.520 MW was only 0.088% different from the detailed design procedure in THERMOFLOW and was only 0.022% different from the STAR Software. The levelized cost of components were calculated considering an interest rate i = 5% and a plant expected life n = 20 year with annual operating time of 8000 h/year. Two cost functions were considered for the case study 2. The first one was related to Aguilar (2005) and the second one is based on AL-Azri (2008) steam turbine cost functions. In this respect, Figs. 14 and 15 show a schematic illustration of shaft power target and TAC in second case study based on Aguilar (2005) and AL-Azri (2008) steam turbine cost functions respectively. 350
i=1(VHP, 90 bar)
300
Saturation Temperature ( C)
Methodology
10.63 MW
i=2(HP, 46 bar) 250
i=3(MP, 15.5 bar)
200
6.88 MW
150
i=4(LP, 2.7 bar) 16.25 MW
100 0
5
10 15 Heat Load(MW)
20
Fig. 13 – The SUGCC of case study 2.
25
1048
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Fig. 14 – TAC and cogeneration potential based on Aguilar (2005) steam turbine cost function obtained from the new method (case study 2).
Fig. 15 – TAC and cogeneration potential based on Al-Azri (2008) steam turbine cost function obtained from the new method (case study 2).
Fig. 16 – TASP-SUGCC based on based on Al-Azri (2008) cost function (case study 2).
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Also, TASP-SUGCC based on AL-Azri (2008) cost function was demonstrated in Fig. 16.
6.
Conclusion
In this paper, new targeting method for estimation of cogeneration potential and TAC has been proposed. The methodology is required to be further extended to accommodate the integration of renewable energy sources, such as solar and wind, for the total site with high accuracy. Another work which could be considered in future is to optimize steam levels for reducing the overall energy consumptions for the site. Moreover, a new cogeneration targeting model developed in this work can be used for the heat recovery of low-grade waste heat in process industries, by addressing a wide range of low-grade heat recovery technologies with high accuracy.
Acknowledgment The authors wish to thank the Iran Power Plant Projects Management Company (MAPNA) for technical and financial supports.
References Aguilar, O., 2005. Design and optimization of flexible utility systems, Ph.D. Thesis, UMIST, Manchester. AL-Azri, N.A., 2008. Integrated approaches to the optimization of process-utility systems. Ph.D. Thesis, Texas A&M University, USA. Bandyopadhyay, S., Varghese, J., Bansal, V., 2010. Targeting for cogeneration potential through total site integration. Appl. Therm. Eng. 30, 6–14. Chemical Engineering Economic Indicators, 2012.. Dhole, V.R., Linnhoff, B., 1993. Total site targets for fuel co-generation, emissions, and cooling. Comput. Chem. Eng. 17 (Suppl.), 101–109. Fodor, Z., Klemes, J., Varbanov, P.S., Walmsley, M.R.W., Atkins, M.J., Walmsley, T., 2012. Total site targeting with stream specific minimum temperature difference. Chem. Eng. Trans. 29, 409–414, http://dx.doi.org/10.3303/CET1229069. Ghannadzadeh, A., Perry, S., Smith, R., 2011. A new shaftwork targeting model for total sites. Chem. Eng. Trans. 25, 917–922. Harell, D.A., 2004. Resource conservation and allocation via process integration. Ph.D. Thesis, Texas A&M University. Hackl, R., Harvey, S., 2012. Total site analysis (TSA) and exergy analysis for shaft work and associated steam and electricity savings in low temperature processes in industrial clusters. Chem. Eng. Trans. 29, 73–78, http://dx.doi.org/10.3303/CET1229013. INDEX MUNDI, 2012. . Kapil, A., Bulatov, I., Smith, R., Kim, J.-K., 2012. Site-wide low-grade heat recovery with a new cogeneration targeting method. Chem. Eng. Res. 90, 677–689. Klemes, J., Dhole, V.R., Raissi, K., Perry, S.J., Puigjaner, L., 1997. Targeting and design methodology for reduction of fuel power and CO2 on total sites. Appl. Therm. Eng. 17 (8–10), 993–1003.
1049
Klemeˇs, J., Friedler, F., Bulatov, I., Varbanov, P., 2010. Sustainability in the process industry—integration and optimization. McGraw-Hill, New York, USA, 362 p. Kravanja, Z., Glavic, P., 1997. Cost targeting for HEN through simultaneous optimization approach: a unified pinch technology and mathematical programming design of large HEN. Comput. Chem. Eng. 21 (8), 833–853. Liew, P.Y., Wan Alwi, S.R., Varbanov, P.S., Manan, Z.A., Klemes, J., 2012. A numerical technique for total site sensitivity analysis. Appl. Therm. Eng. 40, 397–408, http://dx.doi.org/10.1016/j.applthermaleng.2012.02.026. Matsuda, K., Tanaka, S., Endou, M., Iiyoshi, T., 2012. Energy saving study on a large steel plant by total site based pinch technology. Appl. Therm. Eng., http://dx.doi.org/10.1016/j.applthermaleng.2011.11.043. Mavromatis, S.P., Kokossis, A.C., 1998. Conceptual optimisation of utility networks for operational variations—I. Targets and level optimisation. Chem. Eng. Sci. 53 (8), 1585–1608. ˜ M., 2010. Modelling the power Medina-Flores, J.M., Picón-Núnez, production of single and multiple extraction steam turbines. Chem. Eng. Sci. 65 (9), 2811–2820. Mohammad Rozali, N.E., Wan Alwi, S.R., Abdul-Manan, Z., Klemeˇs, J.J., 2012. Design of hybrid power systems with energy losses, design of hybrid power systems with energy losses. Chem. Eng. Trans. 29, 121–126, http://dx.doi.org/10.3303/CET1229021. Mohan, T., El-Halwagi, M.M., 2007. An algebraic targeting approach for effective utilization of biomass in combined heat and power systems through process integration. Clean Technol. Environ. Policy 9 (1), 13–25. Nemet, A., Klemes, J., Kravanja, Z., 2012a. Minimisation of a heat exchanger networks’ cost over its lifetime. Energy 45, 264–276. Nemet, A., Varbanov, P.S., Kapustenko, P., Durgutovic, A., Klemes, J., 2012b. Capital cost targeting of total site heat recovery. Chem. Eng. Trans. 26, 231–236, http://dx.doi.org/10.3303/CET1226039. Perry, S., Klemes, J., Bulatov, I., 2008. Integrating waste and renewable energy to reduce the carbon footprint of locally integrated energy sectors. Energy 33, 1489–1497. Raissi, K., 1994. Total site integration. Ph.D. Thesis, UMIST, Manchester. Salisbury, J.K., 1942. The steam-turbine regerative cycle—an analytical approach. Trans. ASME 64, 231–245. Smith, R., 2005. Chemical process design and integration. Wiley, West Sussex. Sorin, M., Hammache, A., 2005. A new thermodynamic model for shaftwork targeting on total sites. Appl. Therm. Eng. 25, 961–972. STAR Software, Version 2, Center for Process Integration, School of Chemical Engineering & Analytical Science, University of Manchester, UK, Under license of K.N. Toosi University of Technology, Energy Engineering Department. 2008. Thermofloex 18, Thermal Engineering Software for the Power Industry. Thermoflow, Inc., USA. Ulrich, G.D., Vasudevan, P.T., 2006. How to estimate utility cost chemical engineering.. Varbanov, P.S., Doyle, S., Smith, R., 2004. Modelling and optimization of utility systems. Chem. Eng. Res. 82 (5), 561–578. Varbanov, P.S., Klemes, J., 2011. Integration and management of renewables into total sites with variable supply and demand. Comput. Chem. Eng. 35, 1815–1826. Varbanov, P.S., Fodor, Z., Klemes, J., 2012. Total site targeting with process specific minimum temperature difference (DTmin ). Energy 44, 20–28, http://dx.doi.org/10.1016/j.energy.2011.12.025.
Contents lists available at SciVerse ScienceDirect
Chemical Engineering Research and Design journal homepage: www.elsevier.com/locate/cherd
A new targeting method for estimation of cogeneration potential and total annualized cost in process industries M.H. Khoshgoftar Manesh a , S. Khamis Abadi b , M. Amidpour a,∗ , M.H. Hamedi a a
Energy Integration Research Center, Energy Systems Engineering Department, Mechanical Engineering Faculty, K.N. Toosi University of Technology, Tehran, Iran b Department of Energy Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
a b s t r a c t One of the important tasks for optimal design and analyses of site utility systems is targeting total annualized cost and cogeneration potential. This paper introduced a new cogeneration and total annualized cost (TAC) targeting models that were developed to estimate the cogeneration potential of site utility systems and total annualized cost. The procedure which was proposed here provided a consistent, general procedure for determining mass flowrates and efficiencies of the applied turbines. This algorithm utilized the relationship of the entropy with the enthalpy and the isentropic efficiency. It is considered superior to previous works in that it was accurate, did not require any cumbersome simulation for initiation and could be easily traced,which enhance its programmability, considering full lifetime of the utility prices and it can estimate TAC and TASP with diffrent cost functions. Also, the developed model based on the trends of historical prices has been considered for estimation of utility costs. In addition, developed graphical representations based on site utility grand composite curve (SUGCC) were introduced: (1) TAC-SUGCC, to illustrate on the same diagram heat recovery through the steam mains, shaft work production, TAC, temperature of steam mains and steam flow rate simultaneously; and (2) TASP-SUGCC, to demonstrate TASP on the extended SUGCC diagram. Finally, two case studies were used to illustrate the usefulness of the new method. © 2012 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. Keywords: Targeting; Total annualized cost; Cogeneration; Total site
1.
Introduction
Chemical processes usually require steam at different pressure and temperature values for heating and non-heating purposes. In order to provide steam in the required condition, the designer has to decide whether to provide steam in the extreme condition and then let it down to different levels or produce steams separately in different boilers. Many industrial processes operate within total sites (Raissi, 1994), where they are serviced and linked through a common central utility system. This utility system meets the demands for heat and power of individual process units by their indirect heat integration. However, greater benefits in terms of energy and capital cost can be obtained by looking at the entire site.
Details explanation of the total site analysis, grand composite curve and the total site profiles can be seen in Klemeˇs et al. (2010). Total site integration addresses the task of optimizing each process and utility system in the context of the overall site (Sorin and Hammache, 2005). A number of models have been proposed for the early estimation of cogeneration for utility systems using steam turbines. Dhole and Linnhoff (1993) proposed an exergetic model based on the site source–sink profiles. Raissi (1994) presented the T-H model based on the Salisbury approximation, assuming that power is linearly proportional to the difference between the inlet and outlet saturation temperatures. Mavromatis and Kokossis (1998) introduced the non-linear model of THM (turbine hardware model) based on the principle of the Willans’ line in order to incorporate variation of efficiency with turbine size and
∗
Corresponding author. Tel.: +98 21 88677272; fax: +98 21 88677272. E-mail addresses: [email protected] (M.H. Khoshgoftar Manesh), sajad [email protected] (S.K. [email protected] (M. Amidpour), [email protected] (M.H. Hamedi). Received 7 July 2012; Received in revised form 16 November 2012; Accepted 7 December 2012 0263-8762/$ – see front matter © 2012 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cherd.2012.12.002
Abadi),
1040
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Nomenclature h ˙ m P Q˙ s T TAC W C t op
Ctot op Ctot ft SUGCC GCC TASP
specific enthalpy (kJ/kg) mass flow rate (kg/s) pressure (bar) heat load (MW) specific entropy (kJ/kg K) temperature (◦ C) Total annualized cost Shaft work (MW) Costs index corresponding to the number operating periods overall operating plant cost specific costs of utility in time period fraction of time horizon corresponding to each period site utility grand composite curve grand composite curve total annualized sale of products
Greek symbol isentropic efficiency Subscripts is isentropic saturated conditions Sat tot total Superscripts DEM process steam demand GEN process steam generation i steam main net heat or mass load net Op operating costs
operating load. An advanced approach to these concepts, known as top-level analysis, is one that allows for “scoping”, i.e. selecting site processes to target for heat integration improvements proposed by Varbanov et al. (2004). Harell (2004) introduced a graphic technique for estimating the cogeneration potential which utilized the concept of extractable power and header efficiency to establish cogeneration potential. Varbanov et al. (2004) developed the improved turbine hardware model. Sorin and Hammache (2005) also developed an exergetic model based on thermodynamic insights for the Rankine cycle and showed that power was not linear in saturation temperature differences. Mohan and ElHalwagi (2007) presented a linear algebraic approach based on the concept of extractable power and steam main efficiency. Perry et al. (2008) extended the total site concept to a broader spectrum of processes in addition to the industrial process. A potential for the integration of renewable energy sources was introduced to reduce the carbon footprint of a locally integrated. ˜ (2010) proposed a modified Medina-Flores and Picón-Núnez thermodynamic model by keeping the advantages of the THM. Bandyopadhyay et al. (2010) developed a linear model based on the Salisbury approximation and energy balance at steam mains. A new shaft work targeting model, termed the iterative bottom-to-top model (IBTM), was presented by Ghannadzadeh
et al. (2011). Kapil et al. (2012) introduced a new method for estimating cogeneration potential of site utility systems by a combination of bottom-up and top-down procedures. Varbanov and Klemes (2011) proposed an extension of the total site methodology covering industrial, residential, service, business and agricultural customers and the incorporation of renewable energy sources (solar, wind, biomass, and some types of waste), accounting for the often substantial variability on the supply and demand sides and for the use of non-isothermal utilities. Matsuda et al. (2012) studied the heat recovery potential for a large steel plant using total site profile analysis. This is followed by an assessment of the potential benefit of reducing steam demands at various levels by successively optimizing the system in steps of steam demand reduction. This results in a set of curves for steam marginal prices for the system under consideration. Liew et al. (2012) introduced four new contributions: (1) total site sensitivity table (TSST), a tool for exploring the effects of plant shutdown or production changes on heat distribution and utility generation systems over a total site; (2) a new numerical tool for TSHI, the total site problem table algorithm (TS-PTA), which extended the well-established problem table algorithm (PTA) to total site analysis; (3) a simple new method for calculating multiple utility levels in both the PTA and TSPTA; and (4) the total site utility distribution (TSUD) table, which can be used to design a total site utility distribution network. Varbanov et al. (2012) proposed the total site heat recovery targeting using multiple Tmin specifications for the site processes and process-utility interfaces. It is also possible to define and use the Tmin contributions of individual process streams in a process (Kravanja and Glavic, 1997). Fodor et al. (2012) focused on extending traditional total site integration methodology to produce more meaningful utility and heat recovery targets for the process design. This methodology was a further development of a recently extended traditional pinch methodology. The previous extension was on the introduction of using an individual minimum temperature difference (Tmin ) for different processes so that the Tmin is more representative of the specific process. Further, it deals with stream specific Tmin inside each process by setting different T contribution (Tcont ) and also using different Tcont between the process streams and the utility systems. Mohammad Rozali et al. (2012) extended the pinch analysis concept used in Process Integration to determine the minimum electricity targets for systems comprising hybrid renewable energy sources. Power pinch analysis (PoPA) tools described graphical techniques to determine the minimum target for outsourced electricity and the amount of excess electricity for storage during start up and normal operations. Nemet et al. (2012a) proposed a deterministic and stochastic multi-period mixed-integer nonlinear programming (MINLP) models for heat exchanger network (HEN) synthesis to account for future price projections, where the utility cost coefficients are forecasted for the lifetime of the process. Hackl and Harvey (2012) demonstrated how heat integration tools such as total site analysis and exergy analysis can be applied to target for shaft work and hot utility savings for processes and utility systems operating below ambient temperature.
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Mohammad Rozali et al. (2012) extended the graphical and numerical power pinch analysis (PoPA) method by considering the energy losses that occur in the power system conversion, transfer and storage. These losses on the minimum outsourced electrical energy targets and storage capacity are evaluated and the storage cascade table (SCT) of PoPA is developed to include the effect of energy losses in the system design. Nemet et al. (2012a) introduced the capital cost targeting method for total site heat recovery to determine the lower bound on the heat transfer area for meeting the targeted heat recovery on the total site. In this paper, new method for estimation of cogeneration potential and TAC has been proposed. The pervious works cited have not high accuracy for estimation of cogeneration potentials. In this paper, accurate cogeneration targeting and cost estimation models have been proposed. The new method applied for targeting shaftwork, steam production and used with high accuracy, considering degree of superheat in, and estimate TAC and TASP. Also, different models for estimation of costs have been considered. In addition, new graphical representations have been developed to demonstrate the TAC and TASP based on modification on original SUGCC. The utility prices can fluctuate rather quickly and the operating costs may be very different from a year to year. In this regard, the full lifetime of the utility prices is considered in this work. So, the developed model based on the trends of historical prices has been considered for estimation of operating costs.
2.
Methodology
2.1.
Algorithm of proposed procedure
In this section, the new model was presented in detail to target the cogeneration potential for site utility systems. The method uses the site utility grand composite curve (SUGCC), which represents another form of the site composite curves (Klemes et al., 1997). The SUGCC was obtained from the site composite curves by being represented on temperature–enthalpy axes of each steam main by its saturation temperature and steam generation and usage loads, from the source and sinks profiles of the site composites. The differences between steam generation and steam usage set the VHP demand or the supply heat available at each main. The new model calculates the minimum required flow rate from a steam generation unit and the levels of superheat in each steam main based on the heat loads specified by SUGCC. Fig. 1 shows a schematic of the utility system layout on SUGCC diagram. The L given steam mains are indexed by i from the highest pressure steam main, meaning that i is equal to 1, 2, 3 and 4 for very high pressure (VHP), high pressure (HP), medium pressure (MP) and low pressure (LP) steam mains. There is an expansion zone between two steam mains. Each zone are indexed by Z starting from top, i.e. Z = 1 is for VHP-HP, and one single steam turbine is placed in each zone. Fig. 2 shows the thermodynamic expansion of steam at two different pressure levels on a temperature-entropy diagram. The step S1 − S2 shows an isentropic expansion. An isentropic process is an ideal case in which there is no kind of irreversibility, such as mechanical friction and heat losses. Step S1 − S2 is a better representation of what happens in reality. The outlet
1041
Fig. 1 – Schematic of the utility system layout on SUGCC. of the turbine is shifted to the right side which indicates the increase of entropy (state of disorder) caused by losses (ALAzri, 2008). The isentropic efficiency is basically the ratio of the enthalpy difference of step S1 − S2 to that of step S1 –S2 . The isentropic efficiency is a function of the load and, for fixed values of flow rates, it would be better to consider the highest efficiency, assuming to use turbines, for which the calculated flow rate will be the full load (AL-Azri, 2008). In the study by Varbanov et al. (2004), thermodynamic model was used to estimate the isentropic efficiency as follows: is =
Wmax Wis,max
Wis,max =
Wis,max − A B
(1)
where A and B are constants that depend on the turbine and are functions of the saturation temperature. A and B are calculated by Eqs. (2) and (3): A = b0 + b1 Tsat
(2)
B = b2 + b3 Tsat
(3)
The values of these constants are given in Table 1. At the boiler exit, for a given pressure and steam temperature, the enthalpy can be obtained with the aid of steam tables.
Fig. 2 – Thermodynamic expansion of steam at two different pressure levels on a temperature-entropy diagram.
1042
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Table 1 – The regression coefficients used in the isentropic efficiency equation. Back pressure turbines
Condensing turbines
Wmax ≤ 2000 kW
Wmax > 2000 kW
Wmax ≤ 2000 kW
Wmax > 2000 kW
0 1.08 1.097 0.00172
0 4.23 1.155 0.000538
0 0.662 1.191 0.000759
−463 3.53 1.22 0.000148
b0 (kW) b1 (kW/◦ C) b2 b3 (◦ C−1 )
The actual input enthalpy of steam mains are usually provided from the calculations of the previous steam mains. The input isentropic enthalpy of steam main can be obtained in the superheated region. Then, the efficiency is calculated. The actual enthalpy which will serve as the input enthalpy for the next zone is then calculated using the isentropic enthalpies and efficiency by Eq. (4). hi,
actual
= hi−l,
isentropic
− (hi−l,
isentropic
− hi,
isentropic )
(4)
In this study, the calculation of superheat temperature at each steam level was done using the iterative procedure based on a certain desirable amount of superheat in the LP steam main. This superheat was required to be set at 10–20 ◦ C (Smith, 2005). If the degree of superheat in the resulting LP steam main was less than the required level, then, operating conditions of VHP would be updated and iterated until the acceptable superheated conditions would be met for the LP steam main. The mass flow rate of steam expanding through the Z-th ˙ z ) can be calculated by the mass balance for i-th by turbine (m Eq. (5), as shown in Fig. 3: ˙z= m
˙ z−1 + m ˙ DEM m i
˙ GEN −m i
Step 1. Preparation of a model in SUGCC Step 2. Initial estimates of boiler superheat temperature Step 3. Finding initial estimates of mass flow rates passing by each zone, assuming isentropic expansions throughout the levels by Eq. (8)
mz,
initial
=
Q˙ net,i hi − hf,i
(8)
Q˙ net,i = Q˙ iDEM − Q˙ iGEN where Q˙ net,i is the net load at the given level, hi the steam main isentropic enthalpy at the given level, hf is the saturated liquid enthalpy at the given level Step 4. Correcting efficiency by Eq. (1) ˙ NET Step 5. Correcting hi and m for the given steam level i Step 6. Repeating the steps from the second iteration through convergence in a manner until they meet the stopping criteria (Eq. (9))
Z i=1
2
(mZ − mz,new ) ≤ ε
(9)
(5)
˙ GEN where m is the flow rate of steam generated by the process i DEM ˙i is the flow rate of steam demanded by the process, and m which can be calculated by Eqs. (6) and (7): ˙ DEM m = i
Q˙ iDEM Actual hi − hf,i
(6)
˙ GEN = m i
Q˙ iGEN Actual hi − hf,i
(7)
where hf,i is the enthalpy of the saturated liquid enthalpy at the pressure of i-th steam main. The procedure of cogeneration targeting for a given site utility system (Fig. 4) is presented as follows:
Step 7. Checking the superheat temperature when the first loop of algorithm terminates (if it falls below the allowed minimum, the superheat temperature of the boiler is increased and repeats the steps until meeting the desirable amount of superheat in the LP steam main) Step 8. Finding, steam temperatures and mass flows of boiler and each steam mains and efficiency Step 9. Finding the shaft work in each zone Step 10. Generation of final SUGCC Step 11. Finding operating parameters of site utility Step 12. Finding parameters of main components such as boiler and steam turbines Step 13. Calculation of total annualized operating cost in life time of site utility (Section 2.2.1) Step 14. Calculation of total annualized capital cost (Section 2.2.2) Step 15. Calculation of TAC (Section 2.2) Step 16. Generation of TAC-SUGCC and TASP-SUGCC representations Step 17. Targeting cogeneration potential, TAC and TASP
2.2.
Total annualized cost targeting
The total annualized cost is given by the sum of the operating costs over the relevant devices plus the sum of the annualized capital costs (Eq. (10)). TAC = Operation cost + Capital cost Fig. 3 – Mass load balance for i-th steam main.
where TAC is the total annualized cost ($/year)
(10)
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
1043
Fig. 4 – Algorithm of proposed cogeneration and TAC targeting method.
2.2.1.
Modeling of operating costs
The utility prices can fluctuate rather quickly and the estimation of operating costs may be very different from a year to year. It is highly unrealistic to expect constant prices for utilities in the future. Nemet et al. (2012b) considered the full lifetime of the HEN and future utility prices. In this paper, the developed model based on the trends of historical prices has been considered for estimation of operating costs, where the annual utility cost with varying prices is calculated for every period during the lifetime as:
op
Ctot =
op
t
tot (Ct · ft ) · top
(11)
Price projections are based on past prices as described by Ulrich and Vasudevan (2006). Also, other further data needed can be found in INDEX MUNDI (2012) and Chemical Engineering Plant Cost Index – CEPCI (2012).
2.2.2.
Eq. (12) represents a linear formulation to relate the purchase cost of a boiler to its design steam flow rate and pressure. PCbol = A × Mstm-D + B A = 0.249 × Pbol + 47.19
(12)
B = 3.29 × Pbol + 624.6 where PCbol is the boiler purchase cost (1000$) (2002), A and B are regression coefficients for the purchase function (1000$/(kg/s), 1000$), Pbol is the boiler steam design pressure (bar), Mstm-D the boiler design steam flow (kg/s). In this study two functions used to estimate the purchase cost of a steam turbines. First function is based on Aguilar cost functions that represents an equation to estimate the purchase cost of a steam turbine to its design steam flow rate and volumetric flow passing through the unit (Eq. (13)): PCst = 474.5 × Vol + 1111.7 √ Vol = Mst × vin × vout
(13)
Modeling of capital costs
The basis for calculating the overall capital expenditure of the plant is the purchase cost of the equipment. The most significant cost is assumed contributed by the turbine and the boiler. The purchase cost of a boiler mainly depends on its design steam flow rate and pressure, as these parameters define the amount and specifications for the materials of construction.
where PCst is the steam turbine purchase cost (without including electric generator) (1000$) (2002), Vol the average steam volumetric flow rate passing through turbine (m3 /s), Mst the design steam turbine mass flow rate (kg/s), vin is the specific volume of the steam entering the turbine (m3 /kg), vout the specific volume of the steam discharged by the turbine (m3 /kg).
1044
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Second function is based on AL-Azri (2008) equipment cost functions that represents an equation to estimate the purchase cost of a steam turbine to its design shaft power output (Eq. (14)). PCst = 475 × (W × 2.93E − 04)
0.45
(14)
where W is the turbine shaft power output (kW). Eq. (15) represents an equation to estimate the purchase cost of an electric generator to its design output. PCgen = 0.0414 × Wgen + 190.49
(15)
where PCgen is the purchase cost for electric generators (1000$), Wgen is the design output for electric generators (kW). Eq. (16) represents an equation to estimate the purchase cost of a deaerator (Aguilar, 2005). PCdea = 0.958 × Mdea + 30.0
(16)
where PCdea is the deaerator purchase cost (1000$) (2002), Mdea is the design water mass flow from the deaerator (kg/s). In addition, it is possible to update these functions employing the chemical engineering plant cost index (CEPCI) (2012) factor corresponding to the ratio of the actual index over the base year one. The capital costs of the component is converted to annualized cost by using the capital recovery factor CRF (i,n) (Eq. (17))
CRF =
i(1 + i)
Table 2 – Data parameters for case study 1.
Pressure (bar(a)) Saturation temperature (◦ C) Net heat load (MW)
4.
VHP
HP
MP
LP
120 324 0
50 264 50
14 195 40
3 133 85
The case study
To show the applicability of the new method in the total site analysis, two case studies were considered.
n
n
(1 + i) − 1.
(17)
where i is the interest rate, n is the specified plant life.
3.
Fig. 5 – Schematic of the utility system layout on modified SUGCC.
New graphical representation
Due to importance of estimation of total annualized cost and cogeneration production ahead of design, two new graphical representations based on site utility grand composite curve (SUGCC) were introduced: (1) TAC-SUGCC, to illustrate on the same diagram heat recovery through the steam mains, shaft work production, TAC, temperature of steam mains and steam flow rate simultaneously; and (2) TASP-SUGCC, to demonstrate on the same diagram heat recovery through the steam mains, shaft work production, TASP, temperature of steam mains and steam flow rate. As shown in Fig. 5, the left side of horizontal axis shows the TAC for utility and the right side of horizontal axis shows the heat demand of the steam mains. Furthermore, it is helpful to display the system information graphically to visualize the TAC and shaft work production, steam generation and used in each steam main, heat demands and fuel requirement. In the TASP-SUGCC representation, the left side of horizontal axis demonstrates the TASP of site utility and the other side is same as original SUGCC. The left side illustrated the gross income obtained by sale of steam and power in each zone. TASP-SUGCC can visualize the TASP, steam generation and used, steam heat demands, fuel requirement and shaft work production in site utility, in the one diagram.
4.1.
Case study 1
In the first case study, the four considered steam levels were very high pressure (VHP), high pressure (HP), medium pressure (MP) and low pressure (LP) at 120, 50, 14 and 3 bar(a). The heat demand at HP, MP and LP steam levels was 50, 40 and 85 MW. The data parameters related to steam levels were specified as given in Table 2. Fig. 6 shows the necessary input data for cogeneration targeting of the site utility system related to case study 1 (Ghannadzadeh et al., 2011). In this case, the water supplied to the boiler was assumed to be at the temperature of 105 ◦ C and the degree of superheat in LP was assumed to be 40 ◦ C.
Fig. 6 – Total site utility system for case study 1.
1045
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
350
Saturation Temperature ( C)
i=1(VHP, 120 bar) 300
i=2(HP, 50 bar) 250
50 MW
200
i=3(MP, 14 bar) 40 MW
150
i=4(LP, 3 bar) 85 MW
100
Fig. 7 – Total site utility system for case study 2.
0
50
100
150
200
Heat Load(MW)
Fig. 8 – The SUGCC of case study 1.
Fig. 9 – TAC and cogeneration potential based on Aguilar (2005) steam turbine cost function obtained from the new method (case study 1).
Fig. 10 – TAC and cogeneration potential based on Al-Azri (2008) steam turbine cost function obtained from the new method (case study 1).
1046
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Fig. 11 – TAC and cogeneration potential based on THERMOFLOW obtained from the new method (case study 1).
Fig. 12 – TASP-SUGCC based on based on Al-Azri (2008) cost function (case study 1).
4.2.
Case study 2
In the second case study, the four considered steam levels were very high pressure (VHP), high pressure (HP), medium pressure (MP) and low pressure (LP) at 90, 46, 15.5 and 2.7 bar. The heat demand at MP and LP steam levels was 6.88 and 16.25 MW, respectively, and, in this case, the process steam generation at HP level was higher than the process steam demand and heat surplus at HP steam levels was 10.63 MW. The data parameters for each level are specified as given in Table 3. Fig. 7 shows the necessary input data for cogeneration targeting of the site utility system related to case study 2 (Ghannadzadeh et al., 2011). It was assumed that the water supplied to the boiler was at the temperature of 105 ◦ C and the degree of superheat in LP was assumed to be 20 ◦ C.
5.
Results
5.1.
Case study 1
The SUGCC of the first case study is shown in Fig. 8. It presents a schematic of the shaft power target in the second case study obtained from the new method. Table 4 indicates the shaft power targeting results from the main shaft work targeting models and the new method.
Table 3 – Data parameters for case study 2. VHP Pressure (bar(a)) Saturation temperature (◦ C) Net heat load (MW)
90 303 0
HP 46 259 −10.63
MP
LP
15.5 200 6.88
2.7 130 16.25
1047
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Table 4 – Cogeneration targeting results for case study 1. Methodology
Error %
IBTM THM – Model in STAR SHM New method STAR simulation Thermoflow simulation
−10.50 −63.00 +8.99 −0.352 −0.344 –
Total (MW) 34.100 14.100 41.430 37.970 37.973 38.104
VHP-HP (MW)
HP-MP (MW)
MP-LP (MW)
13.490 9.400 18.200 14.740 14.740 14.797
12.28 4.700 14.460 13.520 13.520 13.558
8.330 0 8.770 9.710 9.710 9.749
Table 5 – Cogeneration targeting results for case study 2. Error %
Total (MW)
IBTM THM [Model in STAR] SHM New method STAR simulation Thermoflow simulation
−2.91 −7.33 19.15 −0.26 −0.22 –
4.520 4.400 4.200 5.400 4.521 4.524
As shown in Table 4, the total power target of 34.100 MW from IBTM methodology was significantly different from the detailed design procedure of 38.104 MW with an error of 10.50%. The shaft work target obtained from the THM model of 14.100 MW was 63.00% different from the shaft work obtained from the detailed design procedure. Similarly, SHM model target was 8.73% different from the actual shaft work from the detailed design procedure. These discrepancies in the shaft work targets were due to the assumptions used in these models. The shaft work target obtained from the new method of 37.970 MW was only 0.35% different from the detailed design procedure in THERMOFLOW (Thermofloex 18, 2008) and was only 0.0079% different from the STAR Software. The levelized cost of equipment were computed considering an interest rate i = 5% and a plant expected life n = 20 year with annual operating time of 8000 h/year. Three cost functions were considered for the case study 1. The first one was related to Aguilar (2005). The second one is based on AL-Azri (2008) steam turbine cost function and the third one is based on THERMOFLOW software (Thermofloex 18, 2008). With considering steam turbine cost function based on Aguilar (2005) relations, TAC and cogeneration potential obtained from the new method was demonstrated in Fig. 9. Moreover, based on cost functions of AL-Azri (2008) TAC and cogeneration potential obtained from the new method were illustrated in Fig. 10. In addition, the schematic illustration of shaft power target and TAC in first case study based on THERMOFLOW steam turbine cost function was shown in Fig. 11. Finally, TASP-SUGCC based on AL-Azri (2008) cost function was demonstrated in Fig. 12. The right side of this presentation shows the gross income related to sale of each product generated in site utility such as steam and power in each zone.
5.2.
Case study 2
The SUGCC of the second case study is shown in Fig. 13. A schematic of the shaft power target in the second case study obtained from the new method is demonstrated in Fig. 13. In addition, Table 5 presents the shaft power targeting results from the main shaft work targeting models and the new method.
VHP-HP (MW) 0.80 0.50 1.80 0.57 0.57 0.571
HP-MP (MW)
MP-LP (MW)
1.9 1.9 1.9 2.0 2.0 2.001
1.70 1.80 1.70 1.95 1.951 1.952
As shown in Table 5, the total power target of 4.4 MW from the IBTM methodology was significantly different from the detailed design procedure of 4.524 MW with an error of 2.74%. The shaft work target obtained from the THM model of 4.200 MW was 7.16% different from the shaft work obtained from the detailed design procedure. Similarly, SHM model target was 19.36% different from the actual shaft work from the detailed design procedure. These discrepancies in the shaft work targets were due to the assumptions used in these models. The shaft work target obtained from the new method of 4.520 MW was only 0.088% different from the detailed design procedure in THERMOFLOW and was only 0.022% different from the STAR Software. The levelized cost of components were calculated considering an interest rate i = 5% and a plant expected life n = 20 year with annual operating time of 8000 h/year. Two cost functions were considered for the case study 2. The first one was related to Aguilar (2005) and the second one is based on AL-Azri (2008) steam turbine cost functions. In this respect, Figs. 14 and 15 show a schematic illustration of shaft power target and TAC in second case study based on Aguilar (2005) and AL-Azri (2008) steam turbine cost functions respectively. 350
i=1(VHP, 90 bar)
300
Saturation Temperature ( C)
Methodology
10.63 MW
i=2(HP, 46 bar) 250
i=3(MP, 15.5 bar)
200
6.88 MW
150
i=4(LP, 2.7 bar) 16.25 MW
100 0
5
10 15 Heat Load(MW)
20
Fig. 13 – The SUGCC of case study 2.
25
1048
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Fig. 14 – TAC and cogeneration potential based on Aguilar (2005) steam turbine cost function obtained from the new method (case study 2).
Fig. 15 – TAC and cogeneration potential based on Al-Azri (2008) steam turbine cost function obtained from the new method (case study 2).
Fig. 16 – TASP-SUGCC based on based on Al-Azri (2008) cost function (case study 2).
chemical engineering research and design 9 1 ( 2 0 1 3 ) 1039–1049
Also, TASP-SUGCC based on AL-Azri (2008) cost function was demonstrated in Fig. 16.
6.
Conclusion
In this paper, new targeting method for estimation of cogeneration potential and TAC has been proposed. The methodology is required to be further extended to accommodate the integration of renewable energy sources, such as solar and wind, for the total site with high accuracy. Another work which could be considered in future is to optimize steam levels for reducing the overall energy consumptions for the site. Moreover, a new cogeneration targeting model developed in this work can be used for the heat recovery of low-grade waste heat in process industries, by addressing a wide range of low-grade heat recovery technologies with high accuracy.
Acknowledgment The authors wish to thank the Iran Power Plant Projects Management Company (MAPNA) for technical and financial supports.
References Aguilar, O., 2005. Design and optimization of flexible utility systems, Ph.D. Thesis, UMIST, Manchester. AL-Azri, N.A., 2008. Integrated approaches to the optimization of process-utility systems. Ph.D. Thesis, Texas A&M University, USA. Bandyopadhyay, S., Varghese, J., Bansal, V., 2010. Targeting for cogeneration potential through total site integration. Appl. Therm. Eng. 30, 6–14. Chemical Engineering Economic Indicators, 2012.
1049
Klemeˇs, J., Friedler, F., Bulatov, I., Varbanov, P., 2010. Sustainability in the process industry—integration and optimization. McGraw-Hill, New York, USA, 362 p. Kravanja, Z., Glavic, P., 1997. Cost targeting for HEN through simultaneous optimization approach: a unified pinch technology and mathematical programming design of large HEN. Comput. Chem. Eng. 21 (8), 833–853. Liew, P.Y., Wan Alwi, S.R., Varbanov, P.S., Manan, Z.A., Klemes, J., 2012. A numerical technique for total site sensitivity analysis. Appl. Therm. Eng. 40, 397–408, http://dx.doi.org/10.1016/j.applthermaleng.2012.02.026. Matsuda, K., Tanaka, S., Endou, M., Iiyoshi, T., 2012. Energy saving study on a large steel plant by total site based pinch technology. Appl. Therm. Eng., http://dx.doi.org/10.1016/j.applthermaleng.2011.11.043. Mavromatis, S.P., Kokossis, A.C., 1998. Conceptual optimisation of utility networks for operational variations—I. Targets and level optimisation. Chem. Eng. Sci. 53 (8), 1585–1608. ˜ M., 2010. Modelling the power Medina-Flores, J.M., Picón-Núnez, production of single and multiple extraction steam turbines. Chem. Eng. Sci. 65 (9), 2811–2820. Mohammad Rozali, N.E., Wan Alwi, S.R., Abdul-Manan, Z., Klemeˇs, J.J., 2012. Design of hybrid power systems with energy losses, design of hybrid power systems with energy losses. Chem. Eng. Trans. 29, 121–126, http://dx.doi.org/10.3303/CET1229021. Mohan, T., El-Halwagi, M.M., 2007. An algebraic targeting approach for effective utilization of biomass in combined heat and power systems through process integration. Clean Technol. Environ. Policy 9 (1), 13–25. Nemet, A., Klemes, J., Kravanja, Z., 2012a. Minimisation of a heat exchanger networks’ cost over its lifetime. Energy 45, 264–276. Nemet, A., Varbanov, P.S., Kapustenko, P., Durgutovic, A., Klemes, J., 2012b. Capital cost targeting of total site heat recovery. Chem. Eng. Trans. 26, 231–236, http://dx.doi.org/10.3303/CET1226039. Perry, S., Klemes, J., Bulatov, I., 2008. Integrating waste and renewable energy to reduce the carbon footprint of locally integrated energy sectors. Energy 33, 1489–1497. Raissi, K., 1994. Total site integration. Ph.D. Thesis, UMIST, Manchester. Salisbury, J.K., 1942. The steam-turbine regerative cycle—an analytical approach. Trans. ASME 64, 231–245. Smith, R., 2005. Chemical process design and integration. Wiley, West Sussex. Sorin, M., Hammache, A., 2005. A new thermodynamic model for shaftwork targeting on total sites. Appl. Therm. Eng. 25, 961–972. STAR Software, Version 2, Center for Process Integration, School of Chemical Engineering & Analytical Science, University of Manchester, UK, Under license of K.N. Toosi University of Technology, Energy Engineering Department. 2008. Thermofloex 18, Thermal Engineering Software for the Power Industry. Thermoflow, Inc., USA. Ulrich, G.D., Vasudevan, P.T., 2006. How to estimate utility cost chemical engineering.