A novel flue gas waste heat recovery system for coal-fired ultra-supercritical power plants

Applied Thermal Engineering 67 (2014) 240e249

Contents lists available at ScienceDirect

Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng

A novel flue gas waste heat recovery system for coal-fired ultra-supercritical power plants Gang Xu a, b, Cheng Xu a, Yongping Yang a, *, Yaxiong Fang a, Yuanyuan Li a, Xiaona Song c a

National Thermal Power Engineering & Technology Research Center, North China Electric Power University, Beijing 102206, China Key Laboratory of Low-grade Energy Utilization Technologies & Systems of Ministry of Education, Chongqing University, Chongqing 400044, China c Electrical and Mechanical Practice Center, Beijing Information Science & Technology University, Beijing 100192, China b

h i g h l i g h t s
A novel waste heat recovery system is proposed in this paper.
Energy, exergy and techno-economic analysis are quantitatively conducted.
Better energy-savings of the proposed WHRS is obtained through system integration.
Lower exergy destruction is achieved in the proposed WHRS.
Greater economic benefits have been found in the proposed WHRS.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 September 2013 Accepted 16 March 2014 Available online 25 March 2014

Recovering flue gas waste heat is important in improving power plant efficiency. The most widely method is installing a low-temperature economizer (LTE) after the electrostatic precipitator (ESP) to heat the condensed water, thereby saving the extraction steam from the steam turbine and achieving extra work. The inlet flue gas temperature of the LTE is relatively low, so it can only heat condensed water from low-grade regenerative heaters, resulting in comparatively minor energy savings. After conducting an indepth analysis of the conventional waste heat recovery system (WHRS), this paper proposes a novel WHRS, in which the air preheater is divided into high-temperature (HT) and low-temperature (LT) air preheaters, and the LTE can be situated between the ESP and the LT air preheater. Through system integration, higher-grade extraction steam can be saved, resulting in greater economic benefits. Results show that the net additional power output can reach 9.00 MWe and using the proposed WHRS can yield net benefits up to USD 2.60 million per year, which is much greater than those of conventional WHRS. Exergy destruction is also reduced from 34.1 MWth in the conventional WHRS to 28.5 MWth in the proposed WHRS. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Waste heat recovery Heat rate reduction Exergy destruction Techno-economic analysis Utility boiler

1. Introduction In China, coal-fired power plants consume nearly half of all available coal, and the resulting CO2 emissions account for over 40% of the total nationwide emissions. Therefore, energy conservation in coal-fired power plants is important to China’s energy security and programs for greenhouse gas control. Currently, exhaust gas temperature can reach 120  Ce140  C or even higher [1,2]. The thermal energy of exhaust flue gas dumped into the environment accounts for approximately 50%e80% of a

* Corresponding author. E-mail address: [email protected] (Y. Yang). http://dx.doi.org/10.1016/j.applthermaleng.2014.03.038 1359-4311/Ó 2014 Elsevier Ltd. All rights reserved.

boiler’s thermal loss, and 3%e8% of the plant total energy input. Obviously, there exists a great potential to improve the efficiency of the power plant by the recovery of the waste heat of the flue gas [3,4]. The most widely used method of flue gas heat recovery is installing an auxiliary heat transfer surface, referred to as a lowtemperature economizer (LTE), downstream of the electrostatic precipitator (ESP) to heat a portion of the condensed water. It is well known that, the condensed water is recycled back into the boiler as feed water at temperature ranging from 250  C to 300  C after multistage preheating using steam extracted from different levels in the regenerative heating system. This process is a remarkable energy conservation concept that is widely applied in existing power plants. For the LTE, all the heat needed for

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249

Nomenclature

Abbreviation SG steam generator HPT high-pressure turbine IPT intermediate-pressure turbine LPT low-pressure turbine COND condenser FWP feed water pump CP condenser pump EG electric generator LTE low-temperature economizer WHRS waste heat recovery system GTI Gas Technology Institute US The United States USC ultra-supercritical LHV low heat value DEA deaerator RH regenerative heater ESP electrostatic precipitator NAR net annual revenues O&M operation and maintenance Symbols DPt work output increment, (MWe)

preheating the condensed water originates from the flue gas, instead of the extraction steam, which can save a portion of the extraction steam from the steam turbine. The saved steam is able to pass through the following stages of the steam turbine and continues to expand for more power output, accompanied by improved net power plant efficiency. Currently, the use of the flue gas waste heat is becoming a hot topic both in the industrial community and technological research. In Germany, the Schwarze Pumpe power plant with a 2  800 MW lignite generation unit uses a flue gas division system after the ESP, and uses exhaust energy to heat the condensed water [5]. In China, the Shanghai Waigaoqiao No. 3 power plant uses condensed water at the entrance of the 7th-stage low-pressure regenerative heater (RH) to retrieve heat energy from flue gas in the LTE located after the ESP: this system reduces the design temperature of the flue gas from 125  C to 85  C, which improves boiler efficiency 2%-points and overall unit efficiency by 0.8e0.9%-points [6]. Studies have also examined the structure and materials of heat exchangers of used in waste heat recovery system (WHRS). More specific, Zhao et al. [6] conducted a study on using LTE with spiralfinned tubes for waste heat recovery and the ash deposition flow characteristics of flue gas spiral finned tube economizers. The LTE structure is comprehensively optimized in this paper. Chen et al. [7] investigated technologies for exploiting the large amount of lowgrade heat available from flue gas through industrial condensing boilers, and recover the latent heat of water vapor in flue gas. The Gas Technology Institute (GTI) in the United States investigated the use of transport membrane condenser technology to recover water and latent heat in the exhaust gas and conducted a series of industrial tests and commercial projects under the United States Department of Energy [2]. Plastic heat exchangers were used to cool flue gas temperature down to 50  C. However, the low heat transfer coefficient of plastic LTE requires a large heat exchanger size, which is not widely applicable to large-scale power plants. As for the thermodynamic and economic analysis, Wang [8]

DPf Dpr hf D

DP q Etotal Pnet A

D3 DH DS T0 EAI Ce heq TIC fM fP fT fM&S S b i n r

241

increase in draft fan power, (kW) increase in flue gas resistance, (Pa) induced draft fan efficiency (%) volume flow of flue gas, (m3/s) net increase in work output, (MWe) heat rate of the power plant, (kJ/kWh) net input of power unit, (MWth) net power output, (MWe) energy level exergy change, (kJ/kg) energy change, (kJ/kg) entropy change, (kJ/(kg  C)) environmental temperature, ( C) extra annual income, (million USD) on-grid power tariff, (USD/kWh) the equivalent operation per year, (h/year) total investment capital, (million USD) material correction factor process pressure correction factor process temperature correction factor Marshall and Swift index scale parameter, (m) scale factor fraction interest rate per year,(%) number of years reference

demonstrated the energy-saving principles of the LTE using the equivalent enthalpy drop method, which is commonly used to estimate the off-design performance of steam turbines. Espatolero et al. [3] compared various WHRS configurations by the thermodynamic and economic analysis: an indirect WHRS in the bypass flue gains the best energy savings, increasing efficiency by 1.11% above the reference case. However, the configuration of this arrangement is rather complex, which will enhance the system complexity and reduce the cost-competitiveness of the WHRS. The techno-economic performance of alternative RH configurations in the WHRS has been compared in the literature [9]. However, these studies focused on optimizing LTE configuration in the condensed water side, the characteristics of the LTE arrangement in the flue duct were not considered. The preliminary study and demonstration projects showed the characteristics of the WHRS, including the structural characteristic of the LTE and the energy-saving calculation of the WHRS. A number of valuable achievements have been made. The feasibility of flue gas waste heat recovery was confirmed by techno-economic analysis and the anti-corrosion material technology was also been further developed. However, comprehensive system optimization of WHRS has not received full attention, because of the relatively mature application in existing projects nowadays. Methodologies for thermodynamic analysis are based on the first law of thermodynamics, and the exergy analysis of the WHRS is being disregarded. In fact, WHRS performance can be improved by the process optimization and system integration on the level of the whole tail heating surfaces in the boiler through relatively simple retrofitting. Thus, integrated analysis and system optimization of the tail heating surfaces, rather than focusing on the LTE itself, should be prioritized. Additionally, the exergy analysis of the WHRS is indispensable to determining the energy-saving mechanism of WHRS. In view of this, based on the in-depth analysis of conventional WHRS, we propose a novel WHRS, coupled with comprehensive

242

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249

thermodynamic and techno-economic analysis. The aims of this study as follows: (1) to analyze the influence of key thermodynamic parameters on thermodynamic performance and disclose the cause of the marginally economic benefits of conventional WHRS; (2) to reveal an effective measure to improve the energy-savings and reduce the exergy destruction of the WHRS; and (3) to propose a highly efficient and profitable technical scheme for the flue gas heat recovery projects.

Table 1 Overall performance of the reference unit. Related parameters Coal mass flow rate input Flue gas flow rate Air flow rate Low heat value of coal (LHV) Total energy of coal Input (LHV) Gross electrical output Total auxiliary power Net electrical output Gross efficiency Net efficiency

2. Description of the flue gas WHRS 2.1. LTE applied to waste heat recovery

kg/s kg/s kg/s MJ/kg, ar MWth MWe MWe MWe % %

From another perspective, the capital cost is enhanced as a result of the installation of heat exchangers in the WHRS. Therefore, to assess the WHRS comprehensively, thermodynamics analysis, heat transfer mechanics studies, and techno-economic analysis should be performed.

The installation of an LTE, which can save part of the extraction steam, is the most commonly used method of using the waste heat of flue gas in modern power plants. Fig. 1 shows the configuration of a conventional WHRS integrated with the regenerative system of a power plant. In the RHs, the thermal energy required to preheat the feed water comes from the multistage extraction steam. When an LTE is installed as shown in Fig. 1, part of the waste heat of the flue gas is recovered to heat the low-temperature condensed water, thereby replacing a portion of the extraction steam from the steam turbine. The saved extraction steam can then be able to pass through the following stages of the steam turbine and continues to expand for more output of power. Thus, the low-grade heat of the exhaust flue gas is utilized by the LTE for additional power output and the reduction of the power plant heat rate. In the above scenario, the energy donor is the exhaust flue gas, and the energy acceptor is the condensed water of the regenerative system in the LTE. Therefore, the performance of the WHRS is affected by the characteristic of the flue gas and the parameters of the steam cycle. To be specific, not only the quantities of heat released by the flue gas, but also the parameters of the saved extraction steam affect the power output and economic benefits. In particular, the more heat is released by the flue gas and the higher parameters extraction steam saved in the regenerative system, the better the energy-saving effects of the WHRS will be.

2.2. Reference system: a conventional WHRS in a typical ultrasupercritical (USC) power plant A conventional WHRS in a typical USC power plant is selected as a reference heat recovery system. The live steam is performed at 26.25 MPa, and 600  C. A single reheat is performed at 600  C and 5.85 MPa. Coal used comprise bituminite, with 56.26% carbon, 3.79% hydrogen, 12.11% oxygen, 0.82% nitrogen, 0.17% sulfur, and 18.1% water (wt.%). Temperature at the air inlet and outlet of the air preheater is 25  C and 355  C, respectively. Exhaust flue gas temperature of the air preheater is 130  C. The reference unit details and parameters of the regenerative system without the LTE are listed in Tables 1and 2, respectively. The LTE is installed downstream of the ESP, and its inlet flue gas temperature is equal to that of the exhaust flue gas from the ESP (Fig. 1). The minimum outlet flue gas temperature of the LTE is restricted [10] to prevent corrosion of the LTE. Depending on the 8

EG 1

HPT

3

IPT

LPT

13 4

2

6

5

7 9

18 22

SG

15

RH1

RH2

RH3 FWP

Air Preheater

RH5

10

11

RH6

RH7

COND

12

RH8

14

CP

Deaerator

21 19

ESP

17

LTE

16

20

Wet stack Induced Fan

FGD

Components SG Steam Generator

IPT Inermediate-Pressure Turbine

HPT High-Pressure Turbine

RH Regenerative Heater

ESP Electrostatic Precipitator

LTE Low-temperature Economizer

EG Electric Generator

113.8 1130.1 916.1 21.13 2405.9 1093.20 63.00 1030.20 45.44 42.82

FWP Feedwater Pump

LPT

Low-Pressure Turbine

COND Condenser FGD CP

Flue Gas Desulfurization Condenser Pump

Fig. 1. Schematic of the thermal system of a power plant with an LTE.

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249

243

Table 2 Main parameters of the regenerative heaters of the reference unit. Items

Regenerative system parameters

Extracted steam pressure (MPa) Extracted steam temperature ( C) Extracted steam flow rate (kg/s) Inlet feed water temperature ( C) Outlet feed water temperature ( C)

RH1

RH2

RH3

DEA

RH5

RH6

RH7

RH8

8.584 420.0 48.12 277.0 299.3

6.292 377.1 89.22 226.0 277.0

2.680 479.7 42.08 191.3 226.0

1.249 371.0 27.40 157.7 196.9

0.626 290.8 35.96 125.7 157.7

0.269 194.3 42.30 84.7 125.7

0.0671 88.8 24.50 62.8 84.7

0.0257 65.6 24.42 36.0 62.8

coal grade, flue duct materials, and other factors, the minimum outlet flue gas temperature is set to 95  C. Therefore, the temperature of the condensed water heated in the LTE is also restricted. The condensed water temperature should be lower than the flue gas temperature in the LTE. The minimum heat transfer temperature difference should also be considered to facilitate the heat transfer area of the LTE [11,12]. In this paper, we set the minimum heat transfer temperature difference between the flue gas and the condensed water to 10  C and the mean temperature difference should be above 20  C [11,13]. According to published report [9], the higher-stage extraction steam is saved, the greater energy-savings will be. In addition, energy-saving strongly affect techno-economic performance. Thus, the higher parameters extraction steam saved in the regenerative system, the greater energy savings of the WHRS will be, which in turn bring about better economic performance. In this paper, the optimal scheme in the water side is thus in series between RH7 and RH8. As shown in Fig. 2, the heat released by the flue gas is used to heat the condensed water from 62.8  C to 79.1  C in the LTE and the extraction steam of the RH8 from the steam turbine can be partly saved. 2.3. Proposal of a novel WHRS in the same USC power plant Based on the analysis in Section 2.2, increasing the inlet flue gas temperature and heating the condensed water from the higherstage RHs in the LTE is an effective measure of improving the performance of the WHRS. In addition, the heat transfer temperature difference in the air preheater is relatively large in the conventional WHRS, particularly at the inlet of the ambient air. Thus, to increase the inlet flue gas temperature of the LTE and reduce the temperature difference within the air preheater, we propose a novel WHRS integrated into the same power plant. The configuration of the proposed WHRS is shown in Fig. 3. Contrary to Flue Gas Air Condensed Water

400 350

o

Temperature( C)

300 250 200 150 100 50 0 0

50

100

150

200

250

300

Heat transfer rate(MW th) Fig. 2. T/Q diagram of the conventional WHRS.

350

400

the conventional WHRS, the air preheating system of the proposed system is divided into two stages: the high-temperature (HT) and low-temperature (LT) air preheaters. Ambient air is preheated in the LT air preheater and then enters the HT air preheater (also called the main air preheater) to complete the entire heating process. Temperature at the hot air outlet is fixed at 355  C, same as that in the conventional WHRS. The LTE is arranged downstream of the HT air preheater, particularly installed between the ESP and LT air preheater. In the proposed WHRS, flue gas with a temperature ranging from 140  C to 95  C is used to heat the ambient air in the LT air preheater other than heating the condensed water in the conventional WHRS, thereby increasing the inlet air temperature of the HT air preheater. Simultaneously the temperature of the exhaust gas from the HT air preheater, that is, the inlet of the LTE, can reach up to 174.5  C, which enables the heating of the condensed water from the outlet of the RH6. The extraction steam from the RH5 can then be partly saved, resulting in improved energy-saving performance compared with the conventional WHRS. Fig. 4 shows the heat transfer curves of the proposed WHRS. Compared with the conventional system, it is observed that: (1) Temperatures of the flue gas and condensed water in the LTE are increased dramatically. (2) As a result of the two-stage air heating arrangement, the temperature of the outlet flue gas in the LT air preheater is reduced from 130  C to 95  C, which is accompanied by a reduced mean logarithmic temperature difference. (3) The quantity of heat released by the flue gas remains unchanged and the LTE can heat higher-grade condensed water through the indepth optimization of system integration.

3. Performance evaluation 3.1. System simulation and main assumptions The thermodynamic cycle and energy equilibrium of these two WHRSs based on the same USC coal-fired power plant are simulated using EBSILON Professional software, which is widely used for the design, evaluation and optimization of different types of power plants. The software based on basic physics and conserves the energy and mass balance of all power plant processes. The software also includes an important database that can help to directly calculate the thermodynamic state of the power generation system and can therefore be used for simulating plant conditions at various loads with a high degree of fidelity. In addition, the simulation process has been previously validated [14,15]. The simulation results of these studies are in accordance with the design parameters of the plant and the experimental data. To ensure the precision and reliability of the simulation results, selecting the accurate method and block models is essential. The model details of the main components are shown Table 3. The simulation is approached as a small variation around the nominal conditions of operation, allowing the following assumptions simplified:

244

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249

8

EG HPT 1

IPT

LPT

3 2

4

5

6

13

7 9

15

RH1

18

RH2 RH3 FWP

24

19

SG

11

RH7

COND

12

RH8

14

CP

17

LTE 20

10

RH6

Deaerator D

HT Air Preheater

23

RH5

16

ESP

23

LT Air Preheater 22 21

Wet stack Induced Fan

FGD

IPT Inermediate-Pressure Turbine

SG Steam Generator HPT High-Pressure Turbine

RH Regenerative Heater

ESP Electrostatic Precipitator

LTE Low-temperature Economizer

EG

Electric Generator

LT

FGD CP

FWP Feedwater Pump

HT High-Temperature

LPT

Low-Pressure Turbine

COND Condenser Flue Gas Desulfurization Condenser Pump

Low-Temperature

Fig. 3. Schematic of the thermal system of a power plant with the proposed WHRS. -

-

For the different stages of the high-pressure (LP), intermediatepressure (IP), and low-pressure (LP) turbines, the isentropic efficiencies are constant and equal to 0.87, 0.92, 0.89 and respectively. A constant amount of fuel input and main steam flow rate are chosen for modeling the WHRS integrated with a thermodynamic cycle. The additional power output and improvement in net efficiency are achieved by saving extraction steam.

These hypotheses allow estimating the reduction in steam extraction rate by re-calculation the energy balance of the RH when a part is substituted by the flue gas sensible heat. Constant pressure and turbine extraction temperature are implied. In fact, the saved extraction steam will influence the extraction pressure and temperature in the steam turbine. However, the mass flow of the saved extraction steam is so minute excursions compared with the main Flue Gas Air Condensed Water

400 350

o

3.2. Thermodynamic analysis 3.2.1. Heat transfer calculation The simulation analysis above is based on the energy and mass balance of the power plant processes. However, the area calculation of the heat exchangers, i.e., the air preheater and the LTE, is based on the typical heat transfer equations, which are analyzed in this section. Under the conditions of air preheater and the LTE, the amount of heat transferred by radiation can be neglected. The overall heat transfer coefficient of the LTE is given by [16]:

KLTE ¼

1 a1

1 ðW=ðm2 $KÞÞ þ a12

(1)

where a1 and a2 denote the convective coefficients at the flue gas and the condensed water sides, respectively. High frequency finned tubes with inline positions are selected for the LTE. The convective heat transfer coefficient at the flue gas and the condensed water sides are then calculated as follows [17]:

300

Temperature( C)

steam flow rate, thus in this paper the fluctuation of the extraction pressure and temperature are neglected. The details of the thermodynamic parameters of the main streams in two WHRs are listed in Table 4.

250 200 150

a1 ¼ 0:134ðl=dÞRe0:681 Pr 0:33 ðs=14:85Þ0:2 ðs=tf Þ0:1134 ðW=ðm2 $KÞÞ

100

(2)

50

a2 ¼ 0:021ðl=dÞRe0:8 Pr 0:125 ðW=ðm2 $KÞÞ

0 0

50

100

150

200

250

300

Heat transfer rate(MWth ) Fig. 4. T/Q diagram of the proposed WHRS.

350

400

(3)

where l denotes the thermal conductivity of the flue gas (in Eq. (2)) and condensed water (in Eq. (3)), d represents the equivalent diameter of the fins, Re and Pr are the Reynolds number and the

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249 Table 3 Models details of the main components in EBSILON Professional. Components

Models

Steam generator

Dry ash extraction and single reheat is modeled as a black box Inlet pressure and the isentropic efficiencies are defined of the steam turbine. In most cases, the outlet pressure is defined by the inlet pressure of the following turbine stage. In case of the last turbine stage, the outlet pressure is defined by the inlet pressure of the condenser (0.0058 MPa). Mechanical efficiency ¼ 0.998 Upper terminal temperature difference, and the lower terminal temperature difference of the after-cooler, are to be specified. Pressure loss ¼ 3.3%e5% in steam extraction of different RHs, Heat loss ¼ 0% Inlet temperature (20  C) and pressure (0.1 MPa) of cooling medium is specified, upper temperature difference ¼ 5  C, pressure loss ¼ 0.005 MPa Isentropic efficiencies ¼ 0.8, mechanical efficiency ¼ 0.998, discharge pressure: condenser pump ¼ 1.36 MPa, feedwater pump ¼ 34.65 MPa Generator efficiency ¼ 0.99 Inlet flue gas pressure (0.098 MPa) and inlet air pressure (0.1009 MPa) of air preheater are specified. The pressure loss is calculated by empirical equations in Section 3.2.2.

Steam turbines

RHs

Condenser

Pumps

Electric generator Air preheater and LTE

Prandtl number of the flue gas (In Eq. (2)) and the condensed water (In Eq. (3)), s is the fin pitch, and tf is the fin thickness. For the rotary air preheater, the overall heat transfer coefficient is given by [17]:

Kair preheater ¼

x$Cn 1 x1 $a1

þ x31$a3

ðW=ðm2 $KÞÞ

(4)

where a1 and a3 denote the convective heat transfer coefficients at the flue gas and the air sides, respectively. x denotes the utilization Table 4 Main thermodynamic parameters of these two WHRSs. Steam ID

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Conventional WHRS

Proposed WHRS

Mass flow (kg/s)

Pressure (MPa)

Temperature ( C)

Mass flow (kg/s)

Pressure (MPa)

Temperature ( C)

859.16 48.14 721.80 721.80 89.22 42.06 27.40 595.53 35.96 42.30 5.45 26.19 485.64 542.44 859.16 574.07 574.07 1130.10 1130.10 1130.10 916.10 916.10

26.25 8.584 5.652 6.292 6.292 2.680 1.249 1.249 0.626 0.269 0.0671 0.0257 0.0058 0.0058 34.450 1.294 1.294 0.0980 0.0951 0.0945 0.1009 0.1001

600.0 420.0 600.0 377.1 377.1 479.7 371.0 371.0 290.8 194.3 88.8 65.6 35.4 35.4 299.3 62.8 79.1 378.0 130.0 95.0 25.0 354.5

859.16 48.14 721.80 721.80 89.22 42.08 27.40 595.53 19.25 43.03 25.18 25.09 482.98 539.79 859.16 652.33 652.33 1130.10 1130.10 1130.10 1130.10 916.10 916.10 916.10

26.25 8.584 5.652 6.292 6.292 2.680 1.249 1.249 0.626 0.269 0.0671 0.0257 0.0058 0.0058 34.450 1.174 1.174 0.0980 0.0953 0.0940 0.0936 0.1009 0.1007 0.1001

600.0 420.0 600.0 377.1 377.1 479.7 371.0 371.0 290.8 194.3 88.8 65.6 35.4 35.4 299.3 125.7 139.6 378.0 174.5 140.0 95.0 25.0 83.8 354.5

245

factor (x ¼ 0.9), and Cn represents the factor reflecting the rotational speed (Cn ¼ 1) [17]. x1 (0.46) and x2 (0.42) represent the share of the flue gas and the air in the air preheater, respectively. The convective heat transfer coefficient at the flue gas side and air sides in the air preheater can be calculated as [17]:

a ¼ 0:03  l  Re0:03  Pr 0:4 =dðW=ðm2 $KÞÞ

(5)

where l is the thermal conductivity of the flue gas or air, d is the equivalent diameter of the air preheater, and Re and Pr denote the Reynolds number and the Prandtl number of the steam (flue gas or air), respectively. The heat transfer between the flue gas and air or condensed water is generally has a countercurrent flow configuration, and the logarithmic mean temperature difference between the two streams can be expressed as:

Dt ¼

Dtd  Dtx  ð CÞ lnðDtd =Dtx Þ

(6)

where Dtd denotes the large temperature difference in the heat exchanger and Dtx is the small temperature difference in the heat exchanger. When the overall heat transfer coefficient and the mean temperature difference are determined. The heat transfer capacity, Q, can be achieved by the simulation process. The area of the LTE or the air preheater can be calculated as follows:

A ¼

Q ðm2 Þ K$Dt

(7)

3.2.2. Net additional power output Additional power output is generated by reducing the amount of extraction steam at a constant fuel consumption rate. Thus, the additional power output directly reflects the energy-saving effects of the WHRS. In this study, the increase in gross power output DP of the steam turbine can be calculated using the simulation results and the reference power plant data. Moreover, the auxiliary heating surface installed in the flue duct of the boiler will results in the increase in the flue gas resistance, which inevitably causes increased fan power output. Therefore, the net additional power output is calculated as follows:

DPnet ¼ DP  DPf ðMWÞ

(8)

where DP is the additional power output derived from saved extraction steam, (MW), DPf is the increment in the fan power with WHRS, (MW). The output of an induced fan is affected by the flue gas resistance and the flue gas flow rate. The increase in the induced fan power can be calculated as follows:

DPf ¼

D$Dpr ðMWÞ 1000hf

(9)

In Eq. (9), Dpr (kPa) is the increase in the flue gas pressure drop, hf represents the induced draft fan efficiency (hf ¼ 0.85), and D

denotes the flue gas flow rate (m3/s). For the LTE, the flue gas pressure drop can be further calculated as [18]:

DpLTE ¼ Eu  r  w2y  zðPaÞ

(10)

In Eq. (10), DpLTE refers to the increment of flue gas pressure drop (Pa), Eu represents characteristics of tube resistance, is the

246

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249

density of the flue gas (kg/m3), wy is the average velocity of the flue gas (m/s), and z is the total tube numbers along the direction of flue gas flow. Eu can then be calculates as:

 Eu ¼ a0 Re

b

pf d0

c 

S1 d0

d (11)

where Re stands for Reynolds number, pf/d0 represents the ratio of finned pitch space to wing space, S1/d0 denotes the relative pitch of high frequency finned tubes, and a0, b, c, and d are refer to the empirical data obtained from related design parameter diagrams. The quantitative value of a0, b, c, and d are selected as 1, 2, 3, and 4, respectively. The increase in the heat transfer area increase the pressure resistance of the of the rotary air preheater in the proposed WHRS. The flue gas pressure loss in the air preheater can be calculated as:

Dpair preheater ¼ l0

rv2

l ðPaÞ d 2

(12)

where l0 denotes the frictional resistance factors, l represents the height of the air preheater (m); d refers to the equivalent diameter (m); is stands for the flue gas density (kg/m3); and denotes the flue gas velocity (m/s). 3.2.3. Heat rate reduction In the electricity industry, heat rate is the metric unit most commonly used to track and report the performance of a thermal power plant. Heat rate q represents the amount of fuel energy input needed (usually LHV basis) to produce 1 kWh of net electrical energy output. Considering the net additional power output DPnet, the heat rate reduction is deduced from the following:

Dq ¼ 3600Etotal

DPnet ðkj=kWhÞ Pnet ðPnet þ DPnet Þ

Table 5 Thermodynamic characteristics and energy-saving effects in these two WHRSs. Item LTE Inlet flue gas temperature Outlet flue gas temperature Inlet condensed water temperature Outlet condensed water temperature Saved extraction steam pressure Saved extraction steam enthalpy Logarithmic mean temperature difference Overall heat transfer coefficient Heat transfer area Increase in flue gas pressure drop Air-preheater Inlet flue gas temperature (HT/LT air preheater) Outlet flue gas temperature (HT/LT air-preheater) Inlet air temperature (HT/LT air preheater) Outlet air temperature (HT/LT air preheater) Logarithmic mean temperature difference (HT/LT air-preheater) Overall heat transfer coefficient (HT/LT air-preheater) Heat transfer area (HT/LT airpreheater) Increase in flue gas pressure drop Results Gross power output (P) Gross additional power output (DP) Increase in fan power (DPf) Net additional power output (DPnet) Reduction in heat rate (Dq)

Conventional WHRS

Proposed WHRS



C C C  C MPa kJ/kg  C

130 95 62.8 79.1 0.0671 2624.1 40.8

174.5 140 125.7 139.6 0.626 3037.5 23.1

W/(m2 K) m2 Pa

345.9 18,447 616.2

393.2 27,727 1273.5



C

378/e

378/140



C

130/e

174.5/95



C

25/e

83.8/25



C

354.5/e

354.5/83.8



C

57.24/e

49.8/63.25

14.49

14.66/12.67

421,962

 

W/(m2 K) 2

Pa

e

377,546/ 70,476 205.3

MWe MWe MWe MWe (kJ/kWh) %

1098.35 5.15 0.47 4.68 38.2 0.43

1103.87 10.67 1.67 9.00 72.9 0.81

m

(13)

In Eq. (13), Etotal refers to the total input energy per unit time (MWth), and Pnet refers to the net power output, which takes auxiliary power away from the gross power output. In this study, the net output powers Pnet and Etotal remain constant. Thus, the heat rate reduction has a positive correlation with the increase in work output DPnet. 3.3. Results and analysis Based on the simulation and calculation above, Table 5 shows the main thermodynamic analysis results of the conventional and the proposed WHRSs. As shown in Table 5, based on the LTE, the inlet flue gas temperature in the proposed WHRS is higher than that in the conventional WHRS. Thus, compared with the LTE in the conventional WHRS the LTE in the proposed WHRS can heat higher temperature condensed water and save higher-grade extraction steam, resulting in an increase in the additional power output. A smaller temperature difference clearly leads to a larger LTE area in the proposed WHRS. For the air preheaters, the quantity of the heat released by the flue gas in the proposed system is similar to that in the conventional system. Otherwise, the total mean temperature difference is reduced in the proposed WHRS, leading to approximately 6% increase in the heating area of the air preheating system. The added heating area will increase the flue gas resistance, resulting in the increase in the fan power (1.67 MWe) in the proposed WHRS, which is 1.20 MWe larger than that in the conventional WHRS.

In general, by optimization the system integration, the gross additional power output in the proposed WHRS is 5.52 MWe larger than that in the conventional system. In spite of the increased fan power, the net additional output power and the heat rate reduction in the proposed system are still nearly double of those in the conventional system. 4. Graphical exergy analysis To understand the principles of the energy conservation and distributions of the exergy destruction, we assess the main heating process, including the air preheaters and the LTE in both WHRS by graphical exergy analysis. Compared with the conventional exergy analysis, this method may provide more specific information beyond that of the exergy difference between the output and the input. DH on the abscissa represents the quantity of transformed energy which is related to the first law of thermodynamics, while A on the ordinate denotes the energy quality of energy transformation and it is related to the second law of thermodynamics as follows [4,19]:

A ¼ D3 =DH ¼ 1  T0 ðDs=DHÞ

(14)

where D3 and DH represent the exergy change and energy change during the thermodynamic process, respectively. Ds represents entropy change and T0 denotes the environmental temperature which is set to 298.15 K. The inefficiency or exergy destruction is represented by the shaded area between the curve of energy donor Aed and that of energy acceptor Aea [20]. Obviously, the lower energy level difference between the energy donor and

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249

energy acceptor refers to the better match of energy levels between them. In the WHRSs, the flue gas plays the role of the heat donor. The condensed water and the ambient air being the heat acceptor are heated in the LTE and air preheaters. Fig. 5(a) and (b) illustrate the exergy-destruction distribution, including the air preheating system and the LTE in these two systems. For the air preheaters, Fig. 5(a) shows that the flue gas releases heat from A ¼ 0.54 to A ¼ 0.26 in the air preheater to heat the ambient air from A ¼ 0 to A ¼ 0.52. The large energy level difference between the flue gas and the air causes a large exergy destruction (30.2 MWth) in the conventional WHRS. By contrast, Fig. 5(b) shows that in the proposed WHRS, the flue gas releases heat from A ¼ 0.54 to A ¼ 0.34 in the LT air preheater and further releases heat from 0.38 to 0.19 in the LT air preheater to heat the ambient air, which result from using a two-stage air preheating. Lower exergy level difference between the energy donor and acceptor causes lower exergy loss (26.7 MWth in the two-stage air preheater) and better thermodynamic performance. Likewise, as to the LTE installation in these two systems, the exergy loss in the proposed WHRS is 1.8 MWth, which is 2.1 MWth lower than that in the conventional WHRS because of the lower energy level difference.

0.6

LTE

Air preheater

0.5

Generally speaking, the gross exergy destruction in the proposed WHRS is 5.6 MWth lower than that in the conventional system, which reveals the primary cause of the larger increase in the gross additional power output.

5. Techno-economic analysis The specific techno-economic analysis of the conventional and the proposed WHRSs is conducted in this section. The basic economic assumptions employed in this study as follows: (1) The auxiliary fees are fixed at 15% of the cost of the WHRS. (2) Operation and maintenance (O&M) cost is fixed at 4% of the total capital (TIC) per year [21]. (3) The on-grid power tariff is set at 0.061 USD/kWh, and the annual utilization hours is assumed to be 5500 h per year. (4) The exchange rate of RMB to U.S. dollar is set as 6.25 ￥/$; To comprehensively examine the WHRS, the TIC analysis is necessary. The TIC includes the capital investment cost of the WHRS itself and the related auxiliary fees (e.g., construction and installation fee) of the project [22]. For the conventional WHRS, only the LTE exchanger is required to be consider, i.e., TIC ¼ TICLTE. For the proposed WHRS, the added heat area of the two-stage air preheaters should also be calculated as well as the LTE, i.e., TIC ¼ TICLTE þ TICadded air preheater. The TIC can be estimated using the scaling up method [3,22e24], which can be expressed as:

TIC ¼ TICr

Flue Gas

247

 b S fP fT fM fM&S Sr

(15)

30.18MW th

A

0.4

where S and b denote the scale parameter and scale factor in the present scale, respectively, and TICr and Sr represent the total investment cost and scale parameter of the basic scale, respectively. The correction factors in relation to materials (fM), PeT operating conditions (fP, fT), and price are based on the different currencies from different years. The capital costs are therefore updated for 2013 in millions of dollars and are calculated using the cost index method, which is based on the Marshall and Swift index (fM&S). Table 6 shows the values and units for all the parameters used for the TIC calculation. In the cost analysis, the reference capital cost data of facilities are based on the conventional WHRS [6,26]. The results of the TIC analysis are shown in Table 7. The TIC of the proposed WHRS is 1.7 times greater than that of the conventional system mainly because of larger scales of the LTE and air preheater. Table 8 lists the annual techno-economic analysis results of both systems. In this table, the net annual revenue (NAR) is defined as:

Air

0.3

4.0MW th 0.2

0.1

Condensed Water

0.0 0

50

100

150

200

250

300

350

400

(a) Conventional WHRS HT Air preheater

0.5

LTE

LT Air preheater

18.6MW th

Flue Gas

NAR ¼ EAI  CTIC  CO&M

0.4

Air

where EAI is the extra income per year for the study case, which is calculated as:

1.8MW th

A

(16)

0.3

8.1MW th

EAI ¼ DPnet heq Ce

(17)

0.2

heq being the equivalent operation hours per year and Ce the ongrid power tariff. In addition, the annualized investment capital cost (CTIC) can be calculated as [3,29e31]:

Condensed Water

0.1

Air 0.0 0

50

100

150

200

250

300

350

400

(b) Proposed WHRS Fig. 5. Graphical exergy analysis of conventional WHRS and proposed WHRS.

CTIC ¼ TIC

ið1 þ iÞn ð1 þ iÞn  1

(18)

where TIC stands for the total investment capital analysis above, i refers to the fraction interest rate per year, and n denotes the number of years that the capital has been borrowed over a fixed

248

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249

Table 6 Parameters and correction factors for the cost model in these two WHRSs. Items

Conventional WHRS LTE

Proposed WHRS LTE air preheater

Basic costa (TICr, million USD) Basic scale (Sr, m2) Scale factorb (b) fMc fPc fTc fM&Sd

0.693 13,149 0.68 1.00 1.00 1.00 1.08

0.693 13,149 0.68 1.00 1.00 1.20 1.08

7.07 421,963 0.68 1.00 1.00 0.95 1.05

a TICr of the LTE is taken from a feasibility study of flue gas waste heat recovery project in China 2009 [6]. The TIC of the air preheater of the 1000 MW power plant is obtained from the design notes of the reference power plant 2011 [26]. b The scale factor is based on [3,25]. c The materials (fM) and PeT operating conditions (fP, fT) are based on [23].The materials in the LTE (15NiCuMoNb5) or the air preheater (COR-TEN steel) in the two systems are similar to those used in the basic scale. d The Marshall and Swift index (fM&S) is used in accordance with [27,28].

rate of interest. Here, the discounted rate (i) and the life of equipments (n) are set as 8% and 20 years respectively [29]. To simplify operation and maintenance cost (CO&M) calculation for the last term in Eq. (16), we assume that the CO&M is fixed at 4% of the TIC per year. Table 8 depicts that the EAI of the proposed WHRS is nearly 3.02 million USD, which is 1.44 million USD higher than that of the conventional WHRS. Likewise, the NAR of the proposed WHRS is also nearly double that of the conventional system. From the analysis in this section, the TIC of the proposed WHRS is larger, because of the increase in the exchanger area cost and related auxiliary fees. Nevertheless, the net additional power output in the proposed WHRS is significantly higher than that in the conventional system and it will strongly affect the NAR. Consequently, the NAR of the proposed WHRS can reach 2.60 million USD per year. Thus, by means of integration, the extra income and the net annual revenues have increased, which reflect the benefits resulting from optimization the system integration. 6. Conclusion Based on the analysis of the conventional WHRS, a novel WHRS is proposed in this study. The two systems are compared based on thermodynamics and techno-economics using the data from a typical USC unit. The following important conclusions are drawn. (1) Installing an LTE is a common method used for waste heat recovery from the exhaust flue gas to improve power plant efficiency. Optimizing the WHRS requires steam cycle improvement, flue gas temperature minimization, and a perfect integration between both systems.

Table 7 Total investment capital analysis in these two WHRSs. Items LTE Scale parameter (m2) TICLTE (million USD) Air preheater Scale parameter (m2) TICair preheater (million USD) TICadded air preheater (million USD) TIC parameters TIC (million USD)

Conventional WHRS

Proposed WHRS

18, 447 1.70

27, 727 2.69

421,963 7.07 e

448,023 7.36 0.28

1.70

2.97

Table 8 Annual techno-economic performances of these two WHRSs. Conventional Proposed WHRS WHRS Net additional power output (DPnet, MWe) Extra annual income (EAI, million USD) Annualized investment capital cost (CTIC, million USD) Operation and maintenance cost (CO&M, million USD) Net Annual revenues (NAR, million USD)

4.68 1.58 0.17 0.07 1.34

9.00 3.02 0.30 0.12 2.60

(2) The system integration analysis opens up a novel heuristic optimization scheme. In the proposed WHRS, the air preheating system is divided into two stages and the LTE is arranged between the ESP and the LT air preheater. The LTE can reduce the flue gas temperature from 174.5  C down to 140  C and the extraction steam from the RH5 can be partly substituted. Calculations show that the net additional power output can reach 9.00 MWe with almost 0.81% reduction in heat rate, which is nearly double that of the conventional WHRS. The exergy analysis shows that the exergy destruction in the proposed WHRS is 5.6 MWth lower than that in the conventional WHRS because the energy levels in the former system are better matched. (3) The techno-economic analysis results show that the TIC in the proposed WHRS is significantly larger than that in the conventional WHRS mainly because of the increase in the heat exchanger area and the related auxiliary fees. Nevertheless, because of the outstanding energy-saving performance of the proposed system, the net benefits is much better than the conventional system, with nearly USD 2.60 million per year under 5500 operation hours. Acknowledgements This study was supported by the National Major Fundamental Research Program of China No. (2011CB710706), the National Nature Science Fund of China (No. U1261210), the 111 Project (B12034), and Open Fund of Chongqing University, Key Laboratory of the Ministry of Education of the low-grade energy utilization technologies and systems (KH3381). References [1] Q. Xin, The completion of major power generation in the whole country in June 2010, China Power Enterp. Manage 8 (2010) 95 (in Chinese). [2] K. Hwang, C.h. Song, K. Saito, S. Kawai, Experimental study on titanium heat exchanger used in a gas fired water heater for latent heat recovery, Appl. Therm. Eng. 30 (2010) 2730e2737. [3] S. Espatolero, C. Cortes, L.M. Romeo, Optimization of boiler cold-end and integration with the steam cycle in supercritical units, Appl. Energy 87 (2010) 1651e1660. [4] I.H. Aljundi, Energy and exergy analysis of a steam power plant in Jordan, Appl. Therm. Eng. 29 (2009) 324e328. [5] L. Strömberg, G. Lindgren, J. Jacoby, R. Giering, M. Anheden, U. Burchhardt, H. Altmann, F. Kluger, G.-N. Stamatelopoulos, Update on Vattenfall’s 30 MWth oxyfuel pilot plant in Schwarze Pumpe, Energy Procedia 1 (2009) 581e589. [6] Z.J. Zhao, W.Z. Feng, L. Zhang, et al., Theoretical analysis and engineering practice of heat recovery form exhaust gas of power boilers, J. Power Eng. 29 (2009) 994e997 (in Chinese). [7] Q. Chen, K. Finney, H. Li, X. Zhang, J. Zhou, V. Sharifi, J. Swithenbank, Condensing boiler applications in the process industry, Appl. Energy 89 (2012) 30e36. [8] C. Wang, B. He, S. Sun, Y. Wu, N. Yan, L. Yan, X. Pei, Application of a low pressure economizer for waste heat recovery from the exhaust flue gas in a 600 MW power plant, Energy 48 (2012) 196e202. [9] G. Xu, S.W. Huang, Y.P. Yang, et al., Techno-economic analysis and optimization of the heat recovery of utility boiler flue gas, Appl. Energy 112 (2013) 907e917. [10] R. Saidur, J.U. Ahamed, H.H. Masjuki, Energy, exergy and economic analysis of industrial boilers, Energy Policy 38 (2010) 2188e2197.

G. Xu et al. / Applied Thermal Engineering 67 (2014) 240e249 [11] A. Bahadori, Estimation of combustion flue gas acid dew point during heat recovery and efficiency gain, Appl. Therm. Eng. 31 (2011) 1457e1462. [12] P. Stehlik, Heat transfer as an important subject in waste-to-energy systems, Appl. Therm. Eng. 27 (2007) 1658e1670. [13] L. Westerlund, R. Hermansson, J. Fagerstrom, Flue gas purification and heat recovery: a biomass fired boiler supplied with an open absorption system, Appl. Energy 96 (2012) 444e450. [14] L.G. Wang, et al., Multi-objective optimization of coal-fired power plants using differential evolution, Appl. Energy 115 (2014) 254e264. [15] D. Gewald, et al., Integrated system approach for increase of engine combined cycle efficiency, Energy Convers. Manage. 60 (2012) 36e44. [16] X.J. Shi, D.F. Che, A. Brian, J.M. Gao, An investigation of the performance of compact heat exchanger for latent heat recovery from exhaust flue gases, IJHMT 54 (2011) 606e615. [17] X.H. Wu, General Methods of Calculation and Design for Induaturial Boiler, Standards Press of China, Beijing, 2003 (in Chinese). [18] S.M. Yang, W.Q. Tao, Heat Transfer, Higher Education Press, Beijing, 2006 (in Chinese). [19] M. Ishida, N. Nakagawa, Exergy analysis of a pervaporation system and its combination with a distillation column based on an energy utilization diagram, J. Membr. Sci. 24 (1985) 271e283. [20] H. Jin, H. Hong, B. Wang, et al., A new principle of synthetic cascade utilization of chemical energy and physical energy, Sci. China Ser. Eng. Mater. Sci. 48 (2005) 163e179. [21] H. Jin, Development of Thermal Power Systems Based on Graphical Exergy Analysis (Dissertation for the Doctoral Degree), Tokyo Institute of Technology, Tokyo, 1994.

249

[22] T. Kreutz, R. Williams, S. Consonni, P. Chiesa, Co-production of hydrogen, electricity and CO2 from coal with commercially ready technology. Part B: economic analysis, Int. J. Hydrogen Energy 30 (2005) 769e784. [23] R. Smith, Chemical Process Design and Integration, second ed., John Wiley and Sons, 2005. [24] J. Sweny, Synthetic fuel gas purification by the SELEXOL process, Prepr. Am. Chem. Soc. Div. Fuel Chem. 18 (1976). [25] Q.S. Fu, Thermodynamic Analysis Method for Energy Systems, Xi’an Jiaotong University Press, China, 2005 (in Chinese). [26] China Power Engineering Consulting Group Corporation, Reference Price Index of Thermal Power Engineering Design, China Electric Power Press, Beijing, China, 2011 (in Chinese). [27] R.J. Marshall, D. Lozowsky, G. Ondrey, M. Phelan, K. Torzewski, Chem. Eng. 115 (2008) 64. [28] http://data.stats.gov.cn/ (retrieved on 29.03.14). [29] G. Xu, H. Jin, Y. Yang, et al., A comprehensive techno-economic analysis method for power generation systems with CO2 capture, Int. J. Energy Res. 34 (2010) 321e332. [30] M. Zhao, A.I. Minett, A.T. Harris, A review of techno-economic models for the retrofitting of conventional pulverised-coal power plants for post-combustion capture (PCC) of CO2, Energy Environ. Sci. 6 (2013) 25e40. [31] E. Godoy, S.J. Benz, N.J. Scenna, A strategy for the economic optimization of combined cycle gas turbine power plants by taking advantage of useful thermodynamic relationships, Appl. Therm. Eng. 31 (2011) 852e 871.