Numerical modeling of repowering of a thermal power plant boiler using plasma combustion systems

Energy 103 (2016) 38e48

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Numerical modeling of repowering of a thermal power plant boiler using plasma combustion systems Beycan Ibrahimoglu a, M. Zeki Yilmazoglu b, *, Ahmet Cucen c a

Giresun University, Faculty of Engineering, Department of Energy Systems, Giresun, Turkey Gazi University, Faculty of Engineering, Department of Mechanical Engineering, Ankara, Turkey c Anadolu Plasma Technology Center, Gazi University Technopark, Ankara, Turkey b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 5 May 2015 Received in revised form 12 February 2016 Accepted 22 February 2016

In this study, numerical analyses of repowering of a thermal power plant boiler using plasma combustion systems were performed. In order to reduce the energy consumption of the power plant, fuel-oil burners were disassembled and plasma combustion systems were installed on the surfaces of the boiler. The integration procedure, design data, and boundary conditions were given in detail. Superheater, economizer and tubes (dome) were modeled as porous media and the pressure losses of each section were compared with design data. The power plant was modeled according to the design parameters using the Thermoflex commercial software, in order to find the heat loads of each boiler section. These results were used as input data in CFD (Computational Fluid Dynamics) code. ANSYS Fluent was used for numerical analyses. Temperature contours, velocity vectors, and isosurfaces of temperature in the furnace were compared. According to the results, the integration of the plasma combustion systems to the boiler slightly decreases the velocities at the inlet of each domain. Additional energy from the plasma combustion system has no reverse effect in the case of overheating, especially for convective surfaces. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Repowering Thermal power plant Energy Plasma combustion Plasma

1. Introduction Sustainable production of energy with increasing demand profile is one of the main problems of energy engineers. Coal is the main fuel source with a percentage of 26.8% in total energy demand [1]. In recent years, many research activities have been carried out to decrease the harmful effects of coal combustion in terms of global warming. Most of the coal-fired thermal power plants, operating around the world have major efficiency or availability problems due to ageing. Repowering of these thermal power plants is of great importance to offset the increasing energy demand. Repowering can be defined as increasing the installed capacity, net electric efficiency, and decreasing the emissions per installed capacity of an existing thermal power plant. Generally, a gas turbine is added to the cycle in repowering applications. Feedwater heating, hot windbox, and parallel repowering are the three of the most commonly implemented repowering methods. In thermal power plants, repowering reduces CO2 emissions per installed capacity [1,2]. The most important parameter in repowering applications is the expected life of the components. Therefore, a detailed life * Corresponding author. Tel.: þ90 532 701 4159. E-mail address: [email protected] (M.Z. Yilmazoglu). http://dx.doi.org/10.1016/j.energy.2016.02.130 0360-5442/© 2016 Elsevier Ltd. All rights reserved.

expectancy analysis has to be carried out before repowering. Plasma combustion system application in thermal power plants is another repowering method in order to decrease emissions for existing thermal power plants. Fuel-oil burners are generally in use in thermal power plants for the startup operation and flame stabilization. Plasma activation of coal particles instead of using fuel-oil burners promotes more effective and environmentally friendly combustion [3,4]. Plasma systems are also used for combustion stabilization [5] in utility boiler furnaces. Plasma combustion systems can be used to promote early ignition and enhanced stabilization of a pulverized coal flame. In addition, plasma combustion systems reduce the harmful emissions originated from coal combustion [6]. Ignition of coal by plasma requires less energy compared to the case of using fuel oil or natural gas in thermal power plants for startup and flame stabilization [7e9]. It is explained by additional crushing of coal particles by plasma, production of free radicals, and acceleration of chemical reactions of oxidation. Kanilo et al. [7] showed that using microwave plasmatrons instead of fuel oil burners reduce equivalent energy consumption by 90% in the startup. During the plasma activation, part of the coal/air mixture is fed into the plasmatron, where plasma flame with high energy content induces gasification of coal and partial oxidation of the char carbon. Carbon is mainly

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

oxidized to carbon monoxide at the inlet of the furnace which can be easily ignited. Messerle et al. [8] numerically modeled a thermal power plant boiler with CINAR ICE and PLASMA-COAL 1D code. CFD (Computational Fluid Dynamics) analysis of a tangentially fired thermal power plant boiler with different turbulence models [10,11], gas temperature deviation in the upper furnace area [12,13], unburned carbon and NOx formations [14,15], the effect of the overfire air on NOx formation [16], and operating conditions [17e19] can be found in the literature. Constenla et al. [20] numerically investigated 350 MWe tangentially fired pulverized coal furnace in order to predict the flow characteristics with real operating conditions. Zhou et al. [21] used two-fluid-trajectory model to simulate 3D gas particle flows and coal combustion in a tangentially fired furnace. They offered a grid system rotated by an angle to reduce the computation time. Numerical and experimental results were compared to validate the TFT (Two Fluid Trajectory) model. The combustion behavior of coals has to be identified in order to design the combustion zone [22e25]. Bar-Ziv et al. [26] developed a tool for the prediction of the firing behavior of any coal in a specific boiler. The tool tested for 550 MW opposite-wall and 575 MW tangential-fired utility boilers. A method also developed for fouling and slagging behavior and determining the emissivity of the fly ash in a 50 kW test facility. Numerical modeling of co-firing of biomass and coal enables to discover the combustion problems for nonspherical particles [27e29]. Karampinis et al. [30] investigated cofiring lignite and biomass in a large scale utility boiler. Nonspherical form of the biomass particle, which influences the drag coefficient, devolatilization, and combustion characteristics, was taken into account. Retrofitting of oil burners for biomass injection was suggested. NOx formation in tangentially fired burners was also numerically modeled. Staging combustion [31], various coal types, firing configurations, and boiler sizes and types [32] were investigated numerically. Oxy-fuel combustion systems [33e36] are of great interest due to CO2 capture potential. Zhang et al. [37] compared gray and non-gray WSGGM (weighted-sum-of-graygases models) in oxy-coal furnaces. Non-gray WSGGM is used for both air and oxy-fuel combustion. Particle radiation and gas radiation were compared and non-gray WSGGM with weighting factors for particle radiation was suggested. Habib et al. [38] investigated the characteristics of the oxy-fuel combustion in a gas-fired water tube boiler for different oxygen inlet percentages. Ash recycling and re-burning [39], furnace sorbent injection [40], slagging and fouling prediction [41], and ash deposition were also investigated. Taha et al. [42] modeled ash deposition for co-combustion of MBM (meat and bone meal) and coal in a tangentially fired boiler. Ash deposition on the heat exchange surfaces was modeled on the basis of ash viscosity. In addition, Vuthaluru and Vuthaluru [43] used numerical model to investigate ash related problems in a large scale tangentially fired boiler. Additive injection was found to be one of the effective methods to overcome ash deposition with the optimum location of burner ports. Park et al. [44] combined a 3D CFD model and a 1D steam-water side model to simulate the effects of burner and OFA settings, firing patterns and coal blending on boiler efficiency and also pollutant formation and combustion efficiency. Zhang et al. [45] investigated EulereLagrange (EeL) and EulereEuler (EeE) models in a suddenexpanding coal combustor. The results show that the conventional EeL model can predict CO2 distribution reasonably when the number of particle trajectories is sufficient. The EeE model also gives a reasonable prediction of the trend of the CO2 distribution, but it underestimates the amount of CO2 because the fluctuation of particle temperature is not fully accounted for in the calculation of heterogeneous reaction rates. Drosatos et al. [46] used the macro heat exchanger model in the convective section of the boiler. Schuhbauer et al. [47] developed a detailed boiler model by

39

coupling the fire and steam side. The combustion chamber radiation interaction with convective part was modified in order to get closer results to target values. APROS and ANSYS Fluent were coupled. Baek et al. [48] investigated the effect of the coal blending method on carbon in ash and NOx emissions. The results show that in-furnace blending the method gives the least NOx and carbon in ash. He et al. [49] diagnosed metal surface overheating issues in the reheater section of a boiler. Velocity and temperature distributions were obtained for different working cases in order to obtain the cause of the overheating problem. Edge et al. [50] coupled a 1D process model and a 3D CFD model in order to investigate heat flux in a natural circulation boiler. Kuang et al. [51] also investigated the overfire air angle on flow characteristics in a down-fired furnace. Liu and Bansal [52] integrated multi-objective optimization with CFD to optimize boiler combustion in a coal fired power plant boiler. Modlinski [53] numerically modeled tangentially fired boiler retrofitted with swirl burner. Vuthaluru and Vuthaluru [54] modeled a wall fired furnace for different operating conditions. Particle traces were obtained to determine the residence time in the furnace. Chui et al. [55] specified the improvement strategies for eleven selected boilers in China. Numerical models were used to increase the availability and decrease the emissions. Crnomarkovic et al. [56] investigated radiative heat exchange inside the pulverized lignite fired furnace for the gray radiative properties with thermal equilibrium between phases. Diez et al. [57] reviewed conventional lumped models and semi-empirical approaches used in the online thermal monitoring of the boilers. Online modeling techniques improved by means of integrating offline CFD predictions. Different types of plasmas for different types of combustion systems have been investigated. Positive effects of the plasma systems on combustion dynamics and kinematics were reported [58e60]. In this study, numerical analyses of repowering of a thermal power plant using plasma combustion systems were performed. Retrofitting works, boundary conditions of numerical analyses, and design parameters were given. Fuel-oil burners were disassembled and plasma combustion systems were installed on the surfaces of the boiler. The details and installation descriptions can be found in the next section. The power plant was modeled according to the design parameters using the Thermoflex commercial software in order to find the heat loads of each boiler section. Validation of the results can be found in the previous study [2]. These results were used as input data in CFD code. For numerical analyses, ANSYS Fluent was used. Superheater, tubes, and economizer sections were modeled as porous media in order to model the pressure drop in these sections. Numerical and design data pressure drop values were compared in order to validate the numerical results. Temperature contours, velocity vectors, and isosurfaces of temperature in the furnace were obtained. 2. Retrofitting works and numerical modeling The Soma A thermal power plant began operation in 1957 and served until 2010. Currently, the installed capacity of the power plant is 44 MWe with two units. The boiler was designed to operate with Soma lignite with a lower heating value of 3550 kcal/kg. The ultimate and proximate analyses of Soma/Eynes lignite are given in Table 1. Design data and different operating conditions of one unit are given in Table 2. Operating condition 4 is the constant maximum load of the design data. Because the coal fired power plants are operated in base load to cover the demand, it was expected to work in condition 4 during the operation. Four fuel oil burners, around the corners of the boiler, were installed for the startup and stabilization of the flame. Twelve coal burners, eight of them were in service at condition 4, were also installed around the boiler.

40

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48 Table 1 Properties of SOMA/EYNES coal. Proximate analysis (as received) [wt.%] Moisture Volatile matter Fixed carbon Ash Ultimate analysis (dry basis) [wt.%] C H N O S Lower heating value [kJ/kg]

25.22 32.83 23.55 18.4 39.48 2.95 0.59 12.83 0.53 14,248

Retrofitting works has started in February 2014. Plasma combustion systems were installed instead of the fuel oil burners. Fuel oil burners were repealed and the locations of plasma combustion systems were determined. A photograph taken during the installation of plasma burners is given in Fig. 1. Steam pipes on the selected locations were cut, bended, and welded again as shown in Fig. 2. In order to ensure the tangential combustion, the locations of the burners are given in Figs. 3 and 4, respectively. In Fig. 3, the locations of the coal burners are shown according to the design data. Plasma burners were installed on the surfaces of the boiler instead of corners. In the numerical analyses, initially the boiler was modeled according to the design data. Then, the plasma combustion system was modeled in order to investigate the temperature, pressure and velocity profiles. The SIMPLE algorithm was chosen to describe the pressure and velocity relationship. The keε standard model was used to calculate the turbulent viscosity (mt), turbulent kinetic energy (k), and turbulent dissipation (ε). The conservation equations to represent the interaction between turbulence and chemistry can be modeled with several of models: Eddy dissipation, Eddy dissipation/Finite rate chemistry, Mixture Fraction/PDF model, etc. The Eddy dissipation/Finite rate chemistry model was used to model this interaction. Four combustion reactions were used in combustion calculations, as shown in Eqs. (1)e(4). In Eq. (1) volatile stands for the volatile part of the coal and in Eqs. (2) and (3) C stand for solid part of carbon [27]. The kinetic parameters of these reactions are shown in Table 3. It was considered that the reactive gas mixture was consisted of H2O, O2, CO2, CO and a volatile molecule (denoted as V). The main carbon monoxide and volatile matter releasing mechanism obtained from the coal gasification reactions.

Fig. 1. Installation of plasma burners.

Fig. 2. Bended screen pipes. Table 2 Design data and different operating conditions of Soma A TPP. Operation conditionsa Turbine power [MW] Water/Steam Pressures [bar] Inlet of Economizer Outlet of dome Outlet of Superheater Temperatures [ C] Steam temperature at outlet of superheater Water temperature at inlet of economizer Water temperature at outlet of economizer Gas temperature at inlet of superheater Gas temperature at outlet of economizer Air temperature at outlet of air pre-heater Stack temperature Mass flow rates [t/h] Fuel Combustion gas Steam mass flow rate a

1

2 7

3

4

5

12

17

22

NA

62 59.6 59.4

62.6 61.5 60.7

65.7 64 62.4

70 68 65

72 69.8 66.2

489.7 139 197 826 239 206.5 129

487.7 165 216 872 261 213 142

487 180 230 922 278 222.5 152

486.5 192 242 980 296 226.5 160

486.4 196 245 1000 302 228 162.5

11.4 93.3 51

15.6 121.4 72

20.3 150.8 96

22 161.5 105

6.96 60.6 30

Operation conditions 1: Technical minimum, 2: Constant minimum load, 3: Normal load, 4: Constant maximum load, 5: Transient maximum load.

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

Fig. 3. Locations of coal burners.

Fig. 4. Locations of plasma burners.

Volatile þ 1:546O2 /1:24CO2 þ 1:35H2 O þ 0:019N2

(1)

1 C < S > þ O2 /CO 2

(2)

Table 3 Kinetic parameters of reactions. Reaction number R1 R2 R3 R4

Activation energy 2.027e þ 08 90,000 90,000 1.67e þ 08

Preexponential factor

Temperature exponent

2.119e þ 11 0.07 0.07 2.2e þ 12

0 0 0 0

41

C < S > þ CO2 /2CO

(3)

1 CO þ O2 /CO2 2

(4)

Two different boundary conditions were considered for the heat transfer calculations. In the furnace section, a free stream temperature at the walls was given which corresponds to saturated steam temperature inside the screen tubes. Superheater, tubes, and economizer sections were modeled as porous media and heat transfer values were obtained from Thermoflex results modeled in previous studies [1,2]. Total heat transfer rates of each section were 38.6 MWt for furnace, 14.8 MWt for superheater, and 9.95 MWt for economizer. Internal emissivity was set to 0.8 for considering the radiation effects especially in furnace and superheater sections. DO (Discrete Ordinates) model was used and absorption coefficient was modeled with a domain-based WSGGM. Rosin-Rammler was chosen as coal particle size distribution model with 75 mme200 mm range with 3.5 power coefficient. Mass flow rates of coal and air for conventional and plasma combustion systems are given in Table 4. In Table 4, PA and SA stand for primary and secondary air, respectively. Three different burner stages were available for coal and air inlet in the design data. In the simulations, Stage 1 and Stage 2 were used as coal and air inlet as shown in Fig. 5(a) for conventional combustion system. The locations of the plasma burners are shown in Fig. 5(b). After plasma repowering, Stage 1e2 and plasma stage were used for coal and air inlets. The total mass flow rates of two different combustion systems were kept constant. Numerical modeling of plasma can be simulated by MHD (MagnetoHydroDynamic) module in ANSYS Fluent. However, coupling MHD module and coal combustion reactions is complicated. Instead of coupling MHD module, plasma conditions were modeled as hot air at the inlet of the burner. According to the experimental data with 10 kWel magnetron, inlet temperature and velocity can be assumed as 1273 K and 14 m/s, respectively. For the simplicity, the pipes in the superheater, dome, and economizer sections were modeled as porous media instead of drawing tubes. Pressure loss in these volumes was modeled as porous jump in each interior surface where face permeability and pressure-jump coefficient values varies according to pressure loss. Table 5 shows the pressure jump constants for each domain. In the design data, 5 Pa pressure drop for the gas side in the furnace was also given. However, the furnace was not considered as porous media. The geometry was constructed from the technical drawings of the power plant with the ANSYS Design Modeler tool. Sixteen air inlets, primary and secondary, were located on the corners of the boiler. Moreover, to describe the discrete phase, coal injection points were defined at each primary air inlet. In the burner geometry, some simplifications on the coal and air inlets were taken into account. An equivalent cylindrical area was employed as criterion instead of using the real configuration. This choice was based in terms of providing better element quality during the mesh generation process. Burner stages and sections of the boiler are shown in Fig. 5. Table 4 Comparison of mass flow rates of coal, primary air, and secondary air at the inlet of the burners.

Conventional CS Conventional and Plasma CS

Level

Coal [kg/s]

PA [kg/s]

SA [kg/s]

First stage Second stage First stage Second stage Plasma stage

2.4375 2.4375 2.1615 2.1615 0.552

2.403 2.403 2.037 2.037 4.88

13.617 13.617 11.543 11.537

42

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

Fig. 5. Geometry and sections of the boiler (a) conventional CS (b) conventional & plasma CS.

Table 5 Pressure loss coefficients sections in the boiler.

Superheater Dome (Tubes) Economizer

Face permeability

Pressure jump coefficient

0.00274 0.000486 0.000348

3.07 4.8891 4.184

The preparation of the mesh was generated with the ANSYS Meshing tool. An unconstructed mesh was created using a local refinement over the inlets and curve surfaces together. The cell dimensions were placed in a size range between 0.025 m and 0.1 m of edge length and 0.4 m of global length. Locations of the coal burners at each stage were 4 diverted from the central axis of the corner as shown in Fig. 3. After meshing, approximately 1.5 million tetrahedral elements were obtained. Mean skewness and orthogonality values were 0.3 and 0.85, respectively. In order to obtain mesh independency, the number of the elements was tripled. However, approximately same results were obtained. The convergence criterion was set as RSM (Root Mean Square) with a value of 1e-6.

Fig. 6. Pressure drop across the boiler from design data.

Combustion CFD modeling requires a solution procedure. In the first step, flow analyses were carried out without combustion and heat transfer boundary conditions. In the second step, patching was required to initialize the combustion reactions. In the third step, radiation model was activated and the solution was repeated. In the last step, heat transfer boundary conditions were activated. The solution procedure is given below: 1 Cold simulation: In this step, flow is simulated in order to solve the momentum and keε equations. It is called “cold simulation” because no combustion process occurs in this step. 2 Patching: Boiler temperature is set to 1500 K to initialize the mass transport of species and energy, where 1500 K initial temperature triggers the initialization of the combustion. To minimize the errors, discrete phase was computed in every 20 tries.

Fig. 7. Comparison of total pressure in the boiler: (a) design data and (b) plasma repowering.

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

Fig. 8. Comparison of streamlines in the boiler: (a) design data, (b) streamlines without plasma burners, and (c) streamlines with plasma burners.

Fig. 9. Comparison of velocity values of first stage: (a) design data and (b) plasma repowering.

Fig. 10. Comparison of velocity values of second stage: (a) design data and (b) plasma repowering.

43

44

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

3 Radiation: Radiation modeling is a necessity in boiler combustion analysis especially in the furnace section. DO model was activated. 4 Heat transfer: After the convergence of step 3, heat transfer was activated as heat sinks for each porous media.

3. Results In the numerical modeling of combustion in a tangentially fired boiler, pressure loss in each section has to be calculated. The pressure loss of each section was provided from the design data of the TPP, shown in Fig. 6. According to the design data, the pressure losses were found to be 118 Pa for the superheater, 220 Pa for the tubes, and 270 Pa for the economizer sections. According to the numerical results, the pressure loss of each section was calculated to be 110 Pa, 196 Pa and 216 Pa with the boundary conditions given in Table 5. The results show that the difference between design and numerical data was found to be approximately 7% for the superheater packages, which is an acceptable difference for numerical analyses. Fig. 7 shows the pressure contours in the boiler. It should be noted that the pressure, below the burners section, was lower for conventional systems after plasma repowering. This result directly affects the streamlines along the boiler. In addition, in the upper part of the burners, the pressure variation changed and a more homogenous pressure distribution was obtained after plasma repowering. In Fig. 8, the comparison of streamlines of the coal burners was presented according to the design data and after plasma repowering. In Fig. 8 (a), plasma burners were not activated to determine the characteristics of the streamlines in conventional combustion system. In Fig. 8 (b), plasma burners were activated and streamlines were obtained. In Fig. 8 (c), only plasma burners were activated to investigate the flow behavior of the plasma burners. It is obvious that streamlines became intense at the lower part of the boiler. Plasma burners were installed lower than the coal burners and the average velocities were found to be decreased. It should be noted that the total mass flow rate into the boiler was kept

constant. The residence time of a coal particle will be increased according to the results. One positive effect of this result is, after plasma CS integration, complete combustion can be achieved with increasing residence time. However, the negative effect can be found on the heat transfer mechanism especially for the convective surfaces. Due to the decreasing velocities, Nu and convective heat transfer coefficient were expected to decrease. According to the results, the velocities on the superheater inlet surface were found to be 7.37 m/s for conventional and 7.24 m/s for plasma combustion systems. The velocity profile of each burner stage was investigated before and after the integration of the plasma combustion systems. The flame has to be a rotational characteristic to ensure the mixing and turbulence. As a result of mixing and turbulence, a complete combustion can be achieved. The rotational velocity vectors are given in Figs. 9 and 10 for Stage 1 and 2, respectively. According to the area weighted results, the average velocity decreased from 24.7 m/s to 21.4 m/s after plasma CS integration for Stage 1. In addition, for Stage 2, the velocity decreased from 22.5 m/s to 19.9 m/s. It can be explained by decreasing the mass flow rate of the coal in Stage 1 and 2 according to Table 4. It should be noted that the total mass flow rate of the air and coal were kept constant for both cases. In addition, according to the experimental results, the inlet velocity for plasma CS was given as 14 m/s in numerical analyses. Therefore, it is obvious that the velocity values reduce according to the inlet boundary condition variation; however, the turbulence and mixing in the middle of the planes were obtained according to the vector plots. Although the effective heat transfer mechanism in this section of the boiler is radiation, the decrement in the velocities adversely affect the convection heat transfer. The percentage of this decrement is calculated to be less than 5% which can be ignored for water/steam side calculations. Fig. 11 (a) and (b) show the velocity vectors and temperature contours in plasma burner stage, respectively. As can be seen in Fig. 11 (a), the turbulence was obtained. In the corners of the boiler, stagnation points formed. However, in the middle of the plane, the velocity values were found to be greater than Stage 1 and 2. According to Fig. 11 (b), the highest temperature on the plane was

Fig. 11. (a) Velocity vectors and (b) Temperature distribution on a plane at the plasma inlet section.

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

found to be 1675 K due to the combustion. In the middle of the plane, the temperature drops down to 1100 K due to the swirl effect of the burners. The boundary condition for the walls can be seen in the figure, which was accepted as the temperature of the saturated steam, 495 K. Temperature isosurfaces were obtained to compare the temperature distribution and the effects of plasma combustion systems in the boiler. Fig. 12 shows the temperature isosurfaces for 1550, 1650, 1750, and 1850 K. For each group, the isosurface on the left

45

side shows the conventional CS and the one on the right side shows the plasma integrated CS. For lower temperatures, similar temperature isosurfaces were obtained. However, the similarity between isosurfaces disappeared with an increasing temperature as shown in Fig. 12. According to the energy content of the plasma, higher temperatures were obtained especially in the plasma burners section, as shown in Fig. 12 (d). These analyses were performed to check the temperatures in the superheater section to avoid overheating of the superheating pipes. According to the

Fig. 12. Comparison of temperature isosurfaces: (a) 1550 K, (b) 1650 K, (c) 1750 K, and (d) 1850 K.

46

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

results, higher temperatures caused by the plasma were below the superheater section. Fig. 13 represents temperature contours in mid-plane. The temperature gradients in plasma repowering were different than the design data. Fig. 13 shows that 1880 K and 2033 K were the highest temperature values for conventional and plasma

repowering, respectively. The temperature difference can be explained by additional energy content of the plasma CS. The mass flow rate of coal and air were kept constant as mentioned before. Therefore, it can be concluded that the integration of the plasma CS increases the temperature distribution in the furnace. Consequently, the radiation heat transfer can be increased due to the temperature difference. In order to show the effects of the temperature distribution, Figs. 14 and 15 are given. Results showed that the temperature values after plasma repowering increased for Stage 1 and 2.

4. Conclusions

Fig. 13. Comparison of temperature on a planar plane in the boiler: (a) conventional and (b) plasma repowering.

In this study, CFD analyses of a tangentially fired thermal power plant boiler were performed in order to evaluate the effects of the integration of the plasma combustion systems. Plasma combustion systems can be used instead of fuel-oil burners in order to decrease the start-up energy consumption during the start-up process. For this purpose, the thermal power plant was modeled in the Thermoflex software initially. The results obtained from this simulation were used as heat transfer input values of the boiler sections. Retrofitting works were explained and boundary conditions for plasma repowering case were introduced. Conventional and plasma combustion systems were modeled in ANSYS Fluent. The total mass flow rates into the boiler were kept constant for both cases and the simulations were repeated in order to compare the temperature contours, velocity vectors, and temperature isosurfaces. The pressure loss of each section was found and compared with the design data. For the same mass flow rate into the boiler, the streamlines were obtained. The results show that the velocities slightly decreased after the plasma integration in the boiler. In the case of residence time, the complete combustion can be achieved with lower excess air coefficients. In the case of heat transfer,

Fig. 14. Comparison of temperature in the first burner stage: (a) conventional and (b) plasma repowering.

Fig. 15. Comparison of temperature in the second burner stage: (a) conventional and (b) plasma repowering.

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

decreasing the velocity also decreases the heat transfer coefficient, which negatively affects the boiler thermal load. However, the variations in the velocities were found to be less than 1.5%. The swirl effect was obtained for coal and plasma burners stages. The temperature isosurfaces for 1550e1850 K were obtained in order to evaluate the temperature distribution in the furnace. The effect of additional energy by plasma was shown for higher temperatures. However, it is concluded that plasma integration will not adversely affect the superheater tubes in the case of overheating. The emissions and comparison of numerical results with experimental analyses will be considered in a subsequent study. Acknowledgment The authors gratefully acknowledge TUBITAK, General Directorate of Turkish Coal, and Anadolu Plasma Technology Center for funding this work through the TUBITAK 1511/1120305 project. References [1] Yılmazoglu MZ, Durmaz A, Baker D. Solar repowering of Soma-A thermal power plant. Energy Convers Manag 2012;64:232e7. [2] Yılmazoglu MZ, Durmaz A. Hot windbox repowering of coal-fired thermal power plants. Turk J Eng Env Sci 2013;37:33e41. [3] Askarova AS, Karpenko EI, Messerle VE, Ustimenko AB. Plasma enhancement of coal dust combustion. In: 35th EPS conference on plasma physics, Hersonissos; 2008. [4] Askarova AS, Karpenko EI, Lavrishcheva YI. Plasma supported coal combustion in boiler furnace. IEEE Trans Plasma Sci 2007;35(6):1607e16. [5] Belosevic S, Sijercic M, Stefanovic P. A numerical study of pulverized coal ignition by means of plasma torches in air-coal dust mixture ducts of utility boiler furnaces. Int J Heat Mass Transf 2008;51:1970e8. [6] Gorokhovski MA, Jankoski Z, Lockwood FC, Karpenko EI, Messerle VE, Ustimenko AB. Enhancement of pulverized coal combustion by plasma technology. Combust Sci Technol 2007;179(10):2065e90. [7] Kanilo PM, Kazantsev VI, Rasyuk NI, Schünemann K, Vavriv DM. Microwave plasma combustion of coal. Fuel 2003;82:187e93. [8] Messerle VE, Karpenko EI, Ustimenko AB, Lavrichshev OA. Plasma preparation of coal combustion in power plants. Fuel Process Technol 2013;107:93e8. [9] Karpenko EI, Messerle VE, Ustimenko AB. Plasma-aided solid fuel combustion. Proc Combust Inst 2007;31:3353e60. [10] Hart JT, Naser JA, Witt PJ. Aerodynamics of an isolated slot-burner from a tangentially-fired burner. Appl Math Model 2009;33:3756e67. [11] Tian ZF, Witt PJ, Schwarz P, Yang W. Numerical modelling of Brown coal combustion in a tangentially-fired furnace. In: Seventh international conference on CFD in the minerals and process industries. Melbourne, Australia: CSIRO; 2009. [12] Yin C, Rosendahl L, Condr TJ. Further study of the gas temperature deviation in large-scale tangentially coal-fired boilers. Fuel 2003;82:1127e37. [13] Park HY, Faulkner M, Turrell MD, Stopford PJ, Kang DS. Coupled fluid dynamics and whole plant simulation of coal combustion in a tangentially-fired boiler. Fuel 2010;89:2001e10. [14] Backreedy RI, Jones JM, Ma L, Pourkashanian M, Williams A, Arenillas A, et al. Prediction of unburned carbon and NOx in a tangentially fired power station using single coals and blends. Fuel 2005;84:2196e203. [15] Kuang M, Zhu Q, Li Z, Zhang X. Numerical investigation on combustion and NOx emissions of a down-fired 350 MWe utility boiler with multiple injection and multiple staging: effect of the air stoichiometric ratio in the primary combustion zone. Fuel Process Technol 2013;109:32e42. [16] Diez LI, Cortes C, Pallares J. Numerical investigation of NOx emissions from a tangentially-fired utility boiler under conventional and overfire air operation. Fuel 2008;87:1259e69. [17] Asotani T, Yamashita T, Tominaga H, Uesugi Y, Itaya Y, Mori S. Prediction of ignition behaviour in a tangentially fired pulverized cola boiler using CFD. Fuel 2008;87:482e90. [18] Belosevic S, Sijercic M, Tucakovic D, Crnomarkovic N. A numerical study of a utility boiler tangentially-fired furnace under different operating conditions. Fuel 2008;87:3331e8. [19] Al-Abbas A, Naser J, Dodds D. CFD modelling of air-fired and oxy-fuel combustion of lignite in a 100 kW furnace. Fuel 2011;90:1778e95. [20] Constenla I, Ferrin JL, Saavedra L. Numerical study of a 350 MWe tangentially fired pulverized coal furnace of the As Pontes Power Plant. Fuel Process Technol 2013;116:189e200. [21] Zhou LX, Li L, Li RX, Zhang J. Simulation of 3-D gas-particle flows and coal combustion in a tangentially fired furnace using a two-fluid-trajectory model. Powder Technol 2002;125:226e33. [22] Korytnyi E, Savaliev R, Perelman M, Chudnovsky B, Bar-Ziv E. Computational fluid dynamic simulations of coal-fired utility boilers: an engineering tool. Fuel 2009;88:9e18.

47

[23] Agraniotis M, Stamatis D, Grammelis P, Kakaras E. Numerical investigation on the combustion behaviour of pre-dried Greek lignite. Fuel 2009;88:2385e91. [24] Alvarez L, Gharebaghi M, Jones JM, Pourkashanian M, Williams A, Riaza J, et al. CFD modeling of oxy-coal combustion: prediction of burnout, volatile and NO precursors release. Appl Energy 2013;104:653e65. [25] Al-Abbas AH, Naser J. Computational fluid dynamics modelling of a 550 MW tangentially-fired furnace under different operating conditions. 5th BSME International Conference on Thermal Engineering Procedia Eng 2013;56: 387e92. [26] Bar-Ziv E, Saveliev R, Korytnyi E, Perelman M, Chudnovsky B, Talanker A. Evaluation of performance of Anglo-Mafube bituminous South African coal in 550 MW opposite-wall and 575 MW tangential-fired utility boilers. Fuel Process Technol 2014;123:92e106. [27] Ma L, Gharebaghi M, Porter R, Pourkashanian M, Jones JM, Williams A. Modelling methods for co-fired pulverized fuel furnaces. Fuel 2009;88: 2448e54. [28] Nikolopoulos N, Agraniotis M, Violidakis I, Karampinis E, Nikolopoulos A, Grammelis P, et al. Parametric investigation of a renewable alternative for utilities adopting the co-firing lignite/biomass concept. Fuel 2013;113: 873e97. [29] Gubba SR, Ingham DB, Larsen KJ, Ma L, Pourkashanian M, Tan HZ, et al. Numerical modelling of the co-firing of pulverized coal and straw in a 300 MWe tangentially fired boiler. Fuel Process Technol 2012;104:181e8. [30] Karampinis E, Nikolopoulos N, Nikolopoulos A, Grammelis P, Kakaras E. Numerical investigation Greek lignite/cardoon co-firing in a tangentially fired furnace. Appl Energy 2012;97:514e24. [31] Choi CR, Kim CN. Numerical investigation on the flow, combustion and NOx emission characteristics in a 500 MWe tangentially fired pulverized-coal boiler. Fuel 2009;88:1720e31. [32] Chui E, Gao H. Estimation of NOx emissions from coal-fired utility boilers. Fuel 2010;89. 2977e22984. [33] Al-Abbas AH, Naser J, Dodds D. CFD modelling of air-fired and oxy-fuel combustion in a large-scale furnace at Loy Yang A brown coal power station. Fuel 2012;102:646e65. [34] Black S, Szuhanszki J, Pranzitelli A, Ma L, Stanger PJ, Ingham DB, et al. Effects of firing coal and biomass under oxyfuel conditions in a power plant using CFD modelling. Fuel 2013;113:780e6. [35] Al-Abbas AH, Naser J, Hussein EK. Numerical simulation of brown coal combustion in a 550 MW tangentially-fired furnace under different operating conditions. Fuel 2013;107:688e98. [36] Guo J, Liu Z, Wang P, Huang X, Li J, Xu P, et al. Numerical investigation on oxycombustion characteristics of a 200 MWe tangentially fired boiler. Fuel 2015;140:660e8. [37] Zhang J, Ito T, Ito S, Riechelmann D, Fujimori T. Numerical investigation of oxy-coal combustion in a large-scale furnace: non-gray effect of gas and role of particle radiation. Fuel 2015;139:87e93. [38] Habib MA, Ben-Mansour R, Badr HM, Ahmed SF, Ghoniem AF. Computational fluid dynamic simulation of oxyfuel combustion in a gas-fired water tube boilers. Comput Fluids 2012;56:152e65. [39] Hwang MY, Kim SM, Kim GB, Lee BH, Song JH, Park MS, et al. Simulation studies on direct ash recycling and reburning technology in a tangentially fired 500 MW pulverized coal boiler. Fuel 2013;114:78e87. [40] Shi L, Liu G, Higgins BS, Benson L. Computational modeling of furnace sorbent injection for SO2 removal from coal-fired utility boilers. Fuel Process Technol 2011;92:372e8. [41] Ma Z, Iman F, Lu P, Sears R, Kong L, Rokanuzzaman AS, et al. A comprehensive slagging and fouling prediction tool for coal-fired boilers and its validation/ application. Fuel Process Technol 2007;88:1035e43. [42] Taha TJ, Stam AF, Stam K, Brem G. CFD modeling of ash deposition for cocombustion of MBM with coal in a tangentially fired utility boiler. Fuel Process Technol 2013;114:126e34. [43] Vuthaluru HB, Vuthaluru R. Control of ash related problems in a large scale tangentially fired boiler using CFD modelling. Appl Energy 2010;87:1418e26. [44] Park HY, Baek SH, Kim YJ, Kim TH, Kang DS, Kim DW. Numerical and experimental investigations on the gas temperature deviation in a large scale, advanced low NOx, tangentially fired pulverized coal boiler. Fuel 2013;104: 641e6. [45] Zhang Y, Wei XL, Liu DY. CFD study of a sudden-expanding coal combustor using Euler-Euler and Euler-Lagrange models. Fuel 2010;89:3643e9. [46] Drosatos P, Nikolopoulos N, Agraniotis M, Grigorios I, Grammelis P, Kakaras E. Decoupled CFD simulation of furnace and heat exchangers in a lignite utility boiler. Fuel 2014;117:633e48. [47] Schuhbauer C, Angerer M, Spliethoff H, Kluger F, Tschaffon H. Coupled simulation of a tangentially hard coal fired 700 C boiler. Fuel 2014;122:149e63. [48] Baek SH, Park HY, Ko SH. The effect of the coal blending method in a coal fired boiler on carbon in ash and NOx emissions. Fuel 2014;128:62e70. [49] He B, Zhu L, Wang J, Liu S, Liu B, Cui Y, et al. Computational fluid dynamics based retrofits to reheater panel overheating of No. 3 boiler of Dagang power plant. Comput Fluids 2007;36:435e44. [50] Edge PJ, Heggs PJ, Pourkashanian M, Williams A. An integrated computational fluid dynamics-process model of natural circulation steam generation in a coal-fired power plant. Comput Chem Eng 2011;35:2618e31. [51] Kuang M, Li Z, Ling Z, Jing X, Zhu Q. Effect of overfire air angle on flow characteristics within a small-scale model for a deep-air-staging down-fired furnace. Energy Convers Manag 2014;79:367e76.

48

B. Ibrahimoglu et al. / Energy 103 (2016) 38e48

[52] Liu X, Bansal RC. Integrating multi-objective optimization with computational fluid dynamics to optimize boiler combustion process of a coal fired power plant. Appl Energy 2014;130:658e69. [53] Modlinski N. Computational modeling of a utility boiler tangentially fired furnace retrofitted with swirl burners. Fuel Process Technol 2010;91:1601e8. [54] Vuthaluru R, Vuthaluru HB. Modelling of a wall fired furnace for different operating conditions using Fluent. Fuel Process Technol 2006;87:633e9. [55] Chui EH, Gao H, Majeski AJ, Lee GK. Performance improvement and reduction of emissions from coal-fired utility boilers in China. Energy Sustain Dev 2010;14:206e12. [56] Crnomarkovic N, Sijercic M, Belosevic S, Tucakovic D, Zivanovic T. Radiative heat exchange inside the pulverized lignite fired furnace for the gray radiative

[57]

[58] [59] [60]

properties with thermal equilibrium between phases. Int J Therm Sci 2014;85: 21e8. Diez LI, Cortes C, Campo A. Modelling of pulverized coal boilers: review and validation of on-line simulation techniques. Appl Therm Eng 2005;25: 1516e33. Ju Y, Sun W. Plasma assisted combustion: progress, challenges, and opportunities. Combust Flame 2015;162(3):529e32. Ju Y, Sun W. Plasma assisted combustion: dynamics and chemistry. Prog Energy Combust Sci 2015;48:21e83. Messerle, VE, Ustimenko, AB, Karpenko, YE, Chernetskiy, MY, Dekterev, AA, Filimonov, SA. Modeling and full scale tests of vortex plasma fuel systems for igniting high ash power plant coal. Therm Eng. 62, 6, 442e451.