- Email: [email protected]
Local steam temperature imbalances of coal-fired boilers at very low load
Available online at www.sciencedirect.com Available online at www.sciencedirect.com
ScienceDirect ScienceDirect
Energy Procedia 00 (2017) 000–000 Available online www.sciencedirect.com Available online atatwww.sciencedirect.com Energy Procedia 00 (2017) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy (2017) 000–000 439–446 EnergyProcedia Procedia120 00 (2017) www.elsevier.com/locate/procedia
INFUB - 11th European Conference on Industrial Furnaces and Boilers, INFUB-11 INFUB - 11th European Conference on Industrial Furnaces and Boilers, INFUB-11
Local steam temperature imbalances of coal-fired boilers at very Local steam temperature imbalances of coal-fired boilers at very low load The 15th International Symposium on District Heating and Cooling low load a JensAssessing Hinrich Prause , Moritz Hübela, Dorian Holtzaa, Jürgen Nockeaa, Egon Hasselaa* the heat demand-outdoor a feasibilitya of using the Jens Hinrich Prause , Moritz Hübel , Dorian Holtz , Jürgen Nocke , Egon Hassel * University of Rostock, Albert-Einstein-Straße 2, 18059 Rostock, Germany temperature function for a long-term district heat demand forecast University of Rostock, Albert-Einstein-Straße 2, 18059 Rostock, Germany a
a
I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc a Abstract IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal b Abstract Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France c Département Systèmes Énergétiques et Environnement - IMTplants Atlantique, Alfred Kastler, 44300in Nantes, In order to remain economically viable, even large coal-fired power must 4berueoperated more often partialFrance load. Due to an In order to remain economically viable, coal-firedinpower plants must be operated moreThis often inaffect partialthe load. Due to an unfavourable coal mills configuration, theeven locallarge temperatures the combustion chamber can differ. can temperatures unfavourable mills configuration, local temperatures in the combustion chamber differ. This can affect the in two parallelcoal steam lines so much thatthe even a shut-down of the entire system would be can necessary. The modelling andtemperatures subsequent in two parallel steam so muchway thattoeven a shut-down theto entire would be necessary. The modelling and subsequent simulation of the plantlines is a proven identify problemsofand showsystem the limits of the minimum load. Abstract of the plant is a proven way to identify problems and to show the limits of the minimum load. simulation © 2017 The Authors. Published by Elsevier Ltd. ©District 2017 The Authors. Published Elsevier addressed Ltd. heating networks are by commonly in the literature as one of the most effective solutions for decreasing the © 2017 The Authors. Published by Elsevier Ltd. committee Peer-review under responsibility of the organizing of Peer-review under responsibility of the organizing committee of INFUB-11. INFUB-11 greenhouse under gas emissions from of thethe building sector. These systems require high investments which are returned through the heat Peer-review responsibility organizing committee of INFUB-11. sales. Due to the changed climate conditions and building renovation policies, heat demand in the future could decrease, Keywords: Coal-fired power plant; simulation; local temperatures; steam lines; minimum load prolonging the investment return period. Keywords: Coal-fired power plant; simulation; local temperatures; steam lines; minimum load The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 1. Introduction that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district 1.buildings Introduction renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were The German energyfrom market is currently undergoing fundamental change. andbymore wind and solar energy compared with results a dynamic heat demand model, apreviously developed and More validated the authors. The German energy marketpower. is currently undergoing a fundamental change. More power and more wind and solar energy isThe being converted into electric With the expansion of such highly fluctuating generator larger results showed that when only weather change is considered, the margin of error could be acceptable for facilities, some applications is being converted into electric power. With the expansion of such highly fluctuating power generator facilities, larger amplitudes load steps willwas be lower retrieved. must be provided mainly from thermal power plants. Consequently, (the error inofannual demand than These 20% for all weather scenarios considered). However, after introducing renovation amplitudes of load steps will be retrieved. These must be provided mainly from thermal power plants. Consequently, larger coal-fired power alsouphave to expect an increasingly unsteady power plant schedule. There considered). are many scenarios, the error valueplants increased to 59.5% (depending on the weather and renovation scenarios combination larger coal-fired plantsincreased also have expect an increasingly plant schedule. There areagain. many The value of slope ontoaverage theinrange ofunsteady 3.8% up power toof8% per decade, that corresponds to the situations where itpower iscoefficient beneficial to keep power plantswithin running low load instead shutting down and restarting situations where it is beneficial to keep power plants running in low load instead of shutting down and restarting again. decrease in the the number of heating hours maintaining of 22-139h during the heating season (depending on theancombination weather Decreasing minimum load, while a stable operation point, is therefore important offactor forand a renovation energy scenarios considered). the other hand, function intercept increased (depending Decreasing the minimum load, On while maintaining a of stable operation point,foris 7.8-12.7% therefore per an decade important factor on forthe a sustainable system including increasing shares fluctuating renewables. coupled scenarios). The values suggested could beshares used to the function parameters for the scenarios considered, and sustainable energy system including increasing of modify fluctuating renewables. improve the accuracy of heat demand estimations.
© 2017 The Authors. Published by Elsevier Ltd.
*Peer-review Corresponding author. Tel.: +49-381-4059-665; fax: +49-381-4059-657. under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and *E-mail Corresponding Tel.: +49-381-4059-665; fax: +49-381-4059-657. address: author. [email protected] Cooling. E-mail address: [email protected] 1876-6102 ©Heat 2017demand; The Authors. Published bychange Elsevier Ltd. Keywords: Forecast; Climate 1876-6102 © 2017responsibility The Authors.of Published by Elsevier Ltd. of INFUB-11. Peer-review under the organizing committee Peer-review under responsibility of the organizing committee of INFUB-11.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the organizing committee of INFUB-11 10.1016/j.egypro.2017.07.206
440 2
a
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000
b
Fig. 1. Composition of the German electric energy for a winter scenario in (a) 2011; (b) 2023.
Figure 1 shows the share of electrical energy for a typical winter week in 2011 and a forecast for the year 2023 of the different types of power plants which contribute to the demand for electrical energy in Germany [1]. While in the past the base load has been provided by nuclear power plants, lignite coal-fired power plants and, with some limitations, hard coal-fired power plants, these will only be a minor share in the future. The consumer load is represented in the graphics by the black line. Electrical energy cannot be stored efficiently and on an industrial scale, but supply and demand must be balanced at all times. The indicator for this is the mains frequency. Two different consequences can be observed. On the one hand, the residual load decreases, and on the other hand, the need for balancing power increases. In addition to economic considerations, this is the second reason why it is not always useful to completely shut down thermal power plants in windy and sunny times. Therefore, the use of the power plant at minimum load becomes more important. The investigation of such scenarios is an important part of today's energy research because it presents new challenges for thermal power plants. 2. Modelling All the investigations presented here are based on dynamic power plant simulation. By taking into account the transient thermal equilibrium in the balance equations, physical effects during the load change are visible and traceable. One potential use of the dynamic power plant simulation is to obtain accurate data on temperature and pressure gradients within components and thus to determine the loss of lifetime of system components by a load change. Another example would be to analyze the effects of soot blowers on the dynamic behavior of power plants. Such investigations can be found, in [2] [3] and [4]. This investigation is a purely static one. Therefore, all parameters in the following validation and result plots are related to a fixed power setpoint. The model is based on the open-source, object-oriented programming language Modelica. As a development and simulation environment the commercial tool Dymola is used. Dymola supplies some model libraries as well as the numerical equation solvers. The DASSL (Differential-Algebraic System Solver) has been used for all simulation scenarios. Most models for the components of the power plant come from the non-commercial software ClaRa library(Clausius-Rankine-Cycle). Important material data are from the TILMedia library. For each module, the mass and energy balance is solved for each simulation step. It is constituted as follows:
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause / Energy Procedia 00 (2017) 000–000
441 3
Mass balance: ∑ 𝑚𝑚̇𝑘𝑘 = 𝑘𝑘
Energy balance:
d𝑚𝑚 d𝑡𝑡
(1)
∑ 𝑄𝑄̇𝑖𝑖 + ∑ 𝑊𝑊̇t,𝑗𝑗 + ∑ 𝑚𝑚̇𝑘𝑘 (ℎ𝑘𝑘 + 𝑖𝑖
𝑗𝑗
𝑘𝑘
𝑐𝑐𝑘𝑘2 d𝑈𝑈 + 𝑔𝑔𝑔𝑔𝑘𝑘 ) = 2 d𝑡𝑡
(2)
The heat flows 𝑄𝑄̇𝑖𝑖 , the technical power 𝑊𝑊̇𝑡𝑡,𝑗𝑗 , the enthalpy flows 𝑚𝑚̇𝑘𝑘 ℎ𝑘𝑘 as well as the kinetic and potential energies, which are transferred at the system borders, are taken into account. During the heat transfer the convection, the heat conduction and the heat radiation are considered in the model. The convective heat transfer for turbulent pipe flows is taken into account, according to Gnielinski: 𝑞𝑞̇ 𝐶𝐶 = 𝛼𝛼 ∙ ∆𝑇𝑇𝑚𝑚 𝛼𝛼 =
𝑁𝑁𝑁𝑁 ∙ 𝜆𝜆 𝑑𝑑
𝑁𝑁𝑁𝑁𝑥𝑥,𝑇𝑇 =
(3) (4) (𝜉𝜉 ∙ 8−1 ) ∙ 𝑅𝑅𝑅𝑅 ∙ 𝑃𝑃𝑃𝑃
1 + 12,7 ∙ √𝜉𝜉 ∙ 8−1 ∙
𝜉𝜉 = (1,8 ∙ 𝑙𝑙𝑙𝑙𝑙𝑙10 (𝑅𝑅𝑅𝑅) − 1,5)−2
2 (𝑃𝑃𝑃𝑃 3
− 1)
2 𝑑𝑑𝑖𝑖 3
1 [1 + ( ) ] 3 𝑥𝑥
(5) (6)
Further models used in the simulations can be found in [5] and [6]. In addition to the models for heat transfer, all geometries, material characteristics, characteristic fields for pumps and turbines and, if possible, the entire unit control of the reference power plant are incorporated into the model. To reflect the reality as precisely as possible, the model is calibrated at the end of the parameterization by using measurement data. The result is a fully physical 0D-/1D power plant model, which is individually tailored to the respective power station. Stationary results can also be achieved by dynamic power plant simulation. 3. Reference Power Plant The reference power plant (see Figure 2) is a large lignite coal power plant with a net output of approximately 800 MW per unit. It operates in sliding pressure mode. The combustion chamber is located in the lower part of the boiler. There are 24 burners at three different heights. The dust firing is supplied with lignite through a maximum of eight coal mills arranged symmetrically around the combustion chamber. The tubes of the evaporator are arranged around the inner wall of the combustion chamber. In the upper part of the boiler are the heating surfaces of the four superheaters, the economiser and the reheaters. Between the heating surfaces of the first reheater and the second reheater, between the superheaters two and three, and between the superheaters three and four are injection coolers implemented to control a load-dependent temperature difference within the heating surface.
Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000 Egon Hassel et al. / Energy Procedia 120 (2017) 439–446
4 442
Fig. 2. Scheme of the thermal process of the reference power plant.
In order to be able to use simulation data in a meaningful manner, it is useful to validate them, that is, to compare them with measured values. Figure 3 shows the steam temperature and the steam pressure across the boiler for the stationary full load point as well as for a firing capacity of 60%. The accordance of the results allows the simulation to be extended to other power ranges for which no measured data are available. Through the model of the plant, parameters can be varied, new control strategies can be implemented, special characteristics in the operational measurement data can be analysed, retrofits can be assessed and the influence of certain assemblies on the overall system can be determined. Further information on the possibilities of dynamic simulation can be found in [7]. For dynamic considerations of the system, for example during a load change or start-up or shut-down process, the simulation variables must be validated with corresponding dynamic scenarios. a)
Jens Hinrich Prause / Energy Procedia 00 (2017) 000–000 Egon Hassel et al. / Energy Procedia 120 (2017) 439–446
5 443
b)
Fig. 3. Stationary validation of the (a) steam temperature and (b) pressure in the boiler at 60% and 100% firing capacity.
4. Parallel steam lines in the superheater The design of large-scale boilers separates the steam in the superheater area into parallel lines. There are four parallel lines in the reference power plant. External strands have a three times larger contact area with the surrounding wall (see figure 4). The wall of the boiler in the reference power plant represents the first superheater surface. The steam in the first superheater has a lower temperature, than in the other superheaters. A greater heat flow develops from the wall strands to the wall. Due to this reason, the outer steam lines - the lines close to the walls – become cooler than the centre lines of the boiler. Resulting from the lower temperature in the outer line, the density increases and the volume flow falls. Subsequently, the pressure drop in this line is reduced in comparison to the centre lines, which leads to an even higher mass flow and correlates with an even lower temperature in the outer line. Vice versa, the temperature in the centre lines increases. This effect is also known in the literature and is presented for example in [8].
Fig. 4. Top view of the boiler. Centre and wall line and the corresponding wall surfaces of the first superheater.
444 6
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000
The security system of the boiler monitors these temperature differences in parallel steam lines and can cause an emergency shut-down in case the predefined limits are exceeded. In order to reduce this effect, the parallel strands are mutually mixed (see figure 5). This is done by intersecting steam lines. The injection coolers are also located in each of the intersecting steam lines. In total, there are eight injection coolers in the superheater section. Behind the headers and the injection coolers, which regulate the temperature difference over the corresponding heating surface, the temperature differences are small. The greatest differences are obtained directly behind the heating surface when comparing an outer to an inner line. This process is strongly load-dependent. At low load, the spread of the measured values between the strands increases (see Figure 6). This is mainly due to two reasons. Firstly, the system becomes more vulnerable to local temperature differences between the steam lines, and secondly, the heat flow from one of the outer lines to the wall becomes relatively greater. The largest measured deviation occurs after the third superheater. This is mainly due to the structural size of the superheater. To illustrate this spatial effect within the 0D-/1D simulation software, the strands were modelled individually. The only difference between the lines is the diverging ambient wall surface. Thus, the effects mentioned above also occur in the model. The trend of higher temperature differences at low load is observed in the simulation as well as in the measurement data. Only after the first superheater is the ratio reversed. This is because the first superheater is the wall superheater. In the reference power plant, one line of the first superheater is located in each side wall of the boiler; assuming a uniform distribution of the temperature in the boiler, the outlet temperatures of all four strands must be identical. In the model, a larger wall surface is proportionally allocated to the outer lines and thus a larger transferred heat flow occurs. For all inlet and outlet headers a perfect mixing is presupposed and therefore no temperature difference can be seen in the simulation at these measuring points. At 40% firing power there are no measured data, this is a pure simulation case.
Fig. 5. Steam lines in the superheater.
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause / Energy Procedia 00 (2017) 000–000
445 7
Fig. 6. Measurement data and simulation data of temperature differences of a wall line and a centre line for 40%, 60% and 100% firing capacity.
5. Conclusion Due to the modeling of the reference power plant the operation in partial load can be optimized. As a result, a better integration of renewable energies into the electricity grid can be guaranteed. For thermal power plants, the consequences of lowering the minimum load can be tested without any risks. As a result, power plant operators can commercialize a larger power amplitude in the future and operate economically in a larger power range. Due to the modeling of several strands, temperature differences in the parallel lines can also be shown in 0D-/1D simulations. Both the magnitude and the trend can be represented. Through the dynamic simulation, it is also possible to observe temperature imbalances even during load changes and, if necessary, to keep them as small as possible by modifying the injection cooler control. 6. Outlook In the reference power plant, the coal mills are connected directly to the burners. During the operation of the plant in partial load, coal mills and burners are gradually switched off. When operating the plant with a small, uneven number of coal mills and the corresponding number of burners, local temperature imbalances within the boiler are more likely. The problem of the line temperature differences will therefore be further intensified in a lowered partial load. Likewise, an inhomogeneous mass flow distribution within the boiler can lead to local different deposits, which in turn would lead to a divergent heat flow in the steamlines. A next step is therefore the determination of temperature fields of the flue gas at the level of the superheater part. For this purpose, the boiler must be spatially discretized. By coupling a 3D model with a 0D-/1D model, more exact results can be expected for this question. The gain in knowledge must always be compared with the disadvantage of a significantly increasing computing time.
Fig. 7. positions of the coal mills around the boiler
Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000
8
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446
446
Nomenclature 𝑄𝑄̇𝑖𝑖 𝑊𝑊̇𝑡𝑡,𝑗𝑗 𝑚𝑚̇𝑘𝑘 ℎ𝑘𝑘 𝑞𝑞̇ 𝐶𝐶 𝛼𝛼 𝑇𝑇𝑚𝑚 𝑁𝑁𝑁𝑁 λ d 𝜉𝜉 Re Pr 𝑑𝑑𝑖𝑖
heat flow technical power mass flow specific enthalpy heat flux density heat transfer coefficient mean temperature Nusselt number thermal conductivity diameter discharge coefficient Reynolds number Prandtl number inner diameter
References [1] Ziems, C., Meinke, S., Nocke, J., Weber, H., & Hassel, E. (2012). Kraftwerksbetrieb bei Einspeisung von Windparks und Photovoltaikanlagen. VGB Bericht 333. Rostock: VGB. [2] Hübel, M., Meinke, S., Nocke, J., & Hassel, E. (2015). Identification of Energy Storage Capacities within large-scale Power Plants and Development of Control Strategies to increase marketable Grid Services. San Diego: ASME 2015 Power and Energy Conversion Conference. [3] Berndt, A., Richter, M., Hübel, M., Mutschler, P., Hassel, E., Weber, H., et al. (2014). Regelleistungs-Verschleißmodell für primär- und sekundärgeregelte thermische Kraftwerke im ENTSO-E-Netz. Rostock: VGB PowerTech Journal. [4] Gierow, C., Hübel, M., Nocke, J., Hassel, E. (2015). Mathematical Model of Soot Blowing Influences in Dynamic Power Plant Modelling, 11th International Modelica Conference, 21.-23. September 2015, Versailles, France, Paper. [5] Brunnemann, J., Gottelt, F., Wellner, K., Renz, A., Thuering, A., Roeder, V., Hasenbein, C. (2012). Status of ClaRaCCS: Modelling and simulation of coal-fired power plants with CO2 capture. Proceedings of the 9th International Modelica Conference. [6] ClaRa, ClaRa – Simulation of Clausius-Rankine cycles. www.claralib.com [7] Gottelt, F., Ziems, C., Meinke, S., Haase, M., Nocke, J., & Weber, H. (2009). Auswirkungen von fluktuierendener Windenergieeinspeisung auf das regel-und thermodynamische Betriebsverhalten konventioneller Kraftwerke in Deutschland Bestandsaufnahme und Ableitung zukünftiger Anforderungen. Rostock. VGB Bericht 283. Rostock: VGB. [8] Effenberger, H. (1999). Dampferzeugung. Berlin, Heidelberg: Springer.
ScienceDirect ScienceDirect
Energy Procedia 00 (2017) 000–000 Available online www.sciencedirect.com Available online atatwww.sciencedirect.com Energy Procedia 00 (2017) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy (2017) 000–000 439–446 EnergyProcedia Procedia120 00 (2017) www.elsevier.com/locate/procedia
INFUB - 11th European Conference on Industrial Furnaces and Boilers, INFUB-11 INFUB - 11th European Conference on Industrial Furnaces and Boilers, INFUB-11
Local steam temperature imbalances of coal-fired boilers at very Local steam temperature imbalances of coal-fired boilers at very low load The 15th International Symposium on District Heating and Cooling low load a JensAssessing Hinrich Prause , Moritz Hübela, Dorian Holtzaa, Jürgen Nockeaa, Egon Hasselaa* the heat demand-outdoor a feasibilitya of using the Jens Hinrich Prause , Moritz Hübel , Dorian Holtz , Jürgen Nocke , Egon Hassel * University of Rostock, Albert-Einstein-Straße 2, 18059 Rostock, Germany temperature function for a long-term district heat demand forecast University of Rostock, Albert-Einstein-Straße 2, 18059 Rostock, Germany a
a
I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc a Abstract IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal b Abstract Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France c Département Systèmes Énergétiques et Environnement - IMTplants Atlantique, Alfred Kastler, 44300in Nantes, In order to remain economically viable, even large coal-fired power must 4berueoperated more often partialFrance load. Due to an In order to remain economically viable, coal-firedinpower plants must be operated moreThis often inaffect partialthe load. Due to an unfavourable coal mills configuration, theeven locallarge temperatures the combustion chamber can differ. can temperatures unfavourable mills configuration, local temperatures in the combustion chamber differ. This can affect the in two parallelcoal steam lines so much thatthe even a shut-down of the entire system would be can necessary. The modelling andtemperatures subsequent in two parallel steam so muchway thattoeven a shut-down theto entire would be necessary. The modelling and subsequent simulation of the plantlines is a proven identify problemsofand showsystem the limits of the minimum load. Abstract of the plant is a proven way to identify problems and to show the limits of the minimum load. simulation © 2017 The Authors. Published by Elsevier Ltd. ©District 2017 The Authors. Published Elsevier addressed Ltd. heating networks are by commonly in the literature as one of the most effective solutions for decreasing the © 2017 The Authors. Published by Elsevier Ltd. committee Peer-review under responsibility of the organizing of Peer-review under responsibility of the organizing committee of INFUB-11. INFUB-11 greenhouse under gas emissions from of thethe building sector. These systems require high investments which are returned through the heat Peer-review responsibility organizing committee of INFUB-11. sales. Due to the changed climate conditions and building renovation policies, heat demand in the future could decrease, Keywords: Coal-fired power plant; simulation; local temperatures; steam lines; minimum load prolonging the investment return period. Keywords: Coal-fired power plant; simulation; local temperatures; steam lines; minimum load The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 1. Introduction that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district 1.buildings Introduction renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were The German energyfrom market is currently undergoing fundamental change. andbymore wind and solar energy compared with results a dynamic heat demand model, apreviously developed and More validated the authors. The German energy marketpower. is currently undergoing a fundamental change. More power and more wind and solar energy isThe being converted into electric With the expansion of such highly fluctuating generator larger results showed that when only weather change is considered, the margin of error could be acceptable for facilities, some applications is being converted into electric power. With the expansion of such highly fluctuating power generator facilities, larger amplitudes load steps willwas be lower retrieved. must be provided mainly from thermal power plants. Consequently, (the error inofannual demand than These 20% for all weather scenarios considered). However, after introducing renovation amplitudes of load steps will be retrieved. These must be provided mainly from thermal power plants. Consequently, larger coal-fired power alsouphave to expect an increasingly unsteady power plant schedule. There considered). are many scenarios, the error valueplants increased to 59.5% (depending on the weather and renovation scenarios combination larger coal-fired plantsincreased also have expect an increasingly plant schedule. There areagain. many The value of slope ontoaverage theinrange ofunsteady 3.8% up power toof8% per decade, that corresponds to the situations where itpower iscoefficient beneficial to keep power plantswithin running low load instead shutting down and restarting situations where it is beneficial to keep power plants running in low load instead of shutting down and restarting again. decrease in the the number of heating hours maintaining of 22-139h during the heating season (depending on theancombination weather Decreasing minimum load, while a stable operation point, is therefore important offactor forand a renovation energy scenarios considered). the other hand, function intercept increased (depending Decreasing the minimum load, On while maintaining a of stable operation point,foris 7.8-12.7% therefore per an decade important factor on forthe a sustainable system including increasing shares fluctuating renewables. coupled scenarios). The values suggested could beshares used to the function parameters for the scenarios considered, and sustainable energy system including increasing of modify fluctuating renewables. improve the accuracy of heat demand estimations.
© 2017 The Authors. Published by Elsevier Ltd.
*Peer-review Corresponding author. Tel.: +49-381-4059-665; fax: +49-381-4059-657. under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and *E-mail Corresponding Tel.: +49-381-4059-665; fax: +49-381-4059-657. address: author. [email protected] Cooling. E-mail address: [email protected] 1876-6102 ©Heat 2017demand; The Authors. Published bychange Elsevier Ltd. Keywords: Forecast; Climate 1876-6102 © 2017responsibility The Authors.of Published by Elsevier Ltd. of INFUB-11. Peer-review under the organizing committee Peer-review under responsibility of the organizing committee of INFUB-11.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the organizing committee of INFUB-11 10.1016/j.egypro.2017.07.206
440 2
a
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000
b
Fig. 1. Composition of the German electric energy for a winter scenario in (a) 2011; (b) 2023.
Figure 1 shows the share of electrical energy for a typical winter week in 2011 and a forecast for the year 2023 of the different types of power plants which contribute to the demand for electrical energy in Germany [1]. While in the past the base load has been provided by nuclear power plants, lignite coal-fired power plants and, with some limitations, hard coal-fired power plants, these will only be a minor share in the future. The consumer load is represented in the graphics by the black line. Electrical energy cannot be stored efficiently and on an industrial scale, but supply and demand must be balanced at all times. The indicator for this is the mains frequency. Two different consequences can be observed. On the one hand, the residual load decreases, and on the other hand, the need for balancing power increases. In addition to economic considerations, this is the second reason why it is not always useful to completely shut down thermal power plants in windy and sunny times. Therefore, the use of the power plant at minimum load becomes more important. The investigation of such scenarios is an important part of today's energy research because it presents new challenges for thermal power plants. 2. Modelling All the investigations presented here are based on dynamic power plant simulation. By taking into account the transient thermal equilibrium in the balance equations, physical effects during the load change are visible and traceable. One potential use of the dynamic power plant simulation is to obtain accurate data on temperature and pressure gradients within components and thus to determine the loss of lifetime of system components by a load change. Another example would be to analyze the effects of soot blowers on the dynamic behavior of power plants. Such investigations can be found, in [2] [3] and [4]. This investigation is a purely static one. Therefore, all parameters in the following validation and result plots are related to a fixed power setpoint. The model is based on the open-source, object-oriented programming language Modelica. As a development and simulation environment the commercial tool Dymola is used. Dymola supplies some model libraries as well as the numerical equation solvers. The DASSL (Differential-Algebraic System Solver) has been used for all simulation scenarios. Most models for the components of the power plant come from the non-commercial software ClaRa library(Clausius-Rankine-Cycle). Important material data are from the TILMedia library. For each module, the mass and energy balance is solved for each simulation step. It is constituted as follows:
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause / Energy Procedia 00 (2017) 000–000
441 3
Mass balance: ∑ 𝑚𝑚̇𝑘𝑘 = 𝑘𝑘
Energy balance:
d𝑚𝑚 d𝑡𝑡
(1)
∑ 𝑄𝑄̇𝑖𝑖 + ∑ 𝑊𝑊̇t,𝑗𝑗 + ∑ 𝑚𝑚̇𝑘𝑘 (ℎ𝑘𝑘 + 𝑖𝑖
𝑗𝑗
𝑘𝑘
𝑐𝑐𝑘𝑘2 d𝑈𝑈 + 𝑔𝑔𝑔𝑔𝑘𝑘 ) = 2 d𝑡𝑡
(2)
The heat flows 𝑄𝑄̇𝑖𝑖 , the technical power 𝑊𝑊̇𝑡𝑡,𝑗𝑗 , the enthalpy flows 𝑚𝑚̇𝑘𝑘 ℎ𝑘𝑘 as well as the kinetic and potential energies, which are transferred at the system borders, are taken into account. During the heat transfer the convection, the heat conduction and the heat radiation are considered in the model. The convective heat transfer for turbulent pipe flows is taken into account, according to Gnielinski: 𝑞𝑞̇ 𝐶𝐶 = 𝛼𝛼 ∙ ∆𝑇𝑇𝑚𝑚 𝛼𝛼 =
𝑁𝑁𝑁𝑁 ∙ 𝜆𝜆 𝑑𝑑
𝑁𝑁𝑁𝑁𝑥𝑥,𝑇𝑇 =
(3) (4) (𝜉𝜉 ∙ 8−1 ) ∙ 𝑅𝑅𝑅𝑅 ∙ 𝑃𝑃𝑃𝑃
1 + 12,7 ∙ √𝜉𝜉 ∙ 8−1 ∙
𝜉𝜉 = (1,8 ∙ 𝑙𝑙𝑙𝑙𝑙𝑙10 (𝑅𝑅𝑅𝑅) − 1,5)−2
2 (𝑃𝑃𝑃𝑃 3
− 1)
2 𝑑𝑑𝑖𝑖 3
1 [1 + ( ) ] 3 𝑥𝑥
(5) (6)
Further models used in the simulations can be found in [5] and [6]. In addition to the models for heat transfer, all geometries, material characteristics, characteristic fields for pumps and turbines and, if possible, the entire unit control of the reference power plant are incorporated into the model. To reflect the reality as precisely as possible, the model is calibrated at the end of the parameterization by using measurement data. The result is a fully physical 0D-/1D power plant model, which is individually tailored to the respective power station. Stationary results can also be achieved by dynamic power plant simulation. 3. Reference Power Plant The reference power plant (see Figure 2) is a large lignite coal power plant with a net output of approximately 800 MW per unit. It operates in sliding pressure mode. The combustion chamber is located in the lower part of the boiler. There are 24 burners at three different heights. The dust firing is supplied with lignite through a maximum of eight coal mills arranged symmetrically around the combustion chamber. The tubes of the evaporator are arranged around the inner wall of the combustion chamber. In the upper part of the boiler are the heating surfaces of the four superheaters, the economiser and the reheaters. Between the heating surfaces of the first reheater and the second reheater, between the superheaters two and three, and between the superheaters three and four are injection coolers implemented to control a load-dependent temperature difference within the heating surface.
Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000 Egon Hassel et al. / Energy Procedia 120 (2017) 439–446
4 442
Fig. 2. Scheme of the thermal process of the reference power plant.
In order to be able to use simulation data in a meaningful manner, it is useful to validate them, that is, to compare them with measured values. Figure 3 shows the steam temperature and the steam pressure across the boiler for the stationary full load point as well as for a firing capacity of 60%. The accordance of the results allows the simulation to be extended to other power ranges for which no measured data are available. Through the model of the plant, parameters can be varied, new control strategies can be implemented, special characteristics in the operational measurement data can be analysed, retrofits can be assessed and the influence of certain assemblies on the overall system can be determined. Further information on the possibilities of dynamic simulation can be found in [7]. For dynamic considerations of the system, for example during a load change or start-up or shut-down process, the simulation variables must be validated with corresponding dynamic scenarios. a)
Jens Hinrich Prause / Energy Procedia 00 (2017) 000–000 Egon Hassel et al. / Energy Procedia 120 (2017) 439–446
5 443
b)
Fig. 3. Stationary validation of the (a) steam temperature and (b) pressure in the boiler at 60% and 100% firing capacity.
4. Parallel steam lines in the superheater The design of large-scale boilers separates the steam in the superheater area into parallel lines. There are four parallel lines in the reference power plant. External strands have a three times larger contact area with the surrounding wall (see figure 4). The wall of the boiler in the reference power plant represents the first superheater surface. The steam in the first superheater has a lower temperature, than in the other superheaters. A greater heat flow develops from the wall strands to the wall. Due to this reason, the outer steam lines - the lines close to the walls – become cooler than the centre lines of the boiler. Resulting from the lower temperature in the outer line, the density increases and the volume flow falls. Subsequently, the pressure drop in this line is reduced in comparison to the centre lines, which leads to an even higher mass flow and correlates with an even lower temperature in the outer line. Vice versa, the temperature in the centre lines increases. This effect is also known in the literature and is presented for example in [8].
Fig. 4. Top view of the boiler. Centre and wall line and the corresponding wall surfaces of the first superheater.
444 6
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000
The security system of the boiler monitors these temperature differences in parallel steam lines and can cause an emergency shut-down in case the predefined limits are exceeded. In order to reduce this effect, the parallel strands are mutually mixed (see figure 5). This is done by intersecting steam lines. The injection coolers are also located in each of the intersecting steam lines. In total, there are eight injection coolers in the superheater section. Behind the headers and the injection coolers, which regulate the temperature difference over the corresponding heating surface, the temperature differences are small. The greatest differences are obtained directly behind the heating surface when comparing an outer to an inner line. This process is strongly load-dependent. At low load, the spread of the measured values between the strands increases (see Figure 6). This is mainly due to two reasons. Firstly, the system becomes more vulnerable to local temperature differences between the steam lines, and secondly, the heat flow from one of the outer lines to the wall becomes relatively greater. The largest measured deviation occurs after the third superheater. This is mainly due to the structural size of the superheater. To illustrate this spatial effect within the 0D-/1D simulation software, the strands were modelled individually. The only difference between the lines is the diverging ambient wall surface. Thus, the effects mentioned above also occur in the model. The trend of higher temperature differences at low load is observed in the simulation as well as in the measurement data. Only after the first superheater is the ratio reversed. This is because the first superheater is the wall superheater. In the reference power plant, one line of the first superheater is located in each side wall of the boiler; assuming a uniform distribution of the temperature in the boiler, the outlet temperatures of all four strands must be identical. In the model, a larger wall surface is proportionally allocated to the outer lines and thus a larger transferred heat flow occurs. For all inlet and outlet headers a perfect mixing is presupposed and therefore no temperature difference can be seen in the simulation at these measuring points. At 40% firing power there are no measured data, this is a pure simulation case.
Fig. 5. Steam lines in the superheater.
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446 Jens Hinrich Prause / Energy Procedia 00 (2017) 000–000
445 7
Fig. 6. Measurement data and simulation data of temperature differences of a wall line and a centre line for 40%, 60% and 100% firing capacity.
5. Conclusion Due to the modeling of the reference power plant the operation in partial load can be optimized. As a result, a better integration of renewable energies into the electricity grid can be guaranteed. For thermal power plants, the consequences of lowering the minimum load can be tested without any risks. As a result, power plant operators can commercialize a larger power amplitude in the future and operate economically in a larger power range. Due to the modeling of several strands, temperature differences in the parallel lines can also be shown in 0D-/1D simulations. Both the magnitude and the trend can be represented. Through the dynamic simulation, it is also possible to observe temperature imbalances even during load changes and, if necessary, to keep them as small as possible by modifying the injection cooler control. 6. Outlook In the reference power plant, the coal mills are connected directly to the burners. During the operation of the plant in partial load, coal mills and burners are gradually switched off. When operating the plant with a small, uneven number of coal mills and the corresponding number of burners, local temperature imbalances within the boiler are more likely. The problem of the line temperature differences will therefore be further intensified in a lowered partial load. Likewise, an inhomogeneous mass flow distribution within the boiler can lead to local different deposits, which in turn would lead to a divergent heat flow in the steamlines. A next step is therefore the determination of temperature fields of the flue gas at the level of the superheater part. For this purpose, the boiler must be spatially discretized. By coupling a 3D model with a 0D-/1D model, more exact results can be expected for this question. The gain in knowledge must always be compared with the disadvantage of a significantly increasing computing time.
Fig. 7. positions of the coal mills around the boiler
Jens Hinrich Prause/ Energy Procedia 00 (2017) 000–000
8
Egon Hassel et al. / Energy Procedia 120 (2017) 439–446
446
Nomenclature 𝑄𝑄̇𝑖𝑖 𝑊𝑊̇𝑡𝑡,𝑗𝑗 𝑚𝑚̇𝑘𝑘 ℎ𝑘𝑘 𝑞𝑞̇ 𝐶𝐶 𝛼𝛼 𝑇𝑇𝑚𝑚 𝑁𝑁𝑁𝑁 λ d 𝜉𝜉 Re Pr 𝑑𝑑𝑖𝑖
heat flow technical power mass flow specific enthalpy heat flux density heat transfer coefficient mean temperature Nusselt number thermal conductivity diameter discharge coefficient Reynolds number Prandtl number inner diameter
References [1] Ziems, C., Meinke, S., Nocke, J., Weber, H., & Hassel, E. (2012). Kraftwerksbetrieb bei Einspeisung von Windparks und Photovoltaikanlagen. VGB Bericht 333. Rostock: VGB. [2] Hübel, M., Meinke, S., Nocke, J., & Hassel, E. (2015). Identification of Energy Storage Capacities within large-scale Power Plants and Development of Control Strategies to increase marketable Grid Services. San Diego: ASME 2015 Power and Energy Conversion Conference. [3] Berndt, A., Richter, M., Hübel, M., Mutschler, P., Hassel, E., Weber, H., et al. (2014). Regelleistungs-Verschleißmodell für primär- und sekundärgeregelte thermische Kraftwerke im ENTSO-E-Netz. Rostock: VGB PowerTech Journal. [4] Gierow, C., Hübel, M., Nocke, J., Hassel, E. (2015). Mathematical Model of Soot Blowing Influences in Dynamic Power Plant Modelling, 11th International Modelica Conference, 21.-23. September 2015, Versailles, France, Paper. [5] Brunnemann, J., Gottelt, F., Wellner, K., Renz, A., Thuering, A., Roeder, V., Hasenbein, C. (2012). Status of ClaRaCCS: Modelling and simulation of coal-fired power plants with CO2 capture. Proceedings of the 9th International Modelica Conference. [6] ClaRa, ClaRa – Simulation of Clausius-Rankine cycles. www.claralib.com [7] Gottelt, F., Ziems, C., Meinke, S., Haase, M., Nocke, J., & Weber, H. (2009). Auswirkungen von fluktuierendener Windenergieeinspeisung auf das regel-und thermodynamische Betriebsverhalten konventioneller Kraftwerke in Deutschland Bestandsaufnahme und Ableitung zukünftiger Anforderungen. Rostock. VGB Bericht 283. Rostock: VGB. [8] Effenberger, H. (1999). Dampferzeugung. Berlin, Heidelberg: Springer.