- Email: [email protected]
Response to the Comments by Ronald DiPippo on “Efficiency of geothermal power plants: A worldwide review”
Geothermics 53 (2015) 550–553
Contents lists available at ScienceDirect
Geothermics journal homepage: www.elsevier.com/locate/geothermics
Reply to the Letter to the Editor Response to the Comments by Ronald DiPippo on “Efficiency of geothermal power plants: A worldwide review”
1. Introduction We would like to thank R. DiPippo for his comments. We will provide our rebuttal on all the points he raised. This work uses public domain data from existing and established geothermal power developments around the world, with references given for every data point used. The methodology used is conventional. It has been used for a long time, particularly when the efficiency of energy utilization is related to the geothermal reservoir temperature; please refer to Figure 2 of our paper and also Ogena and Freeston (1988). At the same time, having the efficiency a function of the resource enthalpy is also not new and has been reported in the 2010 Australian Geothermal Reporting Code, AGEA (2010). Our work shows that simple data fitting gives a good match (see our Figure 6). The focus of the work was made clear at more than one point in the paper. This work is to serve as a high level benchmark for geothermal resource assessment which requires an efficiency of conversion to electrical power based on total available thermal power (from the stored heat/volumetric method). This is also the case when calculating the power potential of new wells during discharge/output testing, where the average efficiency of 10% is commonly used (IEA, 2007; Barbier, 2002). We are surprised by R. DiPippo’s response in regards to the conversion efficiency based on the total available thermal power. It is commonly used for resource assessment studies. We recommend some references such as Grant and Bixley (2011); Hochstein and Crosetti (2011); AGEA (2010); Ogena and Freeston (1988); Watson and Maunder (1982) where this method was used. We do agree that geothermal energy is sui generis, but this does not mean we can only understand it through exergy. It also does not mean we cannot compare geothermal power plants with other thermal plants in terms of the input thermal power and electrical power output. Given that not every geothermal resource assessment study will lead to a power station, this method is easy and simple to use for high level evaluation studies. It can also be used for benchmarking new plants against existing geothermal plants, and it simplifies the comparison of geothermal with other types of thermal power plants.
DOI of original article: http://dx.doi.org/10.1016/j.geothermics.2014.06.006. http://dx.doi.org/10.1016/j.geothermics.2014.08.009 0375-6505/© 2014 Elsevier Ltd. All rights reserved.
2. Point 1 We disagree with R. DiPippo, and refer to our response in the introduction. For simplicity, we use the term ‘heat’ to mean thermal power (in kWth or MWth ). We do not mean heat transfer. There are a few points in the paper where we refer to heat transfer or heat loss explicitly. Eq. (1) is thermodynamically correct as it follows the second law of thermodynamics, while accounting for all those losses discussed in the paper. It is similar to the alternative Eq. (1) given by R. DiPippo. The difference is that it uses total thermal power produced from the reservoir (with reference to the triple point of water) rather than exergy which refers to the surroundings or the “dead state”.
3. Point 2 In our opinion, calculating conversion efficiency should not be limited to using just the “utilization efficiency” involving exergy. Using exergy just because it gives a higher and perhaps more “respectable” value than the conversion efficiency discussed in our work, is not, in our view, a justifiable approach. R. DiPippo states that Eq. (1) given in our work is “thermodynamically incorrect”. Then later (in point 3 of his commentary he stated that “it is rather problematic, unconventional and at worst unjustifiable”. However, he provides an example using the same method. This implies that the methodology is not “incorrect”, but simply an alternative approach. Applying the simpler enthalpy-based method, a calculated conversion efficiency of 14.1% is obtained for the example given from the Geysers. We find that the conversion efficiency for geothermal steam turbines has not improved much from that of 1964; it has possibly increased by another 5–6% at the most if new plant equipment are used. We understand that it is the relatively small value of conversion efficiency, compared to that of utilization efficiency from the exergy analysis that R. DiPippo is not comfortable with, as it “puts geothermal power plants in a bad light”. We would like to point out that the net electrical power (MWe) produced by the power plant will not actually be much affected whichever method is used. Geothermal power plants can be considered closed cycles if the reservoir is included, which is what we have done in this work. In this way, conversion efficiency can be compared with other thermal power plants by simply dividing the net MWe produced by the input thermal power MWth from the geothermal reservoir. R. DiPippo repeatedly criticizes the terminology used in our paper (heat, heat rate, heat input). We do agree that this can be a cause of some confusion. However, we would like to point out that simple dimensional analysis shows that there is nothing fundamentally wrong with our work. At the same time, local preferences to the terminology used can vary, depending on what is considered normal. However, it is a point that we have taken on board.
Reply to the Letter to the Editor / Geothermics 53 (2015) 550–553
4. Point 3 Most of the commentary has been addressed in the previous points. Again, this method is conventional and more widely used than the utilization efficiency of geothermal plant described by R. DiPippo in his Eq. (1). Most of the available reported public-domain data did not provide the ambient temperature of the surroundings, i.e. the “dead state”. This “dead state” also changes with the seasons throughout the year. The simple method discussed in our work uses the total thermal power produced from the geothermal reservoir. This is the product of the mass flow rate (kg/s) and the enthalpy (kJ/kg), with the reference being the zero enthalpy of liquid water at the triple point of water. Note that there are a host of factors affecting the conversion efficiency, as discussed in our work, and our method is simply a lumped approach that requires the least number of input parameters. We feel this should be of greatest value in the early stages of development, but it also provides a simple reference to other plants in service and during development. It is also of value when screening data; this will be demonstrated later in our response to point 6.
551
The reference point (dead state) in our approach is taken to be the zero enthalpy of liquid at the triple point of water 0.01 ◦ C and 0.00616 bar (Wagner et al., 2000; IFC, 1967). This is a consistent reference point in most recent steam tables (Watson, 2013). Note that Williams et al. (2008) provide a plot of utilization efficiency as a function of temperature based on reported field data. This effectively serves the same purpose as our efficiency plots when carrying out resource assessment studies. Example: A geothermal well produces 100 kg/s from a liquid dominated geothermal reservoir at 250 ◦ C. Calculate the power potential of the well in MWe? 1. Conversion efficiency (thermal) method: ˙ R hR Power = c m (a) Single flash plant 0.085 × 100 × 1085.7 Power = = 9.2 MWe 1000 (b) Double flash plant 0.094 × 100 × 1085.7 Power = = 10.2 MWe 1000 2. Utilization efficiency (exergy) method: ˙ R eR Power = u m
5. Point 4 It was acknowledged in our work that the power plants working in cold environments are more efficient than those operating Power =
where
e = hR − h0 − T0 (sR − s0 )
(a) For a dead state of 15 ◦ C and using a utilization efficiency of 40% (Williams, 2008)
0.4 × 100 × [1085.7 − 63 − (15 + 273.15) × (2.793 − 0.224)] = 11.3 MWe 1000 (b) For a dead sate of 30 ◦ C and using a utilization efficiency of 40% (Williams, 2008)
Power =
0.4 × 100 × [1085.7 − 125.7 − (30 + 273.15) × (2.793 − 0.224)] = 9.8 MWe 1000
in warmer ambient air conditions. Figure 9 gives a simple example using the Carnot and Triangular efficiencies for two different ambient temperatures. Zarrouk et al. (2014) also shows (from field data) that the produced MWe increases with decline in ambient air temperature. Our work acknowledges the importance of Exergy analysis as a tool when optimizing the production from existing plants. We disagree that plant design “has to be” carried out using exergy analysis and hence our statement “Exergy analysis is normally performed”. Most existing geothermal power plants were not designed using exergy analysis. Discussing this further was outside the scope of our paper. Eqs. (1) and (2) given by R. DiPippo are the basis for the USGS stored heat/volumetric method of resource assessment (Williams et al., 2008; Garg and Combs, 2010). We refer, instead, to the method described in AGEA (2010):
We =
Hth Rf c L·F
(1)
where We is the power plant capacity in kWe; Hth is the theoretical available energy (kJ) in the reservoir from the volumetric method; Rf is the recovery factor; c is the conversion efficiency or conversion factor; F is the power plant load factor/capacity factor; L is the power plant life (converted into seconds). This is a simple case of two different approaches to geothermal resource assessment.
Please note that: the data given by (Williams, 2008) for calculating Utilization efficiency has large variability, and at the same time it does not account for different types of plant. We hope that this simple demonstration shows that the difference between the two methods is relatively small. 6. Point 5 As discussed in our work, there are three components of pressure loss as the geothermal reservoir fluid travels up inside the well (Watson, 2013). dP = dz
dP dz
grav
+
dP dz
accel
+
dP dz
(2) fric
Commonly, wellbore simulators are used to model the flow from the reservoir to the wellhead and different combinations of correlations for flow in wells using the Eq. (2) above (see McGuinness, 2014; Watson, 2013). Our comment was not only about the heat loss (heat transfer) through the casing and cement, but also the loss of pressure in the well due to the above three components which contributes further to enthalpy loss. It is our understanding that R. DiPippo agrees with the three pressure loss components in Eq. (2) above. Generally (in liquid dominated reservoirs) this will account for about 50 to 100 kJ/kg of enthalpy loss from the bottom of the well (reservoir) to the wellhead; depending on well depth, feed zone depth, casing configuration and others. The down-hole pump that consumes electricity to produce the geothermal fluid is simply another parasitic load (electrical power)
552
Reply to the Letter to the Editor / Geothermics 53 (2015) 550–553
used within the power development as discussed in our work. In this case the drop in enthalpy from the reservoir to the wellhead will be less than that given above since the down-hole pump will have to overcome these losses. Our understanding is that R. DiPippo’s view is the same, so we do not see any issue here. Again, we do not have to view the flow inside the wellbore in terms of exergy analysis only; numerical wellbore simulators are considered the normal industry method for doing this.
7. Point 6 It is obvious that the calculated conversion efficiency of the Cerro Prieto plant of 26% (calculated in our work) is unrealistically higher than that of any other reported plant in the world, including dry steam plants, especially considering the reported enthalpy of 1395 kJ/kg and the single-flash power-plant type. However, should this enthalpy be correct, then the most likely explanation is that the reported mass flow rate is less than the actual mass produced or the actual generated power is less than that reported. In our opinion this means that the reported data in the reference is incorrect. This is our reason for not using the data from Cerro Prieto in fitting Eq. (18). We also would like to refer to the recently commissioned and highly efficient (triple flash) Taonga (Nga Awa Purua) power station in New Zealand (referred to in R. DiPippo’s commentary) which uses higher enthalpy fluid (1560 kJ/kg) and produces at a total conversion efficiency of about 17% (calculated in our work). This is much less than that calculated for Cerro Prieto (26% from above). We recommend following the same simple methodology to check the available data. R. DiPippo would probably have discovered the error in the data reported on Cerro Prieto in DiPippo (2012). Although the pipelines, separators, flashing vessels and turbines are thermally insulated, there is no perfect insulation/insulator and there will always be some heat loss (heat transfer). The pipelines can extend for tens of kilometers in some fields resulting in significant thermal losses. These losses are known to increase in winter relative to summer in cold countries. Steam pipelines uses drain pots to release steam condensates, this represents thermal loss and also some mass loss. Note that these mass losses help with the scrubbing of the steam. This heat loss is a deviation from ideal adiabatic conditions. Exergy analysis can serve to identify points that require improvement in existing plant, such as the need to replace/improve insulation on the pipelines. Heat transfer is also used to calculate the thermal power lost from the pipelines and to design the optimum insulation thickness based on insulation cost and the savings from reduced heat loss. Hence our comment and recommendation on keeping the pipelines short. We excluded the 17% from the Taonga (Nga Awa Purua) power station because it is the only data point available from a triple flash plant. We also have excluded other data points from Lihir, Los Humeros and Hachi-jojima which we feel had been under reported or possibly using back pressure turbines.
8. Point 7 This is a simple case of data fitting. We do not see the need to make the function dimensionless. By simply applying the enthalpy (in kJ/kg) one can read off an efficiency value. We note that R. DiPippo agrees that the general shape of the fit is correct. This is similar to having the conversion efficiency as a function of temperature. Fitting the data points with a function allows the calculation of R2 which is important to present to the readers.
In the geothermal industry, and other industries, there are many applied engineering correlations that are based on empirical justification. Figure 6 demonstrates that data fitting gave a good match with an existing model (AGEG, 2010) for single flash and dry steam plants. For binary plants (Figures 10 and 11) it gave a model that sits between the work of DiPippo (2007) and that of Dickson and Fanelli (2003) and Hudson (1988). We like to emphasize that only SI units should be used. 9. Conclusions Our paper was not intended to reduce confidence in geothermal energy, nor mislead readers in any way. As academics and researchers, it is always our duty to report the facts as they are and we have followed a widely used approach to assess efficiency while working within the fundamental laws of thermodynamics. The authors would like to draw attention to the observation that countries and investors typically choose to develop geothermal power on economic and environmental grounds, but not by comparing energy conversion efficiency with other types of thermal plants. There are also a host of commercial, local and political considerations involved in such decisions. Geothermal power generation is known for its reliability and high availability compared with many other thermal plants or other renewable energy sources (e.g. solar and wind). Providing this simple comparison of conversion efficiency with other types of thermal plant can be of significant value and we have received positive feedback on this work. The fundamental difference between the approach of the authors and that of R. DiPippo is the difference in background. The authors approach efficiency from a holistic total geothermal reservoir engineering point of view. In R. DiPippo’s approach, the topic is viewed from the surface facility/plant side. Arguably, both approaches have their place. The authors would also like to stress that the current work is not an alternative to the established thermodynamic methods for power plant design. But, at the same time, we argue that there should not be an imposed restriction on using exergy analysis only. The authors claim that their work is fundamentally correct, valid, applied and should be of practical use to researchers and industry. We do accept the criticism of using the term “heat” instead “thermal power” produced from the reservoir. This was possibly our attempt to simplify the concepts. However, we argue that the remaining criticisms are unfounded, and we would like to leave it to the readers to make their own judgment. PS: We take this opportunity to point out that there is a simple typographical error (during publishing) in equation number 3 in our paper which should be: apc = 1 −
Wapc Wgross
We like to thank R. DiPippo again for his time, input and interest. References AGEA, 2010. Australian Code for Reporting of Exploration Results, Geothermal Resources and Geothermal Reserves, 2nd ed. http://www.pir.sa.gov.au/ data/ assets/pdf file/0005/147875/The Geothermal Reporting Code Ed 2.pdf Garg, S.K., Combs, J., 2010. Appropriate use of USGS volumetric “Heat in Place” method and Monte Carlo Calculations. In: Proceedings of the 34th Workshop on Geothermal Reservoir Engineering, Stanford University, California, 1–3 February, 2010. Grant, M.A., Bixley, P.F., 2011. Simulation. In: Geothermal Reservoir Engineering, 2nd ed. Academic Press, Boston, pp. 201–217 (Chapter 11). Hochstein, M.P., Crosetti, M., 2011. Electric power potential estimates of hightemperature geothermal fields in indonesia and the philippines (a historic review). In: Proceedings of the 33rd New Zealand Geothermal Workshop, Auckland, New Zealand.
Reply to the Letter to the Editor / Geothermics 53 (2015) 550–553 IFC, 1967. A Formulation of the Thermodynamic Properties of Ordinary Water Substance. International Formulation Committee, Düsseldorf, Germany. McGuinness, M.J., 2014. Correct energy conservation in geothermal wellbore simulation. Geothermics 51, 429–433. Ogena, M.S., Freeston, D.H., 1988. Sensitivity analysis of the greater Tongonan field resource assessment. In: Proceedings of the 10th New Zealand Geothermal Workshop, pp. 67–72. Wagner, W., Cooper, J.R., Dittman, A., Kijima, J., Kretzschmar, H.-J., Kruse, A., Mares, R., Oguchi, K., Sato, H., Stöcker, I., Sifner, O., Takaishi, Y., Tanishita, I., Trübenbach, J., Willkommen, Th., 2000. The IAPWS Industrial Formulation 1997 for the thermodynamic properties of water and steam. ASME J. Eng. Gas Turbines Power 122, 150–182. Watson, A., 2013. Geothermal Engineering: Fundamentals and Applications. Springer, ISBN 978-1-4614-8568-1, pp. 336. Watson, A., Maunder, B., 1982. Geothermal resource assessment for power station planning. In: Proceedings the 4th New Zealand Geothermal Workshop, pp. 75–79. Williams, F.C., Reed, M.J., Mariner, R.H., 2008. A Review of Methods Applied by the U.S. Geological Survey in the Assessment of Identified Geothermal Resources, Open-File Report 2008-1296, U.S. Department of the Interior, U.S. Geological Survey. Zarrouk, S.J., Woodhurst, B.C., Morris, C., 2014. Silica scaling in geothermal heat exchangers and its impact on pressure drop and performance: Wairakei Binary Plant, New Zealand. Geothermics 51, 445–459.
553
Sadiq J. Zarrouk ∗ Geothermal Engineering, Department of Engineering Science, The University of Auckland, New Zealand Hyungsul Moon Mighty River Power, 283 Vaughan Road, PO Box 245, Rotorua 3040, New Zealand
∗ Corresponding author at: Department of Engineering Science, the University of Auckland, Private Bag 92019, Auckland, New Zealand. Tel.: +64 9 373 7599x85542; fax: +64 9 373 7468. E-mail addresses: [email protected], [email protected] (S.J. Zarrouk).
Available online 30 August 2014
Contents lists available at ScienceDirect
Geothermics journal homepage: www.elsevier.com/locate/geothermics
Reply to the Letter to the Editor Response to the Comments by Ronald DiPippo on “Efficiency of geothermal power plants: A worldwide review”
1. Introduction We would like to thank R. DiPippo for his comments. We will provide our rebuttal on all the points he raised. This work uses public domain data from existing and established geothermal power developments around the world, with references given for every data point used. The methodology used is conventional. It has been used for a long time, particularly when the efficiency of energy utilization is related to the geothermal reservoir temperature; please refer to Figure 2 of our paper and also Ogena and Freeston (1988). At the same time, having the efficiency a function of the resource enthalpy is also not new and has been reported in the 2010 Australian Geothermal Reporting Code, AGEA (2010). Our work shows that simple data fitting gives a good match (see our Figure 6). The focus of the work was made clear at more than one point in the paper. This work is to serve as a high level benchmark for geothermal resource assessment which requires an efficiency of conversion to electrical power based on total available thermal power (from the stored heat/volumetric method). This is also the case when calculating the power potential of new wells during discharge/output testing, where the average efficiency of 10% is commonly used (IEA, 2007; Barbier, 2002). We are surprised by R. DiPippo’s response in regards to the conversion efficiency based on the total available thermal power. It is commonly used for resource assessment studies. We recommend some references such as Grant and Bixley (2011); Hochstein and Crosetti (2011); AGEA (2010); Ogena and Freeston (1988); Watson and Maunder (1982) where this method was used. We do agree that geothermal energy is sui generis, but this does not mean we can only understand it through exergy. It also does not mean we cannot compare geothermal power plants with other thermal plants in terms of the input thermal power and electrical power output. Given that not every geothermal resource assessment study will lead to a power station, this method is easy and simple to use for high level evaluation studies. It can also be used for benchmarking new plants against existing geothermal plants, and it simplifies the comparison of geothermal with other types of thermal power plants.
DOI of original article: http://dx.doi.org/10.1016/j.geothermics.2014.06.006. http://dx.doi.org/10.1016/j.geothermics.2014.08.009 0375-6505/© 2014 Elsevier Ltd. All rights reserved.
2. Point 1 We disagree with R. DiPippo, and refer to our response in the introduction. For simplicity, we use the term ‘heat’ to mean thermal power (in kWth or MWth ). We do not mean heat transfer. There are a few points in the paper where we refer to heat transfer or heat loss explicitly. Eq. (1) is thermodynamically correct as it follows the second law of thermodynamics, while accounting for all those losses discussed in the paper. It is similar to the alternative Eq. (1) given by R. DiPippo. The difference is that it uses total thermal power produced from the reservoir (with reference to the triple point of water) rather than exergy which refers to the surroundings or the “dead state”.
3. Point 2 In our opinion, calculating conversion efficiency should not be limited to using just the “utilization efficiency” involving exergy. Using exergy just because it gives a higher and perhaps more “respectable” value than the conversion efficiency discussed in our work, is not, in our view, a justifiable approach. R. DiPippo states that Eq. (1) given in our work is “thermodynamically incorrect”. Then later (in point 3 of his commentary he stated that “it is rather problematic, unconventional and at worst unjustifiable”. However, he provides an example using the same method. This implies that the methodology is not “incorrect”, but simply an alternative approach. Applying the simpler enthalpy-based method, a calculated conversion efficiency of 14.1% is obtained for the example given from the Geysers. We find that the conversion efficiency for geothermal steam turbines has not improved much from that of 1964; it has possibly increased by another 5–6% at the most if new plant equipment are used. We understand that it is the relatively small value of conversion efficiency, compared to that of utilization efficiency from the exergy analysis that R. DiPippo is not comfortable with, as it “puts geothermal power plants in a bad light”. We would like to point out that the net electrical power (MWe) produced by the power plant will not actually be much affected whichever method is used. Geothermal power plants can be considered closed cycles if the reservoir is included, which is what we have done in this work. In this way, conversion efficiency can be compared with other thermal power plants by simply dividing the net MWe produced by the input thermal power MWth from the geothermal reservoir. R. DiPippo repeatedly criticizes the terminology used in our paper (heat, heat rate, heat input). We do agree that this can be a cause of some confusion. However, we would like to point out that simple dimensional analysis shows that there is nothing fundamentally wrong with our work. At the same time, local preferences to the terminology used can vary, depending on what is considered normal. However, it is a point that we have taken on board.
Reply to the Letter to the Editor / Geothermics 53 (2015) 550–553
4. Point 3 Most of the commentary has been addressed in the previous points. Again, this method is conventional and more widely used than the utilization efficiency of geothermal plant described by R. DiPippo in his Eq. (1). Most of the available reported public-domain data did not provide the ambient temperature of the surroundings, i.e. the “dead state”. This “dead state” also changes with the seasons throughout the year. The simple method discussed in our work uses the total thermal power produced from the geothermal reservoir. This is the product of the mass flow rate (kg/s) and the enthalpy (kJ/kg), with the reference being the zero enthalpy of liquid water at the triple point of water. Note that there are a host of factors affecting the conversion efficiency, as discussed in our work, and our method is simply a lumped approach that requires the least number of input parameters. We feel this should be of greatest value in the early stages of development, but it also provides a simple reference to other plants in service and during development. It is also of value when screening data; this will be demonstrated later in our response to point 6.
551
The reference point (dead state) in our approach is taken to be the zero enthalpy of liquid at the triple point of water 0.01 ◦ C and 0.00616 bar (Wagner et al., 2000; IFC, 1967). This is a consistent reference point in most recent steam tables (Watson, 2013). Note that Williams et al. (2008) provide a plot of utilization efficiency as a function of temperature based on reported field data. This effectively serves the same purpose as our efficiency plots when carrying out resource assessment studies. Example: A geothermal well produces 100 kg/s from a liquid dominated geothermal reservoir at 250 ◦ C. Calculate the power potential of the well in MWe? 1. Conversion efficiency (thermal) method: ˙ R hR Power = c m (a) Single flash plant 0.085 × 100 × 1085.7 Power = = 9.2 MWe 1000 (b) Double flash plant 0.094 × 100 × 1085.7 Power = = 10.2 MWe 1000 2. Utilization efficiency (exergy) method: ˙ R eR Power = u m
5. Point 4 It was acknowledged in our work that the power plants working in cold environments are more efficient than those operating Power =
where
e = hR − h0 − T0 (sR − s0 )
(a) For a dead state of 15 ◦ C and using a utilization efficiency of 40% (Williams, 2008)
0.4 × 100 × [1085.7 − 63 − (15 + 273.15) × (2.793 − 0.224)] = 11.3 MWe 1000 (b) For a dead sate of 30 ◦ C and using a utilization efficiency of 40% (Williams, 2008)
Power =
0.4 × 100 × [1085.7 − 125.7 − (30 + 273.15) × (2.793 − 0.224)] = 9.8 MWe 1000
in warmer ambient air conditions. Figure 9 gives a simple example using the Carnot and Triangular efficiencies for two different ambient temperatures. Zarrouk et al. (2014) also shows (from field data) that the produced MWe increases with decline in ambient air temperature. Our work acknowledges the importance of Exergy analysis as a tool when optimizing the production from existing plants. We disagree that plant design “has to be” carried out using exergy analysis and hence our statement “Exergy analysis is normally performed”. Most existing geothermal power plants were not designed using exergy analysis. Discussing this further was outside the scope of our paper. Eqs. (1) and (2) given by R. DiPippo are the basis for the USGS stored heat/volumetric method of resource assessment (Williams et al., 2008; Garg and Combs, 2010). We refer, instead, to the method described in AGEA (2010):
We =
Hth Rf c L·F
(1)
where We is the power plant capacity in kWe; Hth is the theoretical available energy (kJ) in the reservoir from the volumetric method; Rf is the recovery factor; c is the conversion efficiency or conversion factor; F is the power plant load factor/capacity factor; L is the power plant life (converted into seconds). This is a simple case of two different approaches to geothermal resource assessment.
Please note that: the data given by (Williams, 2008) for calculating Utilization efficiency has large variability, and at the same time it does not account for different types of plant. We hope that this simple demonstration shows that the difference between the two methods is relatively small. 6. Point 5 As discussed in our work, there are three components of pressure loss as the geothermal reservoir fluid travels up inside the well (Watson, 2013). dP = dz
dP dz
grav
+
dP dz
accel
+
dP dz
(2) fric
Commonly, wellbore simulators are used to model the flow from the reservoir to the wellhead and different combinations of correlations for flow in wells using the Eq. (2) above (see McGuinness, 2014; Watson, 2013). Our comment was not only about the heat loss (heat transfer) through the casing and cement, but also the loss of pressure in the well due to the above three components which contributes further to enthalpy loss. It is our understanding that R. DiPippo agrees with the three pressure loss components in Eq. (2) above. Generally (in liquid dominated reservoirs) this will account for about 50 to 100 kJ/kg of enthalpy loss from the bottom of the well (reservoir) to the wellhead; depending on well depth, feed zone depth, casing configuration and others. The down-hole pump that consumes electricity to produce the geothermal fluid is simply another parasitic load (electrical power)
552
Reply to the Letter to the Editor / Geothermics 53 (2015) 550–553
used within the power development as discussed in our work. In this case the drop in enthalpy from the reservoir to the wellhead will be less than that given above since the down-hole pump will have to overcome these losses. Our understanding is that R. DiPippo’s view is the same, so we do not see any issue here. Again, we do not have to view the flow inside the wellbore in terms of exergy analysis only; numerical wellbore simulators are considered the normal industry method for doing this.
7. Point 6 It is obvious that the calculated conversion efficiency of the Cerro Prieto plant of 26% (calculated in our work) is unrealistically higher than that of any other reported plant in the world, including dry steam plants, especially considering the reported enthalpy of 1395 kJ/kg and the single-flash power-plant type. However, should this enthalpy be correct, then the most likely explanation is that the reported mass flow rate is less than the actual mass produced or the actual generated power is less than that reported. In our opinion this means that the reported data in the reference is incorrect. This is our reason for not using the data from Cerro Prieto in fitting Eq. (18). We also would like to refer to the recently commissioned and highly efficient (triple flash) Taonga (Nga Awa Purua) power station in New Zealand (referred to in R. DiPippo’s commentary) which uses higher enthalpy fluid (1560 kJ/kg) and produces at a total conversion efficiency of about 17% (calculated in our work). This is much less than that calculated for Cerro Prieto (26% from above). We recommend following the same simple methodology to check the available data. R. DiPippo would probably have discovered the error in the data reported on Cerro Prieto in DiPippo (2012). Although the pipelines, separators, flashing vessels and turbines are thermally insulated, there is no perfect insulation/insulator and there will always be some heat loss (heat transfer). The pipelines can extend for tens of kilometers in some fields resulting in significant thermal losses. These losses are known to increase in winter relative to summer in cold countries. Steam pipelines uses drain pots to release steam condensates, this represents thermal loss and also some mass loss. Note that these mass losses help with the scrubbing of the steam. This heat loss is a deviation from ideal adiabatic conditions. Exergy analysis can serve to identify points that require improvement in existing plant, such as the need to replace/improve insulation on the pipelines. Heat transfer is also used to calculate the thermal power lost from the pipelines and to design the optimum insulation thickness based on insulation cost and the savings from reduced heat loss. Hence our comment and recommendation on keeping the pipelines short. We excluded the 17% from the Taonga (Nga Awa Purua) power station because it is the only data point available from a triple flash plant. We also have excluded other data points from Lihir, Los Humeros and Hachi-jojima which we feel had been under reported or possibly using back pressure turbines.
8. Point 7 This is a simple case of data fitting. We do not see the need to make the function dimensionless. By simply applying the enthalpy (in kJ/kg) one can read off an efficiency value. We note that R. DiPippo agrees that the general shape of the fit is correct. This is similar to having the conversion efficiency as a function of temperature. Fitting the data points with a function allows the calculation of R2 which is important to present to the readers.
In the geothermal industry, and other industries, there are many applied engineering correlations that are based on empirical justification. Figure 6 demonstrates that data fitting gave a good match with an existing model (AGEG, 2010) for single flash and dry steam plants. For binary plants (Figures 10 and 11) it gave a model that sits between the work of DiPippo (2007) and that of Dickson and Fanelli (2003) and Hudson (1988). We like to emphasize that only SI units should be used. 9. Conclusions Our paper was not intended to reduce confidence in geothermal energy, nor mislead readers in any way. As academics and researchers, it is always our duty to report the facts as they are and we have followed a widely used approach to assess efficiency while working within the fundamental laws of thermodynamics. The authors would like to draw attention to the observation that countries and investors typically choose to develop geothermal power on economic and environmental grounds, but not by comparing energy conversion efficiency with other types of thermal plants. There are also a host of commercial, local and political considerations involved in such decisions. Geothermal power generation is known for its reliability and high availability compared with many other thermal plants or other renewable energy sources (e.g. solar and wind). Providing this simple comparison of conversion efficiency with other types of thermal plant can be of significant value and we have received positive feedback on this work. The fundamental difference between the approach of the authors and that of R. DiPippo is the difference in background. The authors approach efficiency from a holistic total geothermal reservoir engineering point of view. In R. DiPippo’s approach, the topic is viewed from the surface facility/plant side. Arguably, both approaches have their place. The authors would also like to stress that the current work is not an alternative to the established thermodynamic methods for power plant design. But, at the same time, we argue that there should not be an imposed restriction on using exergy analysis only. The authors claim that their work is fundamentally correct, valid, applied and should be of practical use to researchers and industry. We do accept the criticism of using the term “heat” instead “thermal power” produced from the reservoir. This was possibly our attempt to simplify the concepts. However, we argue that the remaining criticisms are unfounded, and we would like to leave it to the readers to make their own judgment. PS: We take this opportunity to point out that there is a simple typographical error (during publishing) in equation number 3 in our paper which should be: apc = 1 −
Wapc Wgross
We like to thank R. DiPippo again for his time, input and interest. References AGEA, 2010. Australian Code for Reporting of Exploration Results, Geothermal Resources and Geothermal Reserves, 2nd ed. http://www.pir.sa.gov.au/ data/ assets/pdf file/0005/147875/The Geothermal Reporting Code Ed 2.pdf Garg, S.K., Combs, J., 2010. Appropriate use of USGS volumetric “Heat in Place” method and Monte Carlo Calculations. In: Proceedings of the 34th Workshop on Geothermal Reservoir Engineering, Stanford University, California, 1–3 February, 2010. Grant, M.A., Bixley, P.F., 2011. Simulation. In: Geothermal Reservoir Engineering, 2nd ed. Academic Press, Boston, pp. 201–217 (Chapter 11). Hochstein, M.P., Crosetti, M., 2011. Electric power potential estimates of hightemperature geothermal fields in indonesia and the philippines (a historic review). In: Proceedings of the 33rd New Zealand Geothermal Workshop, Auckland, New Zealand.
Reply to the Letter to the Editor / Geothermics 53 (2015) 550–553 IFC, 1967. A Formulation of the Thermodynamic Properties of Ordinary Water Substance. International Formulation Committee, Düsseldorf, Germany. McGuinness, M.J., 2014. Correct energy conservation in geothermal wellbore simulation. Geothermics 51, 429–433. Ogena, M.S., Freeston, D.H., 1988. Sensitivity analysis of the greater Tongonan field resource assessment. In: Proceedings of the 10th New Zealand Geothermal Workshop, pp. 67–72. Wagner, W., Cooper, J.R., Dittman, A., Kijima, J., Kretzschmar, H.-J., Kruse, A., Mares, R., Oguchi, K., Sato, H., Stöcker, I., Sifner, O., Takaishi, Y., Tanishita, I., Trübenbach, J., Willkommen, Th., 2000. The IAPWS Industrial Formulation 1997 for the thermodynamic properties of water and steam. ASME J. Eng. Gas Turbines Power 122, 150–182. Watson, A., 2013. Geothermal Engineering: Fundamentals and Applications. Springer, ISBN 978-1-4614-8568-1, pp. 336. Watson, A., Maunder, B., 1982. Geothermal resource assessment for power station planning. In: Proceedings the 4th New Zealand Geothermal Workshop, pp. 75–79. Williams, F.C., Reed, M.J., Mariner, R.H., 2008. A Review of Methods Applied by the U.S. Geological Survey in the Assessment of Identified Geothermal Resources, Open-File Report 2008-1296, U.S. Department of the Interior, U.S. Geological Survey. Zarrouk, S.J., Woodhurst, B.C., Morris, C., 2014. Silica scaling in geothermal heat exchangers and its impact on pressure drop and performance: Wairakei Binary Plant, New Zealand. Geothermics 51, 445–459.
553
Sadiq J. Zarrouk ∗ Geothermal Engineering, Department of Engineering Science, The University of Auckland, New Zealand Hyungsul Moon Mighty River Power, 283 Vaughan Road, PO Box 245, Rotorua 3040, New Zealand
∗ Corresponding author at: Department of Engineering Science, the University of Auckland, Private Bag 92019, Auckland, New Zealand. Tel.: +64 9 373 7599x85542; fax: +64 9 373 7468. E-mail addresses: [email protected], [email protected] (S.J. Zarrouk).
Available online 30 August 2014