Wind Turbine Build-outs and CO2 emissions in Ireland.
F. Udo
February 9, 2015
Published in the Irish Energy Blog:
http://irishenergyblog.blogspot.nl/2015/03/wind-turbine-build-outs-and-co2.html
1. Introduction.
The Irish electricity distribution grid has seen a rapid evolution in the last 5 years. During the 2010 to 2014 period, 1,4 billion euro was spent to erect 835 MW wind turbines and 0,6 billion on a 500MW high voltage link to England.
This article tries to answer the question: Is this 2 billion euro well spent?
There is much debate about the efficacy of wind energy in supplying electricity to existing distribution grids. In principle it can be measured accurately by monitoring the fuel use of all the generators involved.
The problem is, that this data is not available in the public domain, so one resorts to model calculations using the static characteristics of the generators involved. Dynamic effects due to regulation of the generators and spinning reserves are mostly neglected. The published data about the CO2 emissions in the Irish electricity distribution grid is a case in point.
The Irish grid authority Eirgrid publishes every 15 minutes the total demand, the wind energy produced and CO2 emission calculated in a way as mentioned above).
Using this data) it was possible to show, that in the absence of hydropower in April 2011 the efficacy of insertion of wind power in the grid was less than 40%, or in other words: More than 60% of the wind energy produced did not save any fuel. Wheatley) showed by using detailed output data of generators, that the CO2 reduction in all 2011 was only 70% of the expected value. This is confirmed in the year report 2013 of the Sustainable Energy Authority (SEAI) ).
Nevertheless the CO2 emissions in the year 2011 were at an all time low due to the commissioning of two new CCGT gas power plants. Two questions remain:
A. How close are the results of the Eirgrid CO2 emission calculations to the actual emissions?
B. Did the investments lead to substantial fuel savings?
All numbers used in this article are taken from the websites of Eirgrid and SEAI.
-
2.The CO2 emission derived from fuel input data.
The Year Reports of the SEAI provide the total fuel input and the input fuel mix for the electricity generation per year). The calorific value of each fuel is expressed in a common unit: Kilo Tons Oil Equivalent or ktoe. The electricity produced is given in the same units to enable a comparison of the fuel input and the electricity output.
Appendix 1 presents the data also in this unit to enable a direct verification with the numbers given in the yearly reports of the SEAI. One ktoe = 11,6 GWh.
Appendix 2 presents the calculation of the CO2 intensity for 2006 starting from the composition of the input fuel mix. It is based upon the calorific values and specific CO2 emissions of the different combustibles and on the ratio of total fuel input to total electricity output.
Renewables are always attributed zero emissions in the CO2 balance, but windmills and wood pellets are far from CO2 free. Appendix 3 gives some data for wind turbines in order to show, that the results given in this paper are still underestimates of the real emissions.
The CO2 emissions calculated in Appendix 1 are compared for each year in the period 2006 -2013 to the SEAI numbers. The SEAI numbers given here are also corrected for imports & exports.
Figure 1 shows the comparison of the two sets of data.
The method of calculation of the SEAI numbers is not explicitly given, but the results of the two methods are nearly identical, so both datasets are most probably obtained in the same way. The gradual decline from 2006 to 2010 is explained by an increase of the contribution of gas to the fuel mix. The dip in 2011 is explained by SEAI as due to the commissioning of two new gas plants, which where under-used in subsequent years due to coal getting cheaper….
It is now possible to test the CO2 model calculations of Eirgrid against numbers, which are derived directly from fuel input data.
3. The comparison with the Eirgrid data.
Appendix 2 shows the numbers obtained by integrating the Eirgrid 15 minute data. The numbers are summed over the years 2010 to 2014
The calculated CO2 emission intensities can now be compared to the data derived from fuel inputs for the period 2010 to 2014. Figure 2 presents the comparison between the 3 data sets. The model data are derived from the 15’ data as described in appendix 2. It is clear, that the model underestimates the actual emissions by about 6%.
Note 1:
The fuel input data for 2014 are not yet published, but the Eirgrid 15-minute data clearly indicates that the result for 2014 will be around 520g CO2/kWh. This implies, that the performance of the electricity generating system has not improved between 2010 and 2014.
Note 2:
If the contribution of wind is 20%, the overall CO2 intensity rises by 10 g CO2/kWh. The CO2 content of biofuels is subject to an intense discussion, so this contribution has not been considered.
4. Conclusions
The first observation is, that the Eirgrid model underestimates the real emissions by about 6%. This validates the objections stated in the introduction against the procedure followed by Eirgrid. This proves, that the losses calculated in ref 2 are too low.
The second observation is, that the CO2 intensity has risen significantly since 2010/2011 despite two billion euro spent on an increase in wind turbine capacity from 1260 MW to 2211 MW between 2010 and 2014, and the benefit of the East-West link in 2013 and 2014.
The total generating capacity in Ireland has risen from 6500 MW in 2006 to 9500 MW in 2014 ), while the consumption of electricity has not risen at all during this period. The generating capacity stands now at more than three times the average consumption level, largely due to the build out of wind power and the interconnector to England. This is proof, that wind does not replace dispatchable power.
The effect of this investment is nullified by a slight increase in the use of coal in the fossil fuel mix, as gas became expensive in the recent years. The performance of the system in 2011 shows clearly, that without the extra windmills and without the E-W link, but with the new CCGT gas units operating one does better than with all the new wind turbines and E-W link.
The 600 million euro costing E-W link allows Eirgrid to export the problems of incorporating wind energy to the UK. The Eirgrid data shows, that during the last two years the E-W link served as peak shaver during the day and as a sink for unwanted wind production during the night. A report of SEAI states, that dispatch down of wind energy (curtailment) in 2013 was halved due to the presence of the link.
This bad performance of the “greening” of the electricity supply comes on top of the enormous social and economical cost of littering of the landscape with thousands of 160 meters high rotating monsters.
Appendix 1
Calculation of CO2 intensity from fuel input data.
The table shows the transformation of the fuel input mix to CO2 emission/kWh for 2006.The amounts of input fuel and output electricity are expressed in Kilo Ton Oil Equivalent, a common energy unit to account for the calorific values of the different fuels. One ktoe = 11,6 GWh
The energy input for fossil generated electricity is the sum of all 4 components: 4833 ktoe.
The 2nd column represents the fractional contribution of each fuel.
-The 3rd column shows the amount of CO2 emitted by combustion of 1 kWh of each element of the fuel mix. In other words the 3rd column says: A generator running on coal with 100% efficiency emits 340 g/kWh.
The last column shows the contribution of each fuel component to the CO2 emission of the input mix. This is obtained by multiplying column 2 by column 3.
Column 4 shows, that the fuel mix as used in 2006 produces 269 gram CO2 for the burning of 1 kWh fuel. In this way we change from ktoe to kWh without introducing electricity generation efficiencies for each fuel separately.
The efficacy of the transformation from fuel to electricity follows from the ratio between the total amount of fuel burned and the total electricity produced.
The European rule is to attribute CO2 emissions of exported power to the country of origin, so official numbers for the CO2 emissions include no contribution for imports.
The recently completed East-West connection is used to import up to 500 MW during the day, while during windy nights superfluous wind energy is exported.
Here we study the performance of the Irish system, so imports are subtracted from (and exports are added to) the “total demand” or “energy delivered” to determine the indigenous production.
The total electricity generation in 2006 is given as 2225 ktoe in the SAEI report 2006.
Subtract net imports 153 ktoe
Subtract the contribution from renewables 231 ktoe
Net fossil electricity production 1841 ktoe
The energy conversion efficiency is now output/input: 1841/4833 = 38,1% for the generation from fossil fuels.
Table 1 shows, that 1 kWh fuel mix produces 269 gCO2/kWh, so the production of 1 kWh electricity from fossil fuels emits: 269/0,381 = 706 gCO2/kWh, a surprisingly high number.
Adding the contribution of renewables (231 ktoe) to the 1841 ktoe electricity production increases the indigenous production to 2072 ktoe and increases the apparent energy conversion efficiency to 42,9%. This lowers the apparent CO2 intensity to 628 gCO2/kWh.
Appendix 2:
The CO2 data of Eirgrid.
On the website of Eigrid one finds the following text:
EirGrid, with the support of the Sustainable Energy Authority of Ireland, has together developed the following methodology for calculating CO2 emissions. The rate of carbon emissions is calculated in real time by using the generators MW output, the individual heat rate curves for each power station and the calorific values for each type of fuel used. The heat rate curves are used to determine the efficiency at which a generator burns fuel at any given time. The fuel calorific values are then used to calculate the rate of carbon emissions for the fuel being burned by the generator.
It is clear from this definition, that the model does not include the dynamic behaviour of the system and neglects the spinning reserve. This casts doubt on the level of emissions calculated from the model. The data is presented on the Eirgrid website at intervals of 15 minutes, so one has 2900 lines each month and 35000 lines of data for each year. Publication started in 2010, therefore 2010 contains only two months of data.
Table 1 presents the integral over a year of the Eirgrid data.
The total demand entries are remarkably stable. They are corrected for imports & exports in row 3.
References.
1. www.eirgrid.com
The section “Operations” contains 15 minutes data about the total demand, the CO2 emission and the wind production. The data for the East-West connection are available under the head “East West”. The graphs for wind energy and for the imports from the EW link show a nice anti correlation on Jan 15 2015. Ireland exports the variations in wind energy to the UK….
Note Jan 2016: The 15’ data has been removed from the Eirgrid website
2. F. Udo, Wind energy in the Irish power system.
http://www.clepair.net/IerlandUdo.html
3. J.B. Wheatley: Quantifying CO2 savings from windpower;
Energy Policy, 2013, vol. 63, issue C, pages 89-96.
. This effect has been recognised now by the Sustainable Energy Authority in its year report “Quantifying Ireland’s Fuel and CO2 Emissions Savings from Renewable Electricity in 2012”
Quote page 2:
“Individual fossil-fuel generators run in less efficient modes with renewable electricity generation on the system, showing a 7% increase in the CO2 emissions intensity for such generators.”
In 2012 the part of wind energy on the total was 16%. Neglecting the 4% hydro this implies, that 84% of the total generation was fossil driven and had an efficiency loss of 7%.
This 84% used 1,07 * 0,84 = 0,90 of the fuel it would use without wind, hence the fuel gain is 10% for 16% wind. The efficacy of wind is 10/16 = 63%.
Note:
This is data from an official report. This is a rare example of honesty from a government agency about the
extra fuel necessary to incorporate wind energy into an existing grid..
4. The annual reports for the period 2006 to 2013 are listed under the title “Energy in Ireland”
www.seai.ie/Publications/Statistics_Publications/
5. “More on the Energy Bubble”
http://irishenergyblog.blogspot.nl/2015/01/energy-bub.html
6. The calculation is presented in ref 13 of:
C. le Pair en K. de Groot: De invloed van elektriciteit uit wind op het fossiel brandstofgebruik. (in Dutch)
7. F. Udo: Building wind turbines costs more energy than you think.
http://www.clepair.net/Udo201303payback.html