INTERNSHIP REPORT
EVALUATION OF THE OPERATIONAL PERFORMANCE OF A PARABOLIC TROUGH
CSP PLANT
E U R O P E A N J O I N T M A S T E R S I N M A N A G E M E N T A N D E N G I N E E R I N G O F E N V I R O N M E N T A N D E N E R G Y
A U T H O R :
C A R L O S G U I L L E R M O R U I Z S Á N C H E Z
T U T O R S :
P E D R O M I L L A N E S M O R E N O J A V I E R M U Ñ O Z - A N T Ó N
S E P T E M B E R 2 0 1 3
2
I N T E R N S H I P R E P O R T
EVALUATION OF THE OPERATIONAL P ERFORMANCE OF A PARABOLIC TROUGH CSP PLANT
INDEX
LIST OF FIGURES ... 3
LIST OF TABLES ... 4
ACKNOWLEDGEMENTS ... 5
1 ABSTRACT ... 6
2 COMPANY PROFILE: ELECNOR ... 7
3 SCOPE ... 8
4 BACKGROUND ... 8
4.1 CSP & Parabolic Trough Technology ... 8
4.2 CSP in Spain ... 10
4.3 The Spanish Tariff Deficit ... 10
5 ANALYSIS OF REGULATORY CHANGES IN SPAIN ... 11
5.1 Beginnings of Renewable Energy Regulation ... 11
5.2 Regulatory Consequences of the Tariff Deficit ... 12
6 ELECNOR’S CSP PLANTS ... 15
6.1 Plant Location ... 15
6.2 Plant Configuration ... 15
7 OPERATIONAL PERFORMANCE EVALUATION ... 16
7.1 Semiannual/Annual Evaluation ... 16
7.2 Daily Evaluation ... 21
8 OPERATIONAL IMPROVEMENTS ... 24
9 CRITICAL COMPONENTS ... 26
10 CONCLUSIONS & OUTLOOK ... 30
11 BIBLIOGRAPHY ... 32
3 LIST OF FIGURES
Figure 1. Proportions of the World's energy resources and demand ... 8
Figure 2. Concentration of sunlight using a) parabolic trough collectors, b) linear Fresnel collectors, c) dish/engine systems, and d) central receiver systems ... 9
Figure 3. Schematic of a PT-CSP Plant ... 9
Figure 4. Evolution of the CSP installed capacity (MW) in Spain. ... 10
Figure 5. Average Total Revenues 2012-2013 ... 13
Figure 6. Average Regulated Revenues 2012-2013 ... 14
Figure 7. Geographical Location of the ASTE 1A and ASTE 1B Power Plants ... 15
Figure 8. Simplified Schematic of the ASTE 1A and ASTE 1B Power Plants ... 16
Figure 9. Monthly Power Generation ... 17
Figure 10. Cumulative Percentage of Compliance in Terms of Power Generation ... 18
Figure 11. Cumulative Natural Gas Consumption of the Power Plant ... 19
Figure 12. Water Consumption Ratio of the Power Plants ... 20
Figure 13. Cumulative Electricity Consumption of the Power Plants ... 20
Figure 14. Typical generation curves of the power plants in a sunny day with clear sky ... 21
Figure 15. Daily Evaluation - Differences in generation due to a cold start-up sequence (left) and a turbine trip caused by a component failure (right)... 22
Figure 16. Daily Evaluation - Difference in maximum power generation capacity of the power plants ... 23
Figure 17. Daily Evaluation - Difference in response to cloudy periods caused by different control schemes ... 23
Figure 18. Daily Evaluation - Difference in response to cloudy periods caused by a reflectivity loss. ... 24
Figure 19. Change in power output differences between the two plants ... 24
Figure 20. Improvements in stability and power output in ASTE 1A ... 25
Figure 21. Improvements in stability and power output in ASTE 1B ... 25
Figure 22. Corrective actions by type of intervention ... 27
Figure 23. Corrective actions by plant sub-system ... 27
Figure 24. Corrective actions by plant component ... 28
Figure 25. Energy loss caused by a failure in a primary HTF pump ... 29
4 LIST OF TABLES
Table 1. Overview of Spanish Policy Concerning CSP (1997-2007) ... 11
Table 2. Overview of Regulatory Changes Concerning CSP (2008-2013) ... 12
Table 3. Basic Configuration of the ASTE 1A and ASTE 1B Power Plants ... 15
Table 4. Criteria for the Semiannual Evaluation of the Operation of the Power Plants ... 16
5 ACKNOWLEDGEMENTS
First I would like to thank my mother for making it possible for me to start and fulfill this two-year-long project called ME3. Your help was invaluable.
I would also like to thank my company tutor Pedro Millanes Moreno for providing me with all the materials and information that I needed for the making of this work, and more importantly for making me feel like a true part of the Elecnor team.
Finally I would like to thank Prof. Javier Muñoz Antón for being a great academic tutor, and providing excellent advice and guiding regarding this work during the whole semester.
6 1 ABSTRACT
Renewable energy technologies are in the spotlight. On the one hand, they are called to have a leading role in the battle against climate change. On the other hand, with constantly growing installed capacity and more than generous Feed-in-Tariffs (FiT’s), they might represent the financial demise of the energy sector of some countries. This is the case for concentrated solar power (CSP) in Spain. It is one of the country’s trademark industries and it is also a financial sink, with Feed-in Tariffs that were more than six times higher than the regular market price, before an energy reform had to be introduced.
This work is divided in three main parts. The first part consists of an assessment of the regulatory changes and their impact on the Spanish CSP industry, which by July 2013, amount to at least 17% in revenue losses for the industry without taking into account variable losses related to the use of backup fossil fuels. These changes have created an atmosphere of great uncertainty for the industry, since the profitability of their projects has been severely challenged.
The second part of this report consists of an evaluation of the operational performance of two parabolic trough CSP plants during their commissioning period. Two different methods were used; the first method consists of an annual or semi-annual evaluation where the main intakes and outputs (gas, water and electricity consumption, as well as, power generation) are compared with fixed guaranteed values. The second method consists of a daily evaluation of the power generation curves of the plants.
The first evaluation showed positive results, with both power plants operating within the guaranteed values. The second evaluation fulfilled its objective by highlighting opportunities for improvement for both plants. Problems were brought up by the evaluation and subsequently identified and solved by the operation and maintenance (O&M) crew, increasing the net output of the plants by more than 1.5 MW.
Last but not least, some critical components were successfully identified by means of reviewing the corrective actions performed during the year, opening an opportunity to reduce downtime by having proper spare parts stocks, and addressing design issues with equipment suppliers.
The results presented by this report are not specific to these two power plants. The regulatory changes and the reduction of the FiT’s present a difficult challenge for the whole Spanish renewable energy (RE) industry and a real threat to the future development of these technologies. In the case of CSP, there is still work to do, especially by equipment suppliers, since the most relevant issues found were related to a lack of reliability of the components. These financial and technological problems should be solved jointly by the government, the producers, and the equipment suppliers in order to bring the CSP industry closer to its financial and technological maturity.
7 2 COMPANY PROFILE: ELECNOR
Elecnor was founded in 1958 in Bilbao, Spain as a company originally dedicated to the electric power industry (power lines, substations, lighting, and installations). Through its more than 50 years of continuous growth and diversification, it is now present in several additional business areas which include power generation, railroad construction, telecommunications infrastructure, waste management and water treatment infrastructure, control applications, and even aerospace engineering.
Elecnor’s business figures have been improving throughout the years despite the severe economic crisis that Spain is facing. The company has made the best of its possibilities by basing its strategy on prudent management, and a continuous geographic and business diversification process, especially in markets with a high potential, such as Latin America, Africa, and India.
Presently, Elecnor is one of the leading Spanish companies in engineering, infrastructure construction and development, renewable energy, and new technologies. It is present in more than 50 countries, and employs over 12,000 workers through its four business units: Infrastructures, Renewables, Concessions, and Elecnor Deimos (Elecnor, 2013).
In recent years, the company has established itself in the Spanish renewable energy industry as an important turnkey contractor. Some of its projects include the development and construction of multiple hydro power plants around the world, over 1,000 MW of wind power, and more than 200 MW of photovoltaics (PV).
Elecnor’s mission is to contribute to the economic and technological development, social welfare, and sustainable development of the markets in which it is present. Aligned with this statement, in 2009, with an investment of over 900 million euros, it started the construction of three 50 MW parabolic trough concentrating solar power (PT-CSP) plants in Spain (Elecnor, 2013). These power plants will not only positively impact the environment by producing clean energy, but will also improve the local economy by creating employment opportunities.
The ASTE-1A, ASTE-1B, and ASTEXOL-2 PT-CSP plants started operations in late
2012, marking the beginning of a new business venture for Elecnor, and opening the
door for new business opportunities worldwide. In the midst of an economic crisis, they
also prove to be a challenge, and their success and profitability drift in an atmosphere of
uncertainty due to the continuously changing regulations in Spain.
8 3 SCOPE
This work focuses on the commissioning period of two PT-CSP plants and it is set in a time of crisis for the Spanish CSP industry. The work is composed of an evaluation of the main inputs and outputs of the plants’ performance with respect to fixed operational guarantees.
Since the Spanish government introduced a series of regulatory changes in 2012 concerning CSP, an analysis of these changes and their effects on energy production was made based on a literary review and the actual performance of the plants. The inputs to be assessed include auxiliary gas, water, and electricity consumption; the main output is power generation.
Additionally, critical components will be identified aiming to reduce the plants’ downtime.
4 BACKGROUND
4.1 CSP & PARABOLIC TROUGH TECHNOLOGY
The global energy landscape is constantly changing. Issues like global warming, limited fossil fuel reserves and their increasing prices make the world slowly turn towards renewable energy sources (RES). The sun is a virtually endless source of energy and, as Figure 1 shows, if this energy were to be harnessed properly it could easily cover the world’s energy demand.
Figure 1. Proportions of the World's energy resources and demand. (Institute of Solar Research, 2011)
One way of converting the power of sunlight into electricity is through concentrated solar power (CSP) technologies. Presently, there are four CSP plant technologies: Parabolic trough
9 collectors, linear Fresnel systems, central receiver or solar tower systems, and dish/engine systems. Figure 2 shows the basic working principle of each technology.
Figure 2. Concentration of sunlight using a) parabolic trough collectors, b) linear Fresnel collectors, c) dish/engine systems, and d) central receiver systems. (http://www.volker-quaschning.de, 2003)
Conventional parabolic trough CSP (PT-CSP) plants consist of 2 main systems, the solar field and the power conversion system (PCS), although a thermal energy storage (TES) system is added frequently. In the solar field, a parabolic trough shaped mirror reflects direct radiation onto an absorber tube or heat collecting element (HCE) through which a heat transfer fluid (HTF) flows. The hot HTF then goes through a series of heat exchangers where the industrial process steam is formed. In the PCS, the thermal energy in the steam is transformed into electricity as in any other Rankin cycle. Figure 3 depicts a basic schematic of the configuration of a PT-CSP plant without storage.
Figure 3. Schematic of a PT-CSP Plant. (http://www.volker-quaschning.de, 2003)
a b
c d
10 4.2 CSP IN SPAIN
The Spanish CSP industry is relatively young and characterized by very rapid growth. The first commercial PT-CSP plant in Spain started operations in 2008 (Ciria Repáraz, 2010). With a rating of 50 MW, it was bound to be part of the largest solar complex in Europe at that time, and the first large plant to incorporate thermal energy storage (TES) with molten salts (Solar Millenium AG, 2008). It was followed by several other projects that were made possible by a lavish feed-in tariff (FiT) system. This FiT system nourished the CSP industry and allowed it to grow in an unprecedented way, creating the “CSP bubble”. As Figure 4 shows, the installed capacity of CSP in Spain grew more than 600% in three years, making Spain the absolute leader in the sector.
Figure 4. Evolution of the CSP installed capacity (MW) in Spain. (REE, 2013)
By the end of 2012, Spain is one of the word leaders not only in CSP, but in the whole RE industry in terms of installed capacity. It ranks first in CSP technology with 1,878 MW, second in wind power with 22,213 MW, and third in solar PV with 4,400 MW (Couture & Mischa, 2013).
In April 2013, 54% of the demand was met with RE, and 73.9% of the power production came from non CO2 emitting technologies (REE, 2013), a new record for Spain.
4.3 THE SPANISH TARIFF DEFICIT
Not all that glitters is gold. The fairly generous FiT system, combined with the explosive growth of RE technologies, and a poorly designed policy that subsidizes consumer rates keeping them artificially low, created what is now called the “tariff deficit”. The costs of generating and distributing electricity have exceeded the amount that the utilities can recover from the consumer rates. The problem started around year 2000 and the debt has been growing since. By the end of 2012, the deficit amounted to 25.5 billion euros (CNE, 2013).
282
682
1049
1878
2009 2010 2011 2012
11 In a context of a Spanish economic crisis, and under the pressure of the European Union (EU), Spain was bound to take urgent actions to address the deficit. The Spanish government introduced a series of regulatory changes in order to reduce this deficit. The changes, which will be addressed further on, translate in a revenue loss for the Spanish CSP industry that was not considered in the original business plans. This creates an atmosphere of financial uncertainty, and it forces the CSP plants to make changes in the daily operation to avoid the extra costs.
5 ANALYSIS OF REGULATORY CHANGES IN SPAIN
5.1 BEGINNINGS OF RENEWABLE ENERGY REGULATION
Spain started supporting RES as early as 1980 when Law 82/80 on energy conservation was passed. Since then, the regulations have been evolving, establishing new measures and financial instruments like feed-in tariffs designed to favor the deployment of RE, starting by the Spanish Electric Power Act 54/1997. A brief overview of this policy changes can be seen in Table 1.
Table 1. Overview of Spanish Policy Concerning CSP (1997-2007)
Electric Power Act 54/1997 - Creation of a “Special Scheme” for energy production from RE sources
- Facilities cannot exceed 50 MW
- Incorporation of produced energy into the grid - Payment of a premium to improve the market price - At least 12% of the total demand is to be met by RE by
2010
Royal Decree 2818/1998 - Requirements for registration in the “Special Scheme”
- Conditions of energy delivery - Applicable economic scheme
Royal Decree 436/2004 - Establishment of a methodology to update the legal and economic framework
- Two payment options: Sale to distributor at fixed tariff or sale to market plus premium
- Incentives are calculated as a percentage of the annual average tariff
Royal Decree 661/2007 - Abolishment of linkage of FITs to electricity prices - Change in tariff system introducing cap and floor prices - Changes concerning biomass tariffs
- CSP plants can use backup fuels. The equivalent to 12- 15% of the total yearly production.
12 5.2 REGULATORY CONSEQUENCES OF THE TARIFF DEFICIT
As a result of the aforementioned tariff deficit in the Spanish energy sector, a great number of changes have been introduced to the regulation concerning renewables and especially CSP technologies. Table 2 shows a summary of the main changes that affect the CSP producers.
Table 2. Overview of Regulatory Changes Concerning CSP (2008-2013)
Royal Decree 1614/2010 - Introduction of caps to working hours for CSP and wind projects.
Royal Decree Law 1/2012 - Moratorium on RE projects. Suspension of FiT registration procedures for new plants.
Royal Decree Law 15/2012 - Additional 7% tax on electricity sales, regardless of the source.
- Withdrawal of premiums for the fraction of energy produced with fossil fuels (CSP).
Royal Decree Law 29/2012 - Suspension of FiT for project that don’t meet the construction deadlines.
Royal Decree Law 2/2013 - Elimination of the “pool plus premium” option for electricity sale to the grid.
- Indexation mechanism (IPC) is decoupled from food and fuel prices.
Royal Decree Law 9/2013 - Abolishment of FiT program.
- Specific plants qualify for a special payment that allows a fixed maximum IRR set by the government.
These changes, aimed to reduce the tariff deficit, represent a big blow for the CSP industry in Spain. The moratorium on RET means that there will be a decrease, if not a complete cease, of the investments in RE. The revenues in new projects will be low, since they will not get any incentives, therefore making the investments unattractive. In 2013, only 100 MW of CSP have been installed, and they are remnants of the industry boom (CNE, 2013).
The use of natural gas is a common part of the operation of most CSP plants. Plants usually have an auxiliary gas system which helps with HTF heating. In the winter it prevents the HTF from freezing, during normal operation it makes up for short cloudy periods, and it ensures having enough steam in the turbine to maintain it pressure sealed when it is not synchronized.
With this said, it is clear that the revenue loss is greater if we take into account the withdrawal of the incentives for the proportion of energy produced with fossil fuels.
The 7% additional tax, the elimination of the “pool price plus premium” sale option, and the change in indexation all have a negative impact for the industry, and also represent a considerable decrease in the producer’s revenues. Figure 5 shows the average revenues in €/MWh for the CSP
13 industry in 2012 and part of 2013. It is clear that the removal of the “pool price plus premium”
option has a great impact on the plants, accounting for an estimated revenue loss of over 10%.
The total revenue reduction by the month of June 2013 grows to just under 17% if we take the additional 7% tax into account (CNE, 2013).
Figure 5. Average Total Revenues 2012-2013. (CNE, 2013)
The exact amount of this loss is unknown since the mechanism to assess the use of gas has not been defined yet by the government, but some sources calculate total losses amounting to 30- 40% of the revenues (Asociación PROTERMOSOLAR, 2013).
The problem for the CSP industry does not end there. Indeed, the producers saw their revenues reduced by just under 17%, which should have contributed to the reduction of the deficit. However, to the disappointment of both producers and the government, the deficit reduction was not the expected. The first quarter of 2013 was colder and rainier than predicted;
this altered the energy mix and made the prices for electricity drop below the government’s expectations (Jefatura del Estado, 2013). Lower market prices mean that the government has to compensate for those low prices in order to pay the full fixed tariff. Figure 6 shows the average regulated revenues for CSP. The regulated revenues (total revenues minus pool price) paid in 2013 for the fixed tariff were actually higher than the ones paid in 2012 for the same option.
27,00 28,00 29,00 30,00 31,00 32,00 33,00 34,00 35,00
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec
€/MWh
Pool+Premium 2012 Regulated Tariff 2012 Regulated Tariff 2013
14 Figure 6. Average Regulated Revenues 2012-2013. (CNE, 2013)
In July 2013 the Spanish government decided to abolish the FiT system to keep reducing the deficit, and a new special scheme is introduced. The new regulation states that energy will be sold at the market price, with no premiums. For power plants that cannot meet their financial expenses with the sale of energy, a special compensation will be granted which will allow them to achieve a “reasonable profitability” of approximately 7.5% before taxes (this value will be revised after 6 years) (Jefatura del Estado, 2013). The problem, according to the CSP association PROTERMOSOLAR, is that CSP plants are not low risk projects, on the contrary, they are high risk projects and the should be compensated fairly or a collapse of the industry is possible.
According to some experts, this IRR after taxes is closer to 5% (Morales, 2013). The compensation takes only 3 components into account: the income from the sale of electricity at market price, the operating costs, and the initial investment. These three aspects will be taken in consideration for a “typical” plant and the compensation will be decided. The financing costs have not been taken into account which could make some highly leveraged projects bankrupt.
Since the government hasn’t published what it considers the “typical” plant or the specific methodology for calculating any of this numbers, there is an atmosphere of uncertainty. What is certain is that solar thermal producers will lose a part of their revenues and their financial results will not be as they were expected in their business plans. With some plants being at risk of defaulting their debts, the problem now includes the banks, which must renegotiate the financing terms.
22,00 23,00 24,00 25,00 26,00 27,00 28,00 29,00
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec
€/MWh
Pool + Premium 2012 Regulated Tariff 2012 Regulated Tariff 2013
15 6 ELECNOR’S CSP PLANTS
With the purpose of performing a better analysis of the performance of the plants, this work focuses only on two of the three plants owned by Elecnor.
6.1 PLANT LOCATION
The ASTE 1 CSP plants are located near Ciudad Real, Spain, which is a region that hosts other CSP plants and provides a fairly good amount of Direct Normal Radiation (DNI) as Figure 7 shows.
Figure 7. Geographical Location of the ASTE 1A and ASTE 1B Power Plants. (SolarGIS, 2013)
6.2 PLANT CONFIGURATION
The two power plants, ASTE 1A and ASTE 1B, have minor differences due to the geometry of the property, but have the same main components, which provides an ideal situation to make comparisons. The plant configuration is shown in Table 3 and a general process diagram is shown in Figure 8:
Table 3. Basic Configuration of the ASTE 1A and ASTE 1B Power Plants.
Solar Field
Solar Field Area 392,400 m2
HTF Dowtherm A
Solar Field Inlet Temperature 293 Solar Field Outlet Temperature 393
Thermal Storage No
Ullage System Yes
16 Power Block
Output Capacity 49.9 MW
Cooling Method Wet cooling (cooling towers)
Fossil Backup HTF Natural Gas boiler
Figure 8. Simplified Schematic of the ASTE 1A and ASTE 1B Power Plants. (Flowserve, 2013)
7 OPERATIONAL PERFORMANCE EVALUATION
The operational performance of the power plants takes into account both the power generation and the intakes required for the normal operation of the facility. Four key elements are considered: Energy exported to the grid; and natural gas, water, and electricity consumption.
7.1 SEMIANNUAL/ANNUAL EVALUATION
The plants must operate in a way that allows them to meet guaranteed values fixed in an O&M contract. If the values are not met an economic compensation must be paid. The method for this evaluation is simple. The guaranteed values are shown in Table 4:
Table 4. Criteria for the Semiannual Evaluation of the Operation of the Power Plants.
Energy Production ≥80% of “Theoretical Production” – 1st year
≥90% of “Theoretical Production” – 2nd year
≥100% of “Theoretical Production” – 3rd year Natural Gas Consumption ≤ 15% of yearly energy production
Water Consumption m3/MWh Ratio ≤ 6.24
Electricity Consumption ≤8827 MWh/year
17 In order to calculate the “theoretical production” of energy, the Solar Advisor Model 3.0 (SAM) software created by the National Renewable Energy Laboratory (NREL) is used. The SAM software makes performance predictions for grid-connected energy projects based on input data such as plant design parameters and weather readings (NREL, 2013). Each power plant has three weather stations operating all year. The meteorological data gathered by these weather stations is averaged and introduced to the SAM software, as well as the characteristic parameters of the plants demanded by the software. The real production must amount to the percentage of the “theoretical production” stated in the contract.
The operative results are taken from monthly O&M reports provided by both plants and captured in a single excel file in order to compare the plants’ results and to evaluate if they are meeting the guaranteed values.
7.1.1 ENERGY PRODUCTION
Figure 9 shows the monthly generation of both power plants. The “theoretical production”
values of both plants are similar, and they should be, since the plants are next to each other.
There is a considerable difference in November. However, this is because the power plants started operations at a different date. ASTE 1A started operations on November 1st, 2012, while ASTE 1B started two weeks later. The slight differences in February, March, and April were caused by errors in the weather stations, but the shown values were accepted as valid. The faults were fixed and by the month of May there were practically no difference in the weather readings.
The green line represents the expected energy production according to the typical meteorological year (TMY), or in other words, what the business plan was based on. The results during the first quarter of 2013 were below the TMY expectations; the months of March and April were abnormally cold and humid.
Figure 9. Monthly Power Generation.
0 2.000 4.000 6.000 8.000 10.000 12.000 14.000 16.000 18.000
0 2.000 4.000 6.000 8.000 10.000 12.000 14.000 16.000 18.000
Nov* Dec* Jan* Feb Mar Apr May Jun Jul
MWh
Real P 1A Real P 1B Theoretical P 1A Theoretical P 1B TMY
18 The monthly generation values shown above translate into another chart shown in Figure 10 where the cumulative percentage of compliance is represented. This is the most important value to take into account for the evaluation of the power plants. It shows how much of the energy that the plants were supposed to produce during the year, was actually produced.
Figure 10. Cumulative Percentage of Compliance in Terms of Power Generation (Preal/Ptheory).
ASTE 1B seems to have had an outstanding performance on the first months; however the graph is deceiving if it is not analyzed in parallel with Figure 9. The extremely high values are a result of the use of natural gas and the relatively small amount of energy generated in the winter months. In other words, the energy generated during the winter and the performance during this months, only represents a minor fraction of the yearly values. ASTE 1B burned more gas than ASTE 1A, however the use of natural gas will be addressed further on.
It is safe to say that both plants have been operating according to the expectations. ASTE 1A has produced 94% of the energy it was supposed to produce, according to the SAM software, while ASTE 1B has a percentage of 99%. Both values are well above the minimum acceptable value of 80% in the first year.
7.1.2 RESOURCE USAGE
As it was mentioned before, the intakes of the power plants are also taken into consideration;
the use of gas, water and electricity are assessed, and should be below the values guaranteed by the contract.
7.1.2.1 Natural Gas Consumption
Figure 11 shows the cumulative natural gas consumption as a percentage. The energy provided by the used gas is divided by the total generated energy to obtain this percentage. The graph shows a big decrease in gas consumption, which is brought by two different reasons.
60%
80%
100%
120%
140%
160%
Nov* Dec* Jan* Feb Mar Apr May Jun Jul Aug Sep Oct
Preal/Ptheory 1A Preal/Ptheory 1B
19 Firstly, as the weather gets warmer and less cloudy during the summer, less gas is required for heating the cold HTF, producing gland steam, and making up for small cloudy periods, and more energy is generated in a purely solar mode. Secondly, after the regulatory changes introduced in February 2012, regarding the withdrawal of premiums for the fraction of energy produced with fossil fuels, the company gave the order to limit the gas consumption to the minimum, meaning that no power would be generated with fossil fuels and the use of gas is limited to the heating (anti-freezing) purposes. This represents a major change in the operation strategy of the plant.
Figure 11. Cumulative Natural Gas Consumption of the Power Plants (Gas used / Energy Produced)
Both plants are within the maximum 15% of gas usage stipulated in the contract. The percentage of gas consumption for ASTE 1A and ASTE 1B are 5.4% and 10.1% respectively. It is important to say that this parameter is likely to be modified according to the pending regulatory changes. The new guaranteed value has not been fixed, but changes include the possibility of generating up to 12% of the energy with natural gas plus an additional fixed amount consisting of the volume of gas needed to keep the plant operational (anti-freeze and gland steam purposes).
7.1.2.2 Water Consumption
Water has three main uses in a CSP plant. Around 90% of the water is used for cooling purposes, the rest is divided between the boiler feed water and other uses (e.g. washing mirrors, and service water). Water is withdrawn from the ground and part of it is returned via infiltration basins. However, water is lost in the process, usually through evaporation.
0%
20%
40%
60%
80%
100%
Nov* Dec* Jan* Feb Mar Apr May Jun Jul Aug Sep Oct
Gas Consumption 1A Gas Consumption 1B
20 Figure 12. Water Consumption Ratio of the Power Plants (m3/MWh)
The water consumption is illustrated in Figure 12. The plotted variable is the cumulative ratio between the volume of water consumed and the generated energy, expressed in cubic meters per megawatt hour. The water consumption of both plants is currently complying with the values specified in the contract, 5.6 and 5 m3/MWh, and it has a decreasing trend, which was expected as the plants’ processes are continuously being optimized by the operators.
7.1.2.3 Electricity Consumption
Electricity is consumed by different activities that include turbine starts and stops, electric heat tracing of piping and other components, offline cooling, water treatment, daily office activities, and lighting. Figure 13 illustrates the electricity consumed by the power plant throughout the year.
Figure 13. Cumulative Electricity Consumption of the Power Plants (MWh)
Both of the plants’ electricity use is complying with the guaranteed value. According to the trend shown by the collected data, ASTE 1A will end the year consuming just under 8 GWh and ASTE 1B will consume under 6 GWh in the year. The difference in electricity consumption is
4 5 6 7 8 9 10 11 12
Nov* Dec* Jan* Feb Mar Apr May Jun Jul Aug Sep Oct
m3/MWh
Water Consumption 1A Water Consumption 1B
8827 8827
0 2000 4000 6000 8000 10000
Nov* Dec* Jan* Feb Mar Apr May Jun Jul Aug Sep Oct
MWh
Energy Consumption 1A Energy Consumption 1B
21 obvious; the reason for it is a different offline operation strategy regarding the condenser, which will be detailed in the next chapter.
7.2 DAILY EVALUATION
The daily evaluation only contemplates power generation. It is a new implementation that is not required by the O&M contract. The objective of this evaluation is to identify possible operational improvements by comparing the daily performance of the two power plants, as well as keeping the top layer management informed about the plants’ daily operation. It is an evaluation of a more qualitative than quantitative nature.
In order to perform this evaluation, the instantaneous power generated by both plants is plotted simultaneously in a single graph. Three time periods are defined: the start-up period, the central hours, and the cool-down period. The difference between the instantaneous power of the two plants is then calculated and classified under their corresponding time period. Figure 14 shows an example of the aforementioned graph with the typical daily performance for a sunny day with clear skies.
Figure 14. Typical generation curves of the power plants in a sunny day with clear sky.
Certain details regarding the configuration and normal operating conditions of the power plants need to be taken into consideration in order to perform this assessment. One of the key aspects is that ASTE 1B has a larger HTF volume than ASTE 1A, which means that under normal circumstances, ASTE 1B will take longer to reach an optimal solar field temperature and therefore start generating power slightly later than ASTE 1A. At the same time, it is able to generate slightly more energy during the cool-down period.
-20 -10 0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
Difference ASTE 1A ASTE 1B
Start-up
Central Hours
Cool-down
22 This kind of evaluation allows not only the identification of differences during normal operation, but it also allows abnormalities and operational errors to be noticed, and the losses caused by these events to be quantified. The differences observed may be grouped in two different categories: differences attributable to mistakes in the operation or equipment faults, and differences attributable to operational strategies and plant configuration (e.g. solar field control system and parasitics).
Figure 15 shows some of the differences that are caused by mistakes in the operation and component failures.
Figure 15. Daily Evaluation - Differences in generation due to a cold start-up sequence (left) and a turbine trip caused by a component failure (right).
The graph on the left shows a slow start-up from one of the plants, which translated into a loss of about 15 MWh. Start-ups can be optimized by plant operators and the slower start-ups can be minimized by reaching optimum field temperature before synchronizing. The graph on the right shows two turbine trips that implied a loss of 260 MWh in just one day. Turbine trips are designed to protect the turbine from poor working conditions. They can be caused by different reasons such as human errors, mechanical failure of components, or wrong readings by the instrumentation. In this case, the daily evaluation allows us to identify avoidable generation losses.
Differences caused by different operational strategies and control systems, may also be noticed in these curves. The operation curves for a summer day are shown in Figure 16. In this graph ASTE 1A is able to reach slightly higher instantaneous power values than ASTE 1B. This sort of behavior could be caused by different reasons. In this case, the reason was that the plants had a different solar field temperature, cause by a different solar field arrangement. This issue will be addressed on the next chapter.
-20 -10 0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
Diff. ASTE 1A ASTE 1B
-20 -10 0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
Diff. ASTE 1A ASTE 1B
23 Figure 16. Daily Evaluation - Difference in maximum power generation capacity of the power plants.
Another issue that can be seen on Figure 16 is that ASTE 1A’s signal seems to have more ripple than ASTE 1B’s. In this case this difference is caused by a different control scheme, ASTE 1B’s control was better for temperature stability. The difference caused by the control schemes seems irrelevant. However, if we look for a day with uneven radiation, like the one shown in Figure 17, the real difference can be better appreciated.
It is important to keep in mind that this kind of evaluation is not useful for identifying the actual cause of the problems, since the curves only show the difference in power generation. Its objective is just to identify that there is a problem, and prompt the O&M crew to find the real cause. Two different problems may reflect in the curves in a similar way. For example, Figure 18 also shows a different response to changes in radiation, but in this case it is caused by the lack of mirror washing in ASTE 1A which translates to a reduced reflectivity of the mirrors.
Figure 17. Daily Evaluation - Difference in response to cloudy periods caused by different control schemes.
-20 -10 0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
Diferencia ASTE 1A ASTE 1B
-20 -10 0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
Diff. ASTE 1A ASTE 1B
Diff.
24 Figure 18. Daily Evaluation – Difference in response to cloudy periods caused by a reflectivity loss.
8 OPERATIONAL IMPROVEMENTS
From the last chapter it can be seen that the daily evaluation can point out operative discrepancies that make room for improvements. In this chapter, the main improvements brought by this comparison of the two plants’ operation will be briefly addressed.
A good indicator that the daily evaluation has been effective is the reduction of the differences in the power delivered to the grid by the plants, shown in Figure 19. During the start- up period the differences have been reduced by 45%, during the cool-down period the differences have been reduced by approximately 40%, and during the central hours the differences have been almost eliminated by reducing them by more than 80%.
Figure 19. Change in power output differences between the two plants -20
-10 0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
Diff. ASTE 1A ASTE 1B
0 20 40 60 80 100 120 140 160
Start-up Central Hours Cool-Down
MWh
April July
25 The power generation was increased mainly by means of reducing the parasitic energy consumption of the plants. Less power consumption by the components means less parasitics and more power output.
The overall improvement can be easily appreciated in Figure 20 and Figure 21, where the power generation in the months of April and July are compared for both plants. The timespan of the production should not be taken into account since days are longer in July than they are in April. The total increase in power output is between 1.5-1.9 MW, and operation is more stable.
The improvement might seem small, but if we consider the number of hours the plants operate it becomes a considerable amount of revenue for the plants.
Figure 20. Improvements in stability and power output in ASTE 1A.
Figure 21. Improvements in stability and power output in ASTE 1B.
0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
April July
0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Time
April July
26 The first change does not affect power generation, since it is an offline operating strategy.
ASTE 1B consistently has lower overnight energy consumption than ASTE 1A, this is reflected in the monthly electricity consumption values that were already shown in Figure 13. The reason for this difference is that ASTE 1B is testing the use of a single auxiliary pump to feed the condenser instead of the primary cooling water pumps, reducing power consumption during the night. Something to take into consideration is the fact that the flow rate going through the condenser is lower than the manufacturer’s recommendation. This might result in deposits in the piping; however, the plant’s O&M staff is confident that the manufacturer’s recommendation is too conservative, since the manufacturer sets those values based on generalized conditions instead of specific applications and it overlooks the fact that the plants start and stop every day.
The effectiveness of this strategy will be confirmed or dismissed during the yearly inspection of the equipment.
The improvements that are noticeable in the power generation were made mainly in the solar field control scheme, which allowed higher HTF temperatures to be reached and resulted in a reduction of the energy needed by the pumps to circulate the HTF through the field. It also made the operation more stable. The gain in power output due to these improvements was measured to be between 0.50-0.80 MW. There were several improvements in control schemes, for example in the cooling towers; however, the actual control is out of the scope of this work and will not be addressed in detail.
Another recent improvement was the decision to use both of the primary HTF pumps in parallel instead of a single pump at 100% capacity. This also represented a cut in consumption and therefore an increase in power output of about 0.50 MW.
9 CRITICAL COMPONENTS
In order to include the maintenance part of the O&M in this work, a simple study to identify the critical components of the power plants was performed. The corrective actions executed during a period of 6 months were reviewed and classified by system, sub-system, component, and type. The components were sorted by number of occurrences and plotted in Pareto charts. This is a very extensive analysis, therefore, only the most important points will be outlined in this work.
The objective was to identify the components with the highest number of failures which consequently caused a considerable loss in production and to characterize these failures. The original idea was to compare the corrective actions of the two plants and identify components that consistently affected both of them; however, only one of the plants has been thorough
27 enough with their reports to perform this evaluation. Therefore, the information presented here only takes into account one of the two plants.
The first step was identifying the type of failure in the components. Figure 22 shows corrective actions by type. Around 70% of the maintenance interventions were mechanical, 17%
electrical and only 13% belonged to instrumentation and control (I&C) issues.
Figure 22. Corrective actions by type of intervention.
Identifying the failures by system would not throw any conclusive results, as the systems are too broad, so the failures were directly grouped by subsystem. Figure 23 shows the mentioned grouping where we can clearly identify subsystems with a substantial amount of failures.
Figure 23. Corrective actions by plant sub-system.
0%
20%
40%
60%
80%
100%
120%
0 50 100 150 200 250 300 350 400
Mechanical Electrical I&C
# Ocurrences Cumulative %
0,00%
20,00%
40,00%
60,00%
80,00%
100,00%
120,00%
0 20 40 60 80 100 120 140
# Ocurrences Cumulative %
28 The failures in solar collector assemblies (SCA) are more than two times higher than in any other subsystem. This number is not to be trusted blindly since there was an accident in the plant in the month of April. There was a fire caused by a fault in the temperature control of the solar field that resulted in a fire that accounts for almost half of the interventions reported. In any case, the failure count is still higher than for any other component. The most common failures are HCE and mirror breakage. These usually happen simultaneously; when the glass layer of the HCE breaks it falls onto the mirrors, breaking them. Although HCE and mirror breakage are the largest problem until now, these numbers should decrease as the plant moves on with its operation and the incident is left behind. Luckily, after the incident a stretch of cloudy days occurred and the energy loss was not nearly as large as it was initially estimated. No special preventive maintenance can be given to these elements; the only improvement to this date was a more conservative strategy was implemented to prevent mirror breakage because of high wind speeds.
The second and third sub-systems with the most incidents are the power cycle and steam generation components. These sub-systems are related and include several components within them, most of the actions registered were calibrations and leak repairs which did not impact the power generation. The fourth and fifth categories are the HTF circuit and the auxiliary boiler system. These two subsystems both involve HTF and are more relevant than they appear, as it will be shown in the next phase of this study.
The final evaluation was grouping the corrective actions by component. There were several corrective actions that were link to a system or sub-system, but no component was clearly identified. These actions were excluded from this last analysis in order to present clearer data.
Components with less than 3 registered incidents were also excluded in order to simplify the graph. Figure 24 shows this evaluation, where some critical components can be easily identified.
Figure 24. Corrective actions by plant component.
0,00%
5,00%
10,00%
15,00%
20,00%
25,00%
30,00%
35,00%
40,00%
45,00%
50,00%
0 20 40 60 80 100 120
140 # Ocurrences Cumulative %
29 The SCA’s had already been identified as a critical component, however as it was mentioned before there are not many things that can be done about them. Now we can identify other critical components such as the primary and auxiliary HTF pumps, which account for over 50 corrective actions. The steam generator has 18 incidents however more than half of them are leaks that were repaired without any major problems or energy loss.
The HTF pumps represent one of the biggest challenges for the plants, over the first 8 months of plant operation 8 mechanical seals had to be replaced (including all three plants).
Every time one of these seals needs to be replaced, the plants need to operate at reduced capacity, and depending on the plant configuration a full stop might be required.
Figure 25 shows an incident with a mechanical seal which caused a trip and also caused the plant to operate around 1 MW under its capacity. The graph only shows the day when the incident occurred; the loss on this day was close to 35 MWh. The replacement seal took another two days to arrive from the supplier and the plant lost approximately another 10 MWh each day, for a total loss of 55 MWh plus the cost of the spare parts.
Figure 25. Energy loss caused by a failure in a primary HTF pump.
The problem with the mechanical seals has been addressed with the supplier. However, no root-cause for these issues has been found yet and therefore problems will probably keep emerging. A partial solution in order to reduce the downtime was increasing the spare parts stock for the HTF pumps.
-20 -10 0 10 20 30 40 50 60
7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00
MW
Horas
Diff. ASTE 1A ASTE 1B
30 Another critical component that does not appear on the graphs but needs to be mentioned is the turbine. Turbines are one of the most important components of a CSP plant and one of the few components that cannot be fixed by the maintenance crew. Turbines are sensitive equipment and are usually loaded with several safety measures to ensure their proper performance. A turbine failure can be catastrophic for the financial expectations of the plants’ ownership. Repairs to the turbine can take several weeks and cause production losses that may amount to millions of euros.
In the case of these power plants the only problem that the turbines have presented are a limitation by the manufacturer. The steam inlet valves need to be replaced due to a design flaw and the plants cannot operate at full capacity until this replacement takes place. Since this limitation was applied in the summer months, the decision to postpone the valve replacement until the winter was taken to avoid larger production losses.
10 CONCLUSIONS & OUTLOOK
The CSP industry in Spain is in a very fragile situation. The government’s constant regulation changes keep worsening the conditions not only for CSP, but for RETs in general. The atmosphere of uncertainty has already stopped the growth of this technology in Spain and the changes threaten to ruin the existing CSP projects. The Spanish government needs to pay special attention to their financial previsions and come up with a definitive plan to manage both the energy deficit and the fact that renewable energies are not as profitable as fossil fuels and therefore need a FiT system. Investments in research need to be made in order to lower costs and benefit both the government and the industry in the long term by lowering costs.
From the last paragraph, it can be deduced that although PT-CSP and central receiver plants are operating commercially, CSP technology in general is still not mature enough. On the one hand it is clear that it depends heavily on the government’s financial incentives. On the other hand there are still issues with the plant components. In the case of these two plants, the turbine and the HTF pumps are an example of this lack of maturity, since both components were adapted to CSP from different industries, and they still present problems.
If we look deeper into the HTF pumps the only possible explanation has to be the HTF, according to the suppliers, these pumps have been used before in the refining industry with no apparent complications. However, when it comes to operating with HTF which is always close to its degradation limit, they start to present problems. A solution in the future could be changing the working fluid, seeing how the HTF that is used is toxic and flammable, presenting a hazard to the staff in the plants. Research is currently being done on this matter, and direct steam generation or molten salts may prove to be a good substitute in a not so far future.
31 In terms of the operation of the two plants mentioned, the results have been outstanding, nearing 100% of the possible power generation and operating well within the established guaranteed values.
As a final remark, it is clear that despite almost perfect operation the plants still struggle to be profitable. It is still in the air if the plants will be profitable or not since the Spanish government still hasn´t fixed the parameters to calculate the new incentive. The only thing that is certain is that CSP plants are a risky investment for the moment, and as long as the Spanish government refuses up with a stable and sufficient FiT scheme, it is possible that the Spanish CSP industry will go into a deep recession.
32 11 BIBLIOGRAPHY
Asociación PROTERMOSOLAR. (2013). PROTERMOSOLAR expresa su completo rechazo a la Ley de Medidas Fiscales para la sostenibilidad Energética aprobada por el Congreso.
Boletín de Noticias(1), 2013. Spain.
Ciria Repáraz, I. (2010). Análisis de la tecnología de concetración solar mediante colectores cilindro parabólicos para gereración eléctrica. Escuela Universitaria Ingeniería Técnica Industrial, Zaragoza, Spain.
CNE. (2013). Informe sobre los resultados de la liquidación provisional Nº 14 de 2012 y verifiacaiones practicadas: Sector Eléctrico. Comisión Nacional de Energía, Dirección de Inspección, Liquidaciones y Compensaciones.
Couture, T. D., & Mischa, B. (2013, April 19). Pain in Spain: New Retroactive Changes Hinder Renewable Energy. Retrieved May 17, 0213, from Renewable Energy World:
http://www.renewableenergyworld.com/rea/news/article/2013/04/pain-in-spain-new- retroactive-changes-hinders-renewable-energy
CSP World. (2012, December 6). Spain´s Government to Deal a death blow to CSP. Retrieved April 15, 2013, from CSP World: http://www.csp-world.com/news/20121206/00643/spains- government-deal-death-blow-csp
Elecnor. (2013). Retrieved May 17, 2013, from Elecnor Website: http://www.elecnor.es Flowserve. (2013). Flowserve. Retrieved July 20, 2013, from
http://www.flowserve.com/Industries/Power-Generation/Concentrated-Solar-Power González, A., & Rucinski, T. (2013, February 1). Spain unveils fresh energy reforms that hit renewables.
Retrieved April 12, 2013, from Reuters:
http://ww.reuters.com/asstes/print?aid=USL5N0B1B5J20130201
Institute of Solar Research. (2011). enerMENA CSP Teaching Materials. (T. Fend, & L. Qoaider, Eds.) Germany: German Aerospace Center (DLR).
Jefatura del Estado. (1997, November 27). Ley 54/1997. Boletín Oficial del Estado. Spain.
Jefatura del Estado. (2012, December 28). Ley 15/2012. Boletín Oficial del Estado. Spain.
Jefatura del Estado. (2012, January 28). Real Decreto-ley 1/2012. Boletín Oficial del Estado. Spain.
Jefatura del Estado. (2012, December 31). Real Decreto-ley 29/2012. Boletín Oficial del Estado.
Spain.
Jefatura del Estado. (2013, July 13). Real Decreto Ley 9/2013. Boletín Oficial de Estado. Spain.
Jefatura del Estado. (2013, January 2). Real Decreto-ley 2/2013. Boletín Oficial del Estado. Spain.
Madaeni, S., Sioshansi, R., & Denholm, P. (2012, February). How Thermal Energy Storage Enhances the Economic Viability of concentrating Solar Power. Proceedings of the IEEE, 100(2), 335-347.
