MaintenanceofaSmall ScaleParabolicTrough
MaintenanceofaSmall ScaleParabolicTrough
net/publication/322530317
Article in International Journal of Precision Engineering and Manufacturing-Green Technology · January 2018
DOI: 10.11648/j.ijrse.20170606.12
CITATIONS READS
4 3,604
4 authors:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Kary Ritter on 16 January 2018.
Email address:
[email protected] (K. A. Ritter III), [email protected] (M. J. Prilliman), [email protected] (T. L. Chambers),
[email protected] (J. R. Raush)
*
Corresponding author
Received: December 15, 2017; Accepted: December 27, 2017; Published: January 16, 2018
Abstract: An accurate estimate of fixed operating costs is essential to determine the financial viability of any proposed project.
Although other researchers have reported maintenance costs for large-scale concentrating solar power (CSP) plants in the United
States [1 - 2], there is currently little information available specifically for small-scale CSP or solar Industrial Process Heat (IPH)
plants. This paper discusses the maintenance of an operating small-scale CSP plant in Louisiana over a four year period. The
results are also applicable to a small-scale IPH plant. Maintenance activities and costs are discussed for the collector field, the
power block, and the cooling tower. For the collector field, a study of the degradation of mirror reflectance between washings
was performed for three different types of reflective polymer thin films (3M 1100, 3M 2020, and Konica Minolta). Overall, the
3M 2020 film provided better reflectivity between washings than the other films. An optimized mirror washing schedule was
determined. Optimal mirror washing schedules are very site-dependent, but for this humid subtropical location, the most
economical washing schedule was found to be every 114 days, or approximately three times per year. A recommended
maintenance plan for small-scale CSP and IPH plants is presented and actual maintenance costs over a four year period are
provided. It was found that maintenance costs for small-scale plants are substantially larger than for large-scale plants, and that
maintenance costs for small-scale IPH plants are much lower than for small-scale CSP plants, making IPH applications
significantly more attractive. The average annual maintenance cost for a small-scale CSP plant was found to be approximately
$457/kWe, or $0.27/kWhe. For a small-scale IPH plant the costs were $3.72/m2, $7.81/kWt, and $0.005/kWht.
Keywords: Solar Energy, Concentrating Solar Power, CSP, Maintenance Costs, O&M, Soiling
the equipment installed. The maintenance scope of work for a inspections, collector cleaning, electrical testing, monitoring,
solar power plant typically includes the solar equipment facility maintenance and feedback from operations [7].
The solar specular reflectance in the collector field should feasibility of the installation. The pilot concentrating solar
naturally be kept at its highest level to ensure high global yield, power plant uses reflective parabolic solar troughs to generate
but keeping the solar collectors clean in an economical roughly 500 kWt for process heat or 35 kWe of electricity using
manner is the biggest maintenance challenge for CSP an Organic Rankine Cycle (ORC) turbine-generator system as a
technology [8, 9]. There are many issues related to CSP power block. The system consists of three main components: (i)
performance due to soiling, and technologies are continually the solar collector field, (ii) the power block, and (iii) the
being developed to keep the reflectors clean with minimal use cooling system. Explanations of these components and the
of water [10]. With over 300 publications generated in the last required maintenance for each are presented, along with costs
five years alone, the effects of soiling and particle and an overall recommended maintenance schedule.
accumulation on solar collectors is a high interest topic and a
large part of the current maintenance study presents the
optimization of collector reflectivity in the Louisiana climate
by means of an optimal collector washing schedule.
Cleco Power LLC and the University of Louisiana at
Lafayette (UL Lafayette) completed the installation and
commissioning of a pilot-scale concentrating solar thermal
power plant, the first of its kind in Louisiana, in December
2012 [11 - 14]. All components in the system are
commercially available and have proven to be successful in
other states, but prior to this installation there was no data
available to determine whether the technologies would be
effective in the Louisiana area. Therefore, the pilot plant
provides Louisiana-specific performance and price
information regarding the use of CSP technology in Louisiana.
The plant is also being used to study innovative new CSP Figure 2. Cleco Alternative Energy Center and UL Lafayette START Lab.
technologies [15]. The pilot plant was installed at the UL
Lafayette Energy Research Complex, which includes the
Cleco Alternative Energy Center and the UL Lafayette Solar
2. Solar Collector Maintenance
Technologies Applied Research and Testing (START) Lab, as 2.1. Solar Collectors
shown in Figure 2 below.
The pilot project objectives are to test a CSP system under The solar collector field consists of 12 reflective parabolic
actual conditions in Louisiana, to gain experience in Large Aperture Troughs (LATs), which sit on approximately
maintaining and operating such a system, to determine the 1 acre of land, as shown in Figure 3. Each trough is roughly 12
scalability of the technology, and to determine the overall m long by 7.3 m wide, and has an effective reflective area of
106 Kenneth August Ritter III et al.: Maintenance of a Small-Scale Parabolic Trough Concentrating Solar
Power Plant in Louisiana
87.6 m2. The reflectors consist of thin film polymer complete washing of the system (1050 m2) required
technology provided by 3M and Konica Minolta. Schott approximately 16-man hours and used about 30 gallons of
PTR70 heat collection element (HCE) tubes with 70 deionized water. This cleaning method returned the overall
millimeters outside diameter are employed which, when reflectivity of the aperture to a value near that of the original
combined with the large aperture, results in an industry performance specification of 95.5% [19].
leading concentration ratio (the ratio of the area of collected
radiation to the area of concentrated radiation) of 104. 2.3. Measurement of Reflectivity
Soiling, or degradation of the reflectiveness of the Since the start of plant operation, a micro-TRI-gloss
concentrating reflectors, has a direct impact on the performance glossmeter by Byk-Gardner has been used to measure
of CSP plants [16]. The soiling effect lowers the optical reflectance of the solar reflectors. This device reports a
efficiency of the reflectors, which results in less of the available derived reflectivity value rather than a direct measurement
solar radiation being reflected onto the absorber tube. This [20]. This glossmeter offers easy one-time calibration versus a
results in a smaller energy output and an increase in the reference, and allows the entire collector field (1050 m2) to be
levelized cost of electricity or heat. An economic analysis was measured in less than 1 hour. The data collection method used
performed at the Kramer Junction solar power park located in in this study is unique from that of large-scale power plants in
Boron, California and indicated that maintaining an average that measurements are taken of the entire reflector field rather
field reflectivity above 90% is cost-effective [1]. Frequent than just a representative sample.
reflector washing is therefore required, and the effectiveness of Measurements taken of the specular reflectance showed that
the washing was found to vary with location and time of year. following washing, the reflectors returned to original
2.2. Reflector Washing performance specifications for reflectivity (> 95%) after having
been reduced to less than 80%. The gloss measurements can be
In a previous study of different cleaning methods, the most correlated to specularity of the reflectors, a major factor in
effective was found to be using deionized water and a brush optical efficiency. The degradation rate of the reflectors from
resulting in an average cleaning efficiency of 98.8% in rainy particle accumulation continues to be monitored continuously.
periods and 97.2% in dry seasons [17]. In a recent study a soft
cleaning brush and small amount of water was found to be the 2.4. Experience on Collector Reflectivity
most effective way to clean the thin film polymer without The Cleco Alternative Energy Center’s location in a humid
inflicting surface damage or reducing specular reflectance [18]. subtropical region of the United States presents unique issues
related to soiling, and subsequent thermal losses, of the solar
reflectors. More specifically, its location in southern
Louisiana offers much less soiling in the form of sand and dirt
debris when compared to the Solar Electric Generating
Stations (SEGS) plants on the west coast [1], while the higher
humidity of the area also serves to reduce the effects of soiling.
The amount of rainfall can also have a variable impact on
soiling, with this impact potentially being magnified by
planned SCA rotations to maximize rainfall on the reflector
surfaces. As such, it is essential to determine the optimal
reflector washing schedule and procedures to maintain
near-ideal operating conditions and to reduce thermal losses
without incurring redundant O&M costs from cleaning efforts
that provide little gain in system efficiency.
Figure 3. Reflector Washing.
2.5. Soiling Rate
The washing procedure currently employed at the Louisiana
plant involves using a pressure washer with deionized water Following the May 2017 reflector washing, the glossmeter
and a microfiber cloth attached to a pole brush designed by 3M. was used to take weekly reflectivity measurement of the entire
This brush consists of a long pole attached to a brush head that collector field. The three types of reflective thin film tested in
clamps the microfiber cloth down on a sponge that has running this study were: 3M 1100, 3M 2020, and Konica Minolta. Two
water flowing to it to reduce surface friction. Shown in Figure 3, measurements were taken per panel, so 480 measurements
the reflector washing procedure consists of an initial spray of were taken each week. One goal of this maintenance study
water with a pressure-washer, followed by wiping with the was to determine whether the current bi-annual reflector
brush before the reflectors are sprayed again. washing is cost effective. This determination is based on a
Reflector washing was completed following approximately calculated correlation between the drop in average reflectivity
one year of deployment of the facility. In 2015, a full washing and thermal efficiency of the reflectors as a function of time.
of the East Solar Collector Assembly (SCA) was conducted, The rate of soiling over the summer season of 2017 for the
however the West SCA was not washed for test purposes. The three different types of thin-film is shown in Figure 4.
International Journal of Sustainable and Green Energy 2017; 6(6): 104-111 107
Film Type Degradation Rate (%/day) Reflectivity when New (%) 4 Months Post-Wash (%) Reflectivity after Wash (%)
3M 1100 0.044 94.4 87.6 92.5
3M 2020 0.078 98.9 89.4 97.4
Konica Minolta 0.063 95.3 86.3 93.8
The soiling rate is one of several parameters that go into the It works in a manner similar to a steam turbine generator
equation for determining the optimal cost-effective cleaning system, except that the working fluid for the power block is an
schedule for concentrating reflectors [21]. This equation can organic refrigerant, R245fa, which has a much lower boiling
be expressed by: point than water. The refrigerant working fluid picks up
thermal energy as it passes through a liquid-to-liquid heat
(1) exchanger, where hot water from the solar collector field is on
one side of the heat exchanger, and the refrigerant is on the
Where is the ideal number of days between reflector other side. The hot refrigerant is allowed to expand and create
washing, is the cost of this cleaning per square meter of vapor in a boiler, and then the refrigerant vapor is converted to
surface area, is the optical efficiency of the reflectors, mechanical energy by expanding it through a twin-screw
is the average daily solar energy available per square meter expander system. After the working fluid is expanded through
of surface area at the location in question, is the soiling the expander, it is condensed by passing through another heat
rate of the reflector surface as a percentage of the restored exchanger. This time the hot refrigerant is on one side of the
reflectivity value, and is the energy price, expressed in heat exchanger, while cold water from a cooling tower is on
dollars per kilowatt-hour, at the specified location [21]. the other side. The refrigerant is condensed as it passes
For the Louisiana site the following values were used: W = through the heat exchanger and it is pumped back to the boiler,
$0.78/m2; A0 = 0.62; I0 = 4.5 kWh/m2/day; D = 0.000467; and C and the cycle starts again. The twin-screw expander turns an
= $0.092/kWh, based on the U.S. Energy Information AC generator that produces three-phase electrical power at
Administration’s commercial pricing data for July 2017 [22]. As 480 V and 60 Hz, which is synchronized to the grid. Figure 5
a result, the optimal interval for washing the collectors, Nc, was shows the major components in the power block.
calculated to be 114 days, or about 3 times per year. Louisiana
has one of the lowest electricity cost in the country, which makes
it cost effective to allow the collectors to go longer between
washings. By way of comparison, keeping all other values
constant, states with higher electricity prices such as California
($0.177/kWh) and Hawaii $(0.265/kWh) would have washing
intervals of 82 days and 67 days respectively, to minimize cost.
replaced costing near $2,000. Also in 2016 two hundred Note that the collector field costs are based only on
pounds of refrigerant were added costing over $4,000. The maintenance to the collector field, not to the power block or
variable costs for each plant component tend to average out cooling tower, which would be appropriate if one were
over all components, however, such that the cost in each year designing a plant to provide Industrial Process Heat (IPH)
has been relatively constant over the four year period studied, for a process such as desalination, rather than electricity.
at an average overall maintenance cost of $16,008/year. These results show that the maintenance costs for a
In order to make these cost numbers scalable and comparable small-scale IPH facility would be substantially less than for a
to other technologies, it is helpful to express maintenance costs small-scale CSP plant. This is partly due to the low
in terms of normalized values, such as cost per kWe and cost efficiency of the ORC (7%) when producing electricity at
per collector field area. Previous research [14, 25] has this scale, and it is also partly due to the fact that for process
determined that at this location there is an average of 273 days heat, the maintenance costs for the power block and cooling
per year when the Direct Normal Irradiance (DNI) is high tower do not apply. Even at a small scale, however, the
enough (400 W/m2) to operate the plant. On those days there is annual cost of $7.81/kWt found in this study compares
an average DNI of 679 W/m2 for an average of 6.3 hours per favorably to the current default value of $5/kWt provided in
day, for a total of 1,720 hours of production per year. The the System Advisor Model (SAM) for large-scale IPH plants
collector area is 1050 m2 and the solar to thermal efficiency is [26], which is provided free of charge from the National
approximately 70%. Thus the collector field provides a nominal Renewable Energy Laboratory (NREL).
500 kWt. The Organic Ranking Cycle (ORC) power block has
an efficiency of approximately 7%, resulting in a nominal 6. Maintenance Schedule
electrical power output of $35kWe. Normalized annual costs in
various units are shown in Table 4 below. As noted by Eaton [27], general inspection and evaluation
of the CSP plant should be conducted on a regular basis that
Table 4. Normalized Annual Maintenance Costs for Small-Scale CSP Plants. is consistent with system use and local environment. A
Overall Costs maintenance schedule for the Louisiana plant has been
Annual Cost/kWe $457.36 developed, which may be helpful when planning future
Annual Cost/kWhe $0.27 plants. Table 5 shows the maintenance tasks with the
Collector Field Costs
corresponding frequency the maintenance should be
Annual Cost/m^2 $3.72
Annual Cost/kWt $7.81 performed.
Annual Cost/kWht $ 0.005
Component Required Maintenance Times Per Year Every 2 Yrs Every 5 Yrs
Reflector Washing 3
Tube Pressure Wash 2
Repairing Film 2
Collectors
Soap/Pressure Wash Pillars 1
Inspect Frame and Brakes 1
Gloss Measurement/Recording 52
Change Filters X
Change Gaskets X
Hydraulics
Fluid Level Check 12
Change Fluid X
Power Block Soap/Pressure Wash Casing 4
Check chemicals 12
Cooling Tower
Flush, scrub, refill
Levelling Legs 52
Radiometer Wipe Lenses, Bulbs 365
Calibrate X
Reporting O&M Report 4
Cut Grass 52
Grounds
Spray Weeds 4
Check clock synchronization 365
SCADA
Backup all data 12
The reflector washings have previously been conducted company. The radiometer devices were calibrated in the fall of
twice per year, but the recent soiling study described above 2017.
indicates that three times a year is the most cost effective for
this site. When checking the hydraulic fluid in the East Stow 7. Summary and Conclusions
position, when the cylinders are retracted, the upper site glass
should show full [26]. The monthly cooling tower A study of maintenance activities and their associated
maintenance is handled by the commercial water treatment costs has been performed for a small-scale CSP facility in
110 Kenneth August Ritter III et al.: Maintenance of a Small-Scale Parabolic Trough Concentrating Solar
Power Plant in Louisiana
[6] Zipp, K. (2013), “What Is A Solar Operations And [16] Mejia, F. A., & Kleissl, J. (2013), “Soiling losses for solar
Maintenance Plan?” Retrieved September 21, 2017, from photovoltaic systems in California,” Solar Energy, 95, 357–
https://www.solarpowerworldonline.com/2013/03/what-is-a-s 363. https://doi.org/10.1016/j.solener.2013.06.028.
olar-operations-and-maintenance-plan/.
[17] Fernández-García, A., Álvarez-Rodrigo, L., Martínez-Arcos,
[7] Kostok, K. (2014), “Implementing a Successful O & M L., Aguiar, R., & Márquez-Payés, J. M. (2013). “Study of
Strategy for Solar PV,” White paper. Alectric USA, Miami, different cleaning methods for solar reflectors used in CSP
Florida. Retrieved December 12, 2017, from plants,” Energy Procedia, 49, 80–89.
http://www.alectris.com/library/Alectris%20White%20Paper https://doi.org/10.1016/j.egypro.2014.03.009.
%20-%20Implementing%20a%20Successful%20OM%20Stra
tegy%20for%20Solar%20PV.pdf. [18] Sansom, C., Fernandez-Garcia, A., Sutter, F., Almond, H., &
King, P. (2016). Contact cleaning of polymer film solar
[8] Karim, M., Naamane, S., Delord, C., & Bennouna, A. (2015), reflectors,” In AIP Conference Proceedings (Vol. 1734, pp.
“Study of the Surface Damage of Glass Reflectors Used in 20022-1-20022–8). https://doi.org/10.1063/1.4949046.
Concentrated Solar Power Plants,” Energy Procedia, 69, 106–
115. https://doi.org/10.1016/j.egypro.2015.03.013. [19] Gossamer Space Frames and 3M. (2012), “Large Aperture
Trough (LAT) 73 engineered by Gossamer Space Frames and
[9] Pernick, R., & Wilder, C. (2008), “Utility Solar (USA) Study 3M,” 3M Corporation, St. Paul, Minnesota. Retrieved
Reaching Ten Percent Solar By 2025,” Co-op America December 14, 2017 from
Foundation. Washington, DC. July 2008. Retrieved December http://multimedia.3m.com/mws/media/813507O/large-apertur
12, 2017, from e-trough-lat-73.pdf?fn=GossamerLargeApertureTrough_LAT
https://community-wealth.org/content/utility-solar-assessment 73_DMR_9802425.pdf.
-usa-study-reaching-ten-percent-solar-2025.
[20] Meyen, S., Montecchi, M., Kennedy, C., & Zhu, G., Gray, M.,
[10] Xu, X., Vignarooban, K., Xu, B., Hsu, K., & Kannan, A. M. Crawford, J., Heimer, S., Platzer, W., Heimsath, A., O’Neill,
(2016), “Prospects and problems of concentrating solar power M., Ziegler, S., Brandle, S., Fernandez, A. (2013),
technologies for power generation in the desert regions,” “Parameters and method to evaluate the solar reflectance
Renewable and Sustainable Energy Reviews, 53, 1106–1131. properties of reflector materials for concentrating solar power
https://doi.org/10.1016/j.rser.2015.09.015. technology,” Solar PACES Guidelines. Retrieved December
14, 2017 from
[11] Raush, J., Chambers, T., Russo, B. 2013, “Demonstration of http://elib.dlr.de/84546/1/201306_SolarPACES-Reflectance-G
Pilot Scale Large Aperture Parabolic Trough Organic Rankine uidelines-V2_5.pdf.
Cycle Solar Thermal Power Plant in Louisiana,” Journal of
Power and Energy Engineering, Vol. 1, No. 7, pp. 29 – 39. [21] Bergeron, K. D., & Freese, J. M. (1981), “Cleaning Strategies
Available at: http://dx.doi.org/10.4236/jpee.2013.17006. for Parabolic Trough Solar Collector Fields; Guidelines for
Decisions,” Sandia National Laboratories, Alburquerque, new
[12] Chambers, T. L., Raush, J. R., Massiha, G. H., 2013, “Pilot Mexico. Retrieved December 14, 2017 from
Solar Thermal Power Plant Station in Southwest Louisiana,” https://www.osti.gov/scitech/servlets/purl/6376410.
International Journal of Applied Power Engineering (IJAPE),
Vol. 2, No. 1, April 2013, pp. 31 – 40. ISSN: 2252-8792. [22] U.S. Energy Information Administration (EIA). (2017).
Available at: Electric Power Monthly: with data for July 2017. U.S.
http://www.iaesjournal.com/online/index.php/IJAPE/article/vi Department of Energy. https://doi.org/10.2172/123200.
ew/1941/0.
[23] Electra Therm. (2016), “Installation and Operations Manual,”
[13] Chambers, T. L., Raush, J. R., Russo, B., 2014, “Installation Available from Electra Therm Corporation, Reno, Nevada.
and Operation of Parabolic Trough Organic Cycle Solar
Thermal Power Plant in South Louisiana,” Energy Procedia, [24] SPX Corporation. (2016), “Trouble-Free Hvac Cooling
Vol. 49, 2014, pp. 1107 – 1116. Available at: Towers: A Maintenance Guide,” SPX Cooling Technologies,
http://www.sciencedirect.com/science/article/pii/S187661021 Inc., Overland Park, KS. Retrieved December 14, 2017 from
4005748. http://spxcooling.com/pdf/Cooling-Tower.pdf.
[14] Raush, J. R., Chambers, T. L., Russo, B., Crump, K., 2016, [25] Ezeanya, Kelvin Emeka, “System Advisor Model (SAM)
“Assessment of Local Solar Resource Measurement and Simulation Modeling of a Concentrating Solar Thermal Power
Predictions in South Louisiana,” Energy, Sustainability and Plant, with Comparison to Actual Performance Data,”
Society (2016) 6:18, DOI 10.1186/s13705-016-0083-y, University of Louisiana at Lafayette, Master’s Thesis, Master
Available at: http://rdcu.be/um5E. of Science in Mechanical Engineering, December, 2017.
[15] Raush, J., Chambers, T., 2014, “Initial Field Testing of a [26] NREL System Advisor Model (SAM), available at:
Concentrating Solar Photovoltaic (CSPV) Thermal Hybrid https://sam.nrel.gov/.
Solar Energy Generator Utilizing Large Aperture Parabolic
Trough and Spectrum Selective Mirrors,” International [27] Eaton, J. H. (2016), “Operation and Maintenance Manual
Journal of Sustainable and Green Energy, Vol. 3, No. 6, pp. Large Aperture Trough System – LAT73,” Available from
123 – 131, 2014. Available at: Gossamer Space Frames, Laguna Hills, California.
http://www.sciencepublishinggroup.com/journal/paperinfo.asp
x?journalid=169&doi=10.11648/j.ijrse.20140306.12.