An investigation on factors influencing dust accumulation on CSP mirrors S. Pennetta, S. Yu, P. Borghesani, M. Cholette, John Barry, and Z. Guan Citation: AIP Conference Proceedings 1734, 070024 (2016); doi: 10.1063/1.4949171 View online: http://dx.doi.org/10.1063/1.4949171 View Table of Contents: http://aip.scitation.org/toc/apc/1734/1 Published by the American Institute of Physics Articles you may be interested in Soiling and degradation analysis of solar mirrors AIP Conference Proceedings 1734, 090001 (2016); 10.1063/1.4949186 Specular reflectance of soiled glass mirrors – Study on the impact of incidence angles AIP Conference Proceedings 1734, 130009 (2016); 10.1063/1.4949219 A numerical study on high-pressure water-spray cleaning for CSP reflectors AIP Conference Proceedings 1734, 160001 (2016); 10.1063/1.4949242 Standards for components in concentrating solar thermal power plants - status of the Spanish working group AIP Conference Proceedings 1734, 110003 (2016); 10.1063/1.4949200 Water recovery in a concentrated solar power plant AIP Conference Proceedings 1734, 160014 (2016); 10.1063/1.4949255 Comparing the net cost of CSP-TES to PV deployed with battery storage AIP Conference Proceedings 1734, 080003 (2016); 10.1063/1.4949183
Transcript
An investigation on factors influencing dust accumulation on CSP mirrorsS. Pennetta, S. Yu, P. Borghesani, M. Cholette, John Barry, and Z. Guan
Citation: AIP Conference Proceedings 1734, 070024 (2016); doi: 10.1063/1.4949171View online: http://dx.doi.org/10.1063/1.4949171View Table of Contents: http://aip.scitation.org/toc/apc/1734/1Published by the American Institute of Physics
Articles you may be interested inSoiling and degradation analysis of solar mirrorsAIP Conference Proceedings 1734, 090001 (2016); 10.1063/1.4949186
Specular reflectance of soiled glass mirrors – Study on the impact of incidence anglesAIP Conference Proceedings 1734, 130009 (2016); 10.1063/1.4949219
A numerical study on high-pressure water-spray cleaning for CSP reflectorsAIP Conference Proceedings 1734, 160001 (2016); 10.1063/1.4949242
Standards for components in concentrating solar thermal power plants - status of the Spanish working groupAIP Conference Proceedings 1734, 110003 (2016); 10.1063/1.4949200
Water recovery in a concentrated solar power plantAIP Conference Proceedings 1734, 160014 (2016); 10.1063/1.4949255
Comparing the net cost of CSP-TES to PV deployed with battery storageAIP Conference Proceedings 1734, 080003 (2016); 10.1063/1.4949183
Abstract..The profitability of a CSP plant is highly affected by the efficiency of the solar field: it is essential to maintain mirrors’ reflectivity at high level to avoid thermal power loss. Dust fouling is the main cause of reflectivity loss and cleaning of mirrors is a crucial activity to restore economical level of reflectivity. However, the high cost of cleaning operations requires the study and identification of a balanced plan for the dust removal. The dust generation and transport to the plant site is the first mechanism that needs to be modelled to identify the optimal schedule for cleaning operations and it is highly dependent on weather conditions. Several studies have suggested a dependency of reflectors performance with humidity level, frequency of rainfalls, wind and mirrors’ tilting angle, however rarely quantitative correlation studies have been performed to validate these hypotheses. The aim of this research is to provide an in-depth insight on interaction between the main parameters and airborne dust concentration, providing quantitative information for the development of future mirror dusting models. Outcomes evidence the crucial role of high winds responsible of dust concentration in conjunction with higher wind direction frequencies in the range 60-120°. Actually, in this scenario a perfectly monotonic increase of dust accumulation in the air has been observed with high correspondence of wind direction. A very low effect is provided by the ambient temperature as the contribution of the barometric pressure.
INTRODUCTION
Concentrating Solar Thermal technology has observed a boost of CSP plants installation during the last 20 years. Spain traditionally led this sector thanks to incentives issued from the government since the ’90s with a feed-in tariff system, promoting the expansion of CST technologies across the country. Spain’s experience highlights that appropriate site locations are important to get the maximum performance out of CST technology. CSP plants works with high efficiency in arid to semi-arid region where direct normal irradiance (DNI), is typically above 2000 kWh/m2 [1-3].
Solar fields (SF), power block (PB) and additional thermal energy storages (TES) generally characterize CSP plants. The impact of the solar field in the Operation and Maintenance (O&M) is high, due to cost intensive. Mirror cleaning and reflectance measurements are the most common activities and have a remarkable impact on the levelised cost of energy (LCOE). The fraction of these activities on the total cost of electricity produced by CSP plants is approximately 8% [4]. On the other hand, the cleaning activities are necessary to restore the solar field efficiency, compromised by dust deposition and accumulation on mirror surfaces [5].
Dust particles absorb or scatter solar beams, producing a significant drop of plant reflectivity. Kaldellis et al. [6] and Goossens and Van [7], have studied the impact of dust accumulation on a photovoltaic system and have confirmed that high particle concentration has a remarkable negative effect on power generation. El-Nashar [8] has
SolarPACES 2015AIP Conf. Proc. 1734, 070024-1–070024-7; doi: 10.1063/1.4949171
Published by AIP Publishing. 978-0-7354-1386-3/$30.00
070024-1
studied the seasonal effect of dust deposition on a field of evacuated tube collectors of a solar desalination plant. The system is located near the city of Abu Dhabi, UAE, and the results are therefore relevant to this region. Dust deposition causes a monthly drop in glass tube transmittance of 10 – 18% with large drop in plant production. The author evidences a transmittance decrease from an initial value of 0.98 (clean glass condition) to a low value of 0.6, corresponding to a very dusty glass condition. The production drops from 100% to 40% of the clean collector production level. The degradation of heliostat efficiencies caused by dust accumulation has been studied by Strachan, who has pointed out an average reflectivity decrease of 6.3 and 8.8% respectively for the two types of heliostats investigated in their study [9].
Given a thermal energy loss production of 1.2% for each 1% point of reflectivity drop, the identification of the optimal balance between more and less O&M activities (cleaning) represents an important aspect for CST plant economic feasibility[10]. The optimal balance is extremely related to the history of dust concentration, which in turn is site specific as airborne particulates, dust sources and paths differs with location. Weather conditions affect plant reflectivity, heavy rainfalls can easily restore the plant reflectivity close to the prefixed optimal threshold while light rains or dews improve the adhesion of soil particles on mirrors surface by the cementation process [11]. Roth and Anaya [12] have conducted a study on the effect of rainfall versus the specular reflectance, collectors performances are usually restored to their nearly full capacities after heavy rains. When rainfalls with low intensity happen, only a small gain in reflectance has been observed as represented by the short distance between a trough and a peak. Reflectivity drop of 14% points occurs when strong winds or dust storms are followed by light rains. Heavy rainfalls are able to restore mirror reflectance up to 12% while low intensity atmospheric phenomena cannot handle soiling matters in an effective way. Water droplets generally collect airborne particulates depositing a high concentration of residues during the impact on collectors’ surface. Cuddihy [13] describes the cementation process as the interaction of the soluble part of soil with water droplets resulting with the build-up of soiling layers consisting of water moisture and dust particles. This process is very common to those regions characterized by high dust and humidity level because dust particles, made up of water soluble and insoluble particulate melt with humidity and after a drying period this mixture sticks as cement on surfaces. The development of soiling layers acting as proper surface coatings become harder to remove by standard cleaning procedures. Suresh Kumar et al. [14] identified also temperature, humidity and wind speed as influencing parameters for dust settlement on PV panels. Findings evidenced a loss of efficiency up to 20% with 0.3 g of dust is collected on PV cells.
This study aims at identifying weather conditions responsible of CST plant reflectivity degradation through the analysis of parameters such as temperature, relative humidity, wind speed and wind direction. Data on dust concentration in air has been collected at Collinsville (QLD, Australia) through real time-monitors equipped with swappable Teflon filters of 47mm. In this paper the correlation analysis of climatic variables with dust concentration in air is presented.
MEASUREMENT SETUP AND DATA DESCRIPTION
Measurement System and Site Description
Dust concentration is affected by the ambient conditions and measurements of these parameters are a key step towards the analysis of dust fouling of CST mirror field. The dust concentration data presented in this study has been collected using an Ecotech E-Sampler Light Scattering Monitor whose technical specifications are summarized in table 2.
Smallest data / Reading resolution 1 minute Range 0 – 500 g / m3 or
0 – 1000 g / m3 or 0 – 10 000 g / m3 or 0 – 65 000 g / m3
Resolution +⁄− 3 g/m3 or 2% of reading Pa Operating Temperature − 10 to 50 °C Battery bank size @ 12v 100 Ah Solar array size 125 W
070024-2
The E-Sampler is equipped with a solar power panel, a wind speed and direction sensor(Met One Instruments 034B), and a mounting frame as shown in FIGURE 1.
FIGURE 1 Real-time dust monitor system at Collinsville
This real-time monitor is characterized by a dual technology, the light scattering technology and the gravimetric method. The first one provided real time measurements of airborne particulate by light scatter, while the second provides the calibration factor for the measurement of the light scattering and particulate characterization for size and properties analysis by the 47 mm filter system. Each 30 days the filter is exchanged with a new one and dust accumulated is sent to be analysed in Queensland University of Technology laboratories for concentration, size and physical and chemical analysis. The real time monitoring system is remotely controlled providing the data of dust concentration, wind speed, wind direction, temperature, humidity, date and time for each filter collected from the monitor.
Data Description
The data study considers a 12 months set of data from 1/06/2013 to 1/06/2014 recorded at Collinsville in state of Queensland, Australia. This data set of information presents different time sampling rate: at the beginning the E-sampler monitor has been set at 1 minute of time sampling rate not good for the amount of memory available to store all the data during a day. Indeed once the memory is full the E-sampler proceed to overwrite the oldest records. To prevent this, a second time sampling rate has been set at15 minutes leaving enough time for dust accumulating on the filter. Moreover three months of missing data from November 2013 to March 2014 are due modem connection issues.
The data discussed in this paper includes a series of five channels:
TABLE 2 Measurement list Measurement Unit Resolution Symbol
Dust concentration in air g/m3 Variable Temperature °C 0.1 °C Atmospheric pressure Pa 20 Pa Wind speed m/s 0.1 m/s Wind direction Degrees (CW from North) 1 deg
Wind direction is measured clockwise from North, as indicated in FIGURE 2. The data, despite being sampled with an uneven sample rate, presents the same sampling for all channels, allowing for direct correlation. Therefore the time information is neglected, with a progressive -numbering of the samples of each channel and . Data samples with the same are always simultaneously measured.
070024-3
Site Description
The environment of our dust monitor is shown in Google Maps at the following coordinates 20°32′36″S 147°48′25″E (FIGURE 2). The real time monitor is screened by a coal power station on the southwest, and surrounded by two coal mines: the Collinsville mine 5km on the west and the Sonoma mine 8 km on the southeast. An arid open landscape surrounds the site on the north and east, with few bushes and no trees in a 500 m range from our monitoring station.
FIGURE 2 Real-time monitor located at Collinsville in the state of Queensland, Australia (Google Maps)
METHODOLOGY
A correlation study is proposed in this paper between the different weather parameters (temperature, pressure, wind speed and wind direction) and the dust concentration in air .
The first step in the analysis is the transformation of the raw data to account for the particular characteristics of dust concentration and wind measurements. As visible in FIGURE 3 (a-b), the dust concentration shows an extremely leptokurtic distribution, with most points with very low values and few occurrences of very high concentrations. For this reason a logarithmic transformation is performed on the concentration data, with to avoid non-finite for null raw concentrations. The result of the transformation is shown in the diagrams of FIGURE 3 (c-d). The second key transformation is operated on the wind direction, bringing all the measurements in the range -180°:180°, thus recognising the “circular” periodicity of the measurement.
The second step in the correlation analysis was the clustering of the data for the calculation of the statistics of each cluster. As visible for the concentration in FIGURE 3 (c) the data is highly stratified, as a consequence of the setting of the analog-to-digital converter (ADC). The stratification of the data is time-independent for all measurements except for the dust concentration channel, which shows a variation in the ADC settings in the last part of the dataset. The stratification is used as a maximum resolution clustering size for all measurements except dust concentration.
FIGURE 3 Dust accumulation during the year: (a) raw data, (b) distribution of raw data, (c) log-concentration and (d)
distribution of the log-concentration
070024-4
Therefore, for each -th weather parameter the concentration data is clustered in a series of clusters ,
, where is the number of different levels present in . Each cluster is thus defined as:
(1)
For each cluster the mean concentration is then calculated as . Similarly the standard deviation is obtained as . The results of this procedure are presented in correlation plots in the following section.
RESULTS
The results of the correlation analysis are presented in FIGURE 4. Temperature (FIGURE 4 (a)) shows a macroscopically positive correlation with concentration, with the highest
concentration levels recorded at the highest temperatures. A local minimum at 28°C has a possible explanation in seasonal effects and requires further investigation.
The effect of atmospheric pressure (FIGURE 4 (b)) is much clearer, with a substantially flat curve, except for very low and very high pressure levels. This could be caused by the strong connection between extreme pressure levels and rainfall, with sunny days in case of high pressure states and rainy days in case of low pressure. The higher dispersion of data for lower pressures is surely dependent on the scarcity of rain events but may also suggest a more complex relationship in case of low precipitation, as suggested in literature.
The correlation with wind speed (FIGURE 4 (c)) confirms the hypothesis of the crucial role of high winds in the generation and transport of airborne dust. The curve shows a clear almost perfectly monotonic increase with an almost flat low speed part and a rapidly increasing slope for moderate to high winds. This suggests that a level of approximately 3-4 m/s represents a threshold for significant effects of wind speed on dust concentration.
Finally, FIGURE 4 (d) shows a preferential direction for dust transport to the site, with higher average concentrations recorded for easterly winds. FIGURE 5 represents the same information on a polar plot, where the increase in average dust concentration for winds in the range 60-120° is clearer (blue dots).
FIGURE 4 Correlation analysis of dust concentration (red points represent maximum e minimum values, blue points represent the average ones) vs. (a) temperature, (b) atmospheric pressure, (c) wind speed and (d) wind direction with 0° equal to North.
0 10 20 30 40 5010
0
101
102
AT(°C)
Dus
t con
c. (
g/m
3 )
9.75 9.8 9.85 9.9 9.95 10 10.05
x 104
100
101
102
BP(Pa)
Dus
t con
c. (
g/m
3 )
0 2 4 6 8 1010
0
101
102
WS(m/s)
Dus
t con
c. (
g/m
3 )
-200 -150 -100 -50 0 50 100 150 20010
0
101
102
WD(deg)
Dus
t con
c. (
g/m
3 )
10(a)
10(b)
102(c) 10
2(d)
070024-5
The diagram also presents the frequencies of the different wind direction records (red histograms). The higher number of occurrences in the right half plane explains the lower dispersion of the mean dust concentrations. The presence of power plant buildings screening winds from the south west, partially explained the low impact of the coal mines on the dust distribution. Moreover, the Collinsville mine saw scarce and discontinuous operation during the acquisition period. The high dust concentrations caused by easterly winds can be explained with the presence of a corridor of flat land with scarce vegetation and the Collinsville town, in addition to the road, mainly used by mine and construction trucks.
The impact of the relative humidity on the dust accumulation requires further investigations, since preliminary analysis evidences a high fluctuation during the day with a different scenario during a longer period of time. The reasons that can justify this scenario might be found in the role played by other factors such as wind speed, wind direction and temperature.
FIGURE 5 Wind diredction polar plot: correlation analysis with dust concentration (blue dots) and wind direction event
frequency (red histograms)
CONCLUSION
In this paper a correlation study between dust concentration and relevant weather parameters has been considered in a one year time interval, from June 2013 to June 2014 at Collinsville (Queensland, Australia).
Results show strong correlations, especially with wind speed and direction, suggesting the possibility to develop a model based prediction of the dust concentration in air. This model would benefit the prediction of mirror dusting in CSP plants, in turn allowing for an optimised planning of cleaning activities.
Further data collection could improve the following study by the development of a large dataset useful to identify seasonal trends of dust concentration versus the parameters considered in this study.
ACKNOWLEDGEMENT
This research was performed as part of the Australian Solar Thermal Research Initiative (ASTRI), a project supported by the Australian Government, through the Australian Renewable Energy Agency (ARENA). We also acknowledge the support of the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Queensland University of Technology (QUT) and the University of Queensland (UQ).
30
210
60
240
90 (E)270 (W)
120
300
150
330
180 (S)
0 (N)
070024-6
REFERENCES
1. A. Fernández-García, L. Álvarez-Rodrigo, L. Martínez-Arcos, R. Aguiar, and J. Márquez-Payés, "Study of Different Cleaning Methods for Solar Reflectors Used in CSP Plants," Energy Procedia, vol. 49, pp. 80-89, 2014.
2. A. Fernández-García, M. E. Cantos-Soto, J. León, and R. López-Martín, "Optimization of some key aspects of CSP plants maintenance," ed: SolarPACES, 2010.
3. A. Fernandez-Garcia, E. Zarza, L. Valenzuela, and M. Pérez, "Parabolic-trough solar collectors and their applications," Renewable and Sustainable Energy Reviews, vol. 14, pp. 1695-1721, 2010.
4. K. Lovegrove and W. Stein, Concentrating solar power technology: Principles, developments and applications: Elsevier, 2012.
5. G. E. Cohen, D. W. Kearney, and G. J. Kolb, "Final report on the operation and maintenance improvement program for concentrating solar power plants," Usage listed is raw water usage and assumed to be withdrawal rate. Consumption rate approximated from, vol. 90, pp. 30-31, 1999.
6. J. K. Kaldellis, P. Fragos, and M. Kapsali, "Systematic experimental study of the pollution deposition impact on the energy yield of photovoltaic installations," Renewable Energy, vol. 36, pp. 2717-2724, 10// 2011.
7. D. Goossens and E. Van Kerschaever, "Aeolian dust deposition on photovoltaic solar cells: the effects of wind velocity and airborne dust concentration on cell performance," Solar Energy, vol. 66, pp. 277-289, 7// 1999.
8. A. M. El-Nashar, "Seasonal effect of dust deposition on a field of evacuated tube collectors on the performance of a solar desalination plant," Desalination, vol. 239, pp. 66-81, 4// 2009.
9. J. W. Strachan and R. M. Houser, "Testing and evaluation of large-area heliostats for solar thermal applications," Sandia National Laboratories, Albuquerque, NM, Report No. SAND92-1381, 1993.
10. B. J. Anglani F., Dekkers W., "CFD Modelling of a Water-Jet Cleaning Process for Concentrated Solar Therma CST Systems," Third Southern African Solar Energy Conference, 2015.
11. Z. Guan, S. Yu, and H. Gurgenci, "Collinsville solar thermal project: Solar mirror cleaning requirements," University of Queensland, Brisbane, Australia, 2014.
12. E. P. Roth and A. J. Anaya, "The Effect of Natural Soiling and Cleaning on the Size Distribution of Particles Deposited on Glass Mirrors," Journal of Solar Energy Engineering, vol. 102, pp. 248-256, 1980.
13. E. F. Cuddihy, "SURFACE SOILING: THEORETICAL MECHANISMS AND EVALUATION OF LOW-SOILING COATINGS," Williamsburg, VA, USA, 1983, pp. 379-396 JPPUD4.
14. E. S. Kumar, B. Sarkar, and D. Behera, "Soiling and dust impact on the efficiency and the maximum power point in the photovoltaic modules," in International Journal of Engineering Research and Technology, 2013.
