Reflector Soiling and Cleaning Methods Fabian Wolfertstetter, SFERA Summer School Hornberg, May 15th 16th 2013

Size: px

Start display at page:

Download "Reflector Soiling and Cleaning Methods Fabian Wolfertstetter, SFERA Summer School Hornberg, May 15th 16th 2013"

May Sparks
5 years ago
Views:

Reflector Soiling and Cleaning Methods Fabian Wolfertstetter, DLR @

1 Reflector Soiling and Cleaning Methods Fabian Wolfertstetter, SFERA Summer School Hornberg, May 15th 16th

2 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 2 Overview 1. Introduction 2. Measurement 3. Nature of dirt 1. Microscopic 2. Macroscopic 3. Weather influences 4. Mitigation techniques 1. Natural cleaning 2. Artificial cleaning 3. Preventing dirt 5. Cleaning strategies 6. Summary and outlook Literature review presented No claim to be complete No best practices available No standard procedures for measurement or cleaning established up to now

3 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 3 Nomenclature Soiling is the process of surfaces getting dirty during exposure to the environment Dirt is everything that settles on the surfaces, it includes: Aerosol Particles, composed of Salt particles (very small crystals) Mineral Dust (anorganic, from bedrock, travels very long distances) Biological matter like pollen, bacteria, algae Other biological matter like lichen, algae, bird droppings, etc Sand, e.g. in dust storms Cleanliness is the reduction of reflectivity of a mirror compared to its clean state. Values between 0 and 1 Cleaning sets cleanliness to 1 and reflectivity to some 90-95% depending on the mirror material

4 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 4 Introduction In general: very little CSP-specific research in the field Losses of 1% to 25% per week measured. High variation depending on location Scales at least linearly with reduction of CSP plant efficiency Highly underestimated efficiency limiting factor in CSP Most of research has been done for Photovoltaics Different optical properties: PV is affected in terms of transmission loss of cover glass (absorption), CSP optics are directional, i.e. also scattering plays a role CSP more sensitive to soiling than PV In practice constant reflectivity losses of e.g. 5% are assumed is this reasonable?!? PV CSP Glass mirror

5 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 5 Dirt Everyday-phenomenon Very complex and variable in its nature There is artistic solutions and there is serious solutions

6 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 6 Dirt basic principles 1. Settling of particles to the surface due to gravity 2. Adhesion with humidity/ Van der Waals/ electrostatics 3. Cementation with salt and other chemicals 4. Other: e.g. bird droppings, melting of dust on absorber windows or glass tubes 5. Washing of mirrors by rain (snow) Soiling

7 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 7 Measurement of soiling Several hand measurement devices for specular reflectivity available on the market: Devices & Services Reflectometer R15 Abengoa s Condor Reflectometer SOC 410 Solar FW2 For power plants, only punctual measurements available Research offers other non-standard-methods

8 Slide 7 FW2 nrel paper on comparison of different reflectometers Fabian Wolfertstetter;

SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 8 Devices and Services Reflectometer D & S reflectometer 15 is the reference in most publications Used as the industry standard

9 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 8 Devices and Services Reflectometer D & S reflectometer 15 is the reference in most publications Used as the industry standard Disadvantages: Punctual measurement points Time consuming Single wavelength Measurement results are operator dependent Advantage: Absolute reflectivity values measured directly in power plant [Devices and Services Manual]

SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 9 Other hand measurement devices Different models available for hand measurements in a CSP plant

10 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 9 Other hand measurement devices Different models available for hand measurements in a CSP plant Advantage: measurement directly in power plant Portable reflectometers from left to right: Condor (Abengoa), Devices and Services 15R, SOC 410 Solar [Crawford 2012]

11 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 10 Intercomparison of Reflectometers Optical properties: Devices & Services R15 Resolution and accuracy Accuracy not specified Wavelengths Acceptance angle 550nm or 660nm 7, 15, 25, 46 mrad Condor (Abengoa) ± 5% 435, 525, 650, 780, 940, 1050 nm 408 mrad Surface Optics Corporation 410 Solar ± , , , , , , nm Integrating sphere, diffuse and total

12 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 11 Intercomparison of Reflectometers Intercomparison not convincing [Crawford 2012]

13 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 12 Automatic device TraCS Automatic real time measurement of a test mirror with sun spectrum High time resolution allows for identification of weather influences on soiling Validated with hand measurement devices TraCS cleanliness Reference cleanliness cleanliness factor Jun Jun Jun Jun Jun Ju [Wolfertstetter, 2012]

14 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 13 Microscopic nature of Dirt Learn about the mechanisms of adhesion Origin of dust Optical properties of dust particles

15 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 14 Dirt seen at 50x magnification on a mirror What a beauty!

16 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 15 Different types of dirt seen in microscope 50x 100x a) c) b) d) Character of single particles: a) Saharan dust, compare [Avila 1997] b) Local dust Character of soiling patterns: c) Dried water droplet d) High salt content with water forming film

17 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 16 Dust is very different Sand Particles collected at different sites a) Sample from Libya b) Iraq c) Libya in SEM (~120 µm) d) Oman in SEM

18 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 17 Zoom in: Dirt on surfaces is different Left: sample collected from around NREL research facility Right: sample washed off mirrors exposed at same site Note: scales are the same [Nelson 2011]

19 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 18 Zoom closer with SEM Left: mirror soil Right: ground soil Note: many coagulated particles [Nelson 2011]

20 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 19 Optical system for size distribution determination Scanning microscopic method Size distributions detemined Compared to airborne dust [Biryukov 1999, 1996]

21 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 20 Optical effects of dirt Specular reflectance as a function of acceptance angle and dust accumulation Beam profile is slightly altered due to scattering Another result of Pettits study: Effect of particles on specular reflectance ~5 times greater than on hemispherical reflectance [Pettit 1980]

22 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 21 Composition of Dirt Dirt components compositions measured in many studies Conclusions to influence on soiling not drawn in sufficient manner [Nelson 2011]

23 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 22 Macroscopic The following are general observations, but again, there is very few publications investigating the macroscopic effects for CSP plant types: Dirt is very inhomogeneously distributed on a single mirror panel Back structure of mirrors plays a role Mirrors at the edges of a field soil differently than those in the middle

SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 23 Intra-field variation Bowman compared different soiling rates of CPV elements within the same

24 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 23 Intra-field variation Bowman compared different soiling rates of CPV elements within the same field High variations shown Comparison to pyrheliometer (NIP) behaviour: High winds speeds: CPVs soil faster than pyrheliometer Low wind speeds: pyrheliometer soils faster

25 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 24 Weather influence Very few publications Dependency on different mirror inclinations well documented for a few measurement sites Sandstorms in different regions

weekly measurements of sample mirrors to weather parameters

26 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 25 Weather influences Study from Abu Dhabi comparing weekly measurements of sample mirrors to weather parameters Best correlation with DHI, still far from being proven [Tahboub 2012]

27 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 26 Weather influences [Tahboub 2012]

28 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 27 Red Rain Atmospheric research treats red rains as light rains at high atmospheric aerosol concentrations Red rain event statistics For Southern France [Avila 1997]

29 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 28 Red Rain impact Measurement curve measured in Oujda, Morocco with TraCS instrument during a red rain event cleanliness factor Jun Jun Jun Jun Jun Ju [Wolfertstetter 2012]

30 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 29 Mitigation

31 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 30 Natural cleaning Rain (and snow) will clean mirrors only above a certain intensity Light rains can have the opposite effect Light rain washes out the suspended particles in the atmosphere and deposits them on the mirror surface If rain is heavy enough less dirt will stay on the mirrors Dew can wash the mirrors in some cases Dew can also glue smallest particles to surfaces Wind has been shown to have no cleaning effect

32 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 31 Cleaning systems Many different vehicles and methods for cleaning documented Studies on best parameters for cleaning Water pressure Water salt level Scrubbing vs spraying Some studies on automatic coulomb cleaning

33 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 32 Cleaning systems Traditional cleaning method, Mr Twister and Cleaning Truck at Cramer Junction power plant [Cohen 1999]

34 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 33 Cleaning systems Brush and pressurerized cleaning systems

35 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 34 Cleaning systems Paris automatic cleaning robot from Sener [Vicente 2012]

moving robot cleans line per line Novatec cleaning robot for Fresnel systems Runs

36 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 35 Cleaning systems More automized systems: Esolar cleaning robot Regular heliostat field configuration Continuously moving robot cleans line per line Novatec cleaning robot for Fresnel systems Runs automatically along linear concentrators Very little water consumption Nighttime cleaning [Schell 2011]

37 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 36 Coulomb force cleaning AC voltage induced on the mirror surface can help clean mirrors Particles are charged so they get pushed around by voltages Deaggregation effect (on the right) Energy balance?!? [Beryukov, 1999]

efficiency depends on water temperature and applied

38 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 37 Washing optimization Washing efficiency depends on water temperature and applied pressure in systems with no mechanical force [Fernández García 2011]

39 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 38 Washing optimization

40 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 39 Washing optimization Kattke and Vant Hull: Scrubbing is typically five times slower than spraying, but requires half the water Scrubbing removes dirt film, spraying only removes dust Dependency of costs on target reflectivity

41 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 40 Cleaning strategies Only relevant study supplies theoretical formulas for optimization Focus on dirt film and dirt particle removal [Kattke 2012]

42 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 41 Preventing dirt Nightly stow positions of heliostats and troughs avoid settling of dirt Anti-soiling coatings for solar mirrors Wind fences Paved roads, not dirt roads in power plants Vegetation, location, etc

43 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 42 Preventing dirt Anti soiling coating by Flabeg shows better behaviour than standard mirror surfaces

44 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 43 Summary PV literature not applicable to CSP Measurement techniques questionable Soiling is has very big effects Soiling is still poorly understood

45 SFERA Summer School, Hornberg, Germany, May 15 th 26 th 2013 Slide 44 Thank you for your attention Literature sources: Sarver, T., et al A comprehensive review of the impact of dust on the use of solar energy: history, investigations, results, literature and mitigation approaches, Renewable and Sustainable Energy Reviews 22 (2013) , 2013 Bowman, J., et al Analysis of soilingrates at the victor valley community college CPV site, AIP conf. Proc. 1407, 312 (2011); doi: / Wolfertstetter et al A novel method for automatic real-time monitoring of mirror soiling rates, SolarPACES conference proceedings, Marrakech, Morocco 2012 Schwarberg,F., Enhanced solar mirrors with anti-soiling coating, SolarPACES conference proceedings, Marrakech, Morocco 2012 Biryukov et al an optical system for the quantitative study of particulate contamination on solar collector surfaces, Solar Energy Vol 66, No 5, pp , 1999 Biryukov et al DEGRADATION OF OPTICAL PROPERTIES OF SOLAR COLLECTORS DUE TO THE AMBIENT DUST DEPOSITION AS A FUNCTION OF PARTICLE SIZE, Journal of Aerosol Sciences, Vol 27, pp 37-38, 1996 Kattke et al optimum target reflectivity for heliostat mirror washing, SolarPACES conference proceedings, Marrakech, Morocco 2012 Deffenbaugh et al THE EFFECT OF DUST ACCUMULATION ON LINE-FOCUS PARABOLIC TROUGH SOLAR COLLECTOR PERFORMANCE, Solar Energy Vol. 36 No, 2, pp, , 1986 Nelson et al UNDERSTANDING SOIL ADHESION IN CONCENTRATING SOLAR POWER PLANTS: A NOVEL ANALYSIS OF SOIL CHARACTERISTICS, SolarPACES conference proceedings, Granada, Spain, 2011 Crawford et al, A Comparison of Three Portable Reflectometers for Use in Operations and Maintenance of CSP Plants, SolarPACES conference proceedings, Marrakech, Morocco, 2012 Cohen et al, FINAL REPORT ON THE OPERATION AND MAINTENANCE IMPROVEMENT PROGRAM FOR CONCENTRATING SOLAR POWER PLANTS, SAND Unlimited Release, June 1999 Tahboub et al Solar Mirrors Soiling Campaign Abu Dhabi SolarPACES conference proceedings Marrakech, Morocco, 2012 Pettit et al WAVELENGTH DEPENDENT SCATTERING CAUSED BY DUST ACCUMULATION ON SOLAR MIRRORS Solar Energy Materials 3 (1980) pp 1-20 Schell et al Design and evaluation of esolar s heliostat fields, Solar Energy 85 (2011), pp Vicente et al PARIS A Parabolic Trough Autonomous Robotic Cleaning System, SolarPACES conference proceedings Marrakech, Morocco, 2012 Nelson et al, UNDERSTANDING SOIL ADHESION IN CONCENTRATING SOLAR POWER PLANTS: A NOVEL ANALYSIS OF SOIL CHARACTERISTICS, SolarPACES conference proceedings Granada, Spain, 2011

Trip Report SolarPACES September 2011

Trip Report SolarPACES September 2011 Introduction The 2011 SolarPACES conference was held in the town of Granada, in Spain, this year. The conference dealt with solar power and chemical energy systems.