AUTONOMOUS SOLAR FIELD AND RECEIVER INSPECTIONS BASED ON POLARIMETRIC-ENHANCED IMAGING

Abstract

Provided herein are systems and related methods of performing solar field and receiver inspections based on polarimetric-enhanced imaging.

Description

TECHNICAL FIELD

This disclosure relates to apply polarimetric-enhanced imaging systems for efficient and accurate autonomous field inspections.

BACKGROUND

During typical operation of a concentrating solar power (CSP) plant, a large portion of the energy (˜45%) can be lost due to various imperfect conditions, such as blocking, shading, mirror soiling, tracking and canting errors, etc. Among factors that determine CSP collector instantaneous optical efficiency, the energy loss from non-perpendicular solar incidence on the heliostats and error in tracking the sun path during a day accounts for ˜20% optical loss. The mirror reflectance is an average 87.8% between washings due to soiling, much smaller than that of a clean mirror (as high as 93-96%). The spillage efficiency is often assumed high (−99%), but in practice the operations can be far less ideal and much larger than 1% loss occurs as a result of heliostat canting and alignment errors. Currently, the inspection of CSP collectors typically relies on conventional imaging techniques or human eyes. However, neither provides the required accuracy and speed, thus resulting in a large loss in optical efficiency due to tracking error, mirror soiling (˜5%), alignment error and defects (1% to 1.5%). Additionally, small cracks on solar absorber tubes and CSP plant piping can rapidly propagate and subsequently cause final failure. However, the current imaging methods do not provide sufficient contrast to identify small defects before propagation.

SUMMARY

This disclosure relates to systems and methods for polarization-based autonomous concentrating solar power (CSP) inspection that allow fast, high-contrast, and accurate inspection of collector fields and receivers. Embodiments combine intensity and polarization-based detection of heliostat edges and corners, thereby enabling fast and accurate detection of heliostat optical errors using drone-based techniques for autonomous inspections and improving solar field collector efficiency, in particular with regard to CSP plants where the canting, tracking, and aiming are less than ideal. The autonomous detection of soiling conditions and defects improves efficiency in scheduling heliostat washing and repair, and accordingly increases the solar-weighted heliostat reflectance and collector optical efficiency. Infrared polarimetric imaging, in combination with thermal and visible imaging, allows early detection of receiver defects before damage, thereby extending the receiver lifetime.

Advantages of the described embodiments include but are not limited to the following. Visible full-stokes polarimetric images provide higher contrast for mirror edges, corners, scratches, cracks, and soiling. Both visible and infrared (thermal) polarimetric images provide higher contrast for solar receiver tube cracks and scratches. Integration of polarimetric imagers onto drones allows realization of autonomous field inspection in CSP plants, instead of labor-intensive slow inspection carried out by humans.

Systems and methods described herein combine intensity and polarization information of optical images to realize high speed and accurate detection of heliostat edges and corners, which can be used for quick evaluation of heliostat optical errors based on optical models developed for heliostats. This is particularly applicable for CSP plants where the canting, tracking, and aiming is less than ideal and can ultimately reduce optical loss due to canting and tracking errors and increase energy efficiency up to >1% (best-effort performance). The autonomous detection of soiling conditions and defects can improve the efficiency in scheduling heliostat washing and repair, and as a result, can potentially increase the solar-weighted heliostat reflectance and the collector optical efficiency up to 4%. Visible and infrared polarimetric imaging, in combination with thermal and visible imaging, allow early detection of receiver defects before damage and thus extending the receiver lifetime. This will mark a new direction towards increasing the receiver lifetime towards at least a 40-year span.

The details of one or more embodiments of the subject matter of this disclosure are set forth in the accompanying drawings and the description. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.

In another aspect, the present disclosure provides an imaging system that includes a plurality of imaging devices configured for imaging visible and infrared wavelengths received from one or more components of a solar power plant. The system also includes a controller operably connected, or connectable, to the plurality of imaging devices, which controller comprises, or capable of accessing, computer readable media comprising non-transitory computer executable instructions which, when executed by at least one electronic processor, perform at least: receiving one or more visible and infrared images of the one or more components of the solar power plant from the plurality of imaging devices, and identifying one or more errors, defects, and/or soiling conditions of the one or more components of the solar power plant from the one or more visible and infrared images. In some embodiments, one or more of the plurality of imaging devices are mounted onto unmanned autonomous vehicles (UAVs). In some embodiments, the solar power plant comprises a concentrating solar power plant (CSP). In some embodiments, one or more of the plurality of imaging devices are mounted onto one or more power towers of the CSP. In some embodiments, one or more of the plurality of imaging devices comprise polarimetric imaging devices. In some embodiments, the polarimetric imaging devices comprise one or more full-stokes polarimetric imaging sensors. In some embodiments, one or more of the plurality of imaging devices comprise one or more integrated silicon metasurfaces with metallic nanowire polarizers. In some embodiments, one or more of the plurality of imaging devices comprise one or more CMOS image sensors. In some embodiments, the controller is configured to identify the one or more errors, defects, and/or soiling conditions in less than about 1 second of receiving the receiving the one or more visible and infrared images and with a probability of detection of more than about 0.95. In some embodiments, the controller is configured to receive different polarization parameters selected from the group consisting of: intensity, degree of linear polarization (DOLP), angle of polarization (AOP), Stokes parameters (S₁, S₂, S₃), and degree of polarization (DOP).

In another aspect, the present disclosure provides a method of conducting an autonomous solar power field inspection using a plurality of imaging devices. The method includes receiving one or more visible and infrared images of one or more components of the solar power plant from the plurality of imaging devices. The method also includes identifying one or more errors, defects, and/or soiling conditions of the one or more components of the solar power plant from the one or more visible and infrared images. In some embodiments, one or more of the plurality of imaging devices are mounted onto unmanned autonomous vehicles (UAVs).

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 depicts an example of a system for drone-based and polarimetric imaging for intelligent inspections of concentrating solar power (CSP).

FIGS. 2A-2D show aspects of polarization detection by an integrated silicon metasurface with metallic nanowire polarizers.

FIGS. 3A and 3B depict front and side views, respectively, of polarimetric imaging drones.

FIG. 4A-E shows an overview of a full-stokes polarimetric imaging sensor.

FIG. 5A shows an image of a mirror in the background of blue sky and clouds. The arrows indicate the scratches or dust particles on the mirror surface. FIGS. 5B and 5C show DOLP and AOP images of the mirror, respectively. FIGS. 5D and 5E show edge detection images obtained by standard algorithm from FIGS. 5A and 5B, respectively.

FIG. 6A shows mirror (5×5 cm²) image in blue sky background taken with a conventional camera. FIG. 6B shows the corresponding microscope images. FIG. 6C shows reflection spectra of mirror surfaces with four different soiling conditions. FIG. 6D show DOP and DOLP images taken by the full-stokes polarimetric imaging setup.

FIGS. 7A-7D show images related to polarimetric imaging of solar receiver tubes. FIG. 7A is an optical image taken by a conventional camera. FIG. 7B is an AOP image taken by a visible polarimetric imager. FIG. 7C is a thermal image of the object in FIG. 7D taken by a thermal polarimetric imager (scale bar: 1 cm).

DETAILED DESCRIPTION

This disclosure describes ultra-compact polarimetric imaging systems for efficient and accurate autonomous concentrating solar power (CSP) field inspections. Implementations include ultra-compact chip-integrated full-stokes polarimetric imaging devices for visible and infrared wavelengths. The polarimetric imagers can be mounted onto drones, i.e., unmanned autonomous vehicles (UAVs) or CSP power towers for continuous field monitoring and inspections. Applications of the ultra-compact polarimetric imaging systems include detection of edges and corners of heliostats and facets, monitoring of mirror soiling and quantitative reporting of solar-weighted reflectance, and inspection and monitoring of receivers for hot spots, tube defects, and paint degradation in visible and infrared ranges to provide temperature and polarization-based inspection.

FIG. 1 depicts an example of a system for drone-based and polarimetric imaging for intelligent inspections of concentrating solar power (CSP) plants. The system 100 depicted in FIG. 1 is configured to achieve >95% success rate of detecting heliostat edges and corners with a detection speed <1 second, increase optical efficiency of collector fields by >5% by improving spillage efficiency and mirror clearness, detect the peak flux of receiver tubes within 5% accuracy, identify defects before catastrophic damage, reduce maintenance cost, and improve inspection speed of CSP plants.

Combining drone-based techniques with intensity and polarization-based detection of heliostat edges and corners allows fast and accurate detection of heliostat optical errors, thereby improving solar field collector efficiency, which can be particularly important for CSP plants in which canting, tracking, and aiming are less than ideal. The autonomous detection of soiling conditions and defects improves the efficiency in scheduling heliostat washing and repair, and accordingly increases the solar-weighted heliostat reflectance and the collector optical efficiency. Infrared polarimetric imaging, in combination with thermal and visible imaging, allows early detection of receiver defects before damage, thereby extending receiver lifetime.

Systems including hybrid metasurface structures for ultra-compact circular polarization filters with large extinction ratio are described, in which rotation of the chip is not required. FIGS. 2A-2D show aspects of polarization detection by an integrated silicon metasurface with metallic nanowire polarizers. FIG. 2A depicts a metasurface structure integrated on a fused silica substrate. FIG. 2B shows scanning electron microscope (SEM) images of a fabricated silicon metasurface and nanograting array. FIG. 2C shows extinction ratios >30 of 12 pixels in the metasurface-based polarization filter array, measured with a filter centered around 630 nm with 50 nm bandwidth. Extinction ratios >30 have been demonstrated for visible light and transmission efficiency >60%. The metasurface-based CP filters have been integrated with nanograting-based LP filters onto a CMOS imager to realize a polarimetric imaging sensor (imaging speed >60 fps). FIG. 2D (top) shows an array imaged using a linear polarization incidence. FIG. 2D (bottom) shows a prototype polarimetric imaging array (32 by 32).

Chip-integrated full-stokes polarimetric imaging can be achieved via integrating rationally designed nanostructures directly onto CMOS image sensors. Device designs for IR cameras have been demonstrated. With a single shot, these cameras can obtain complete polarization information for each pixel, i.e., all four stokes parameters, and only simple mathematic operations are typically needed to generate polarimetric images for various parameters, such as DOP, DOLP, DOP, AOP, and the like.

Systems and methods of this disclosure include full-stokes polarimetric imaging systems with the same volume factor and weight as conventional imaging systems, which can be integrated onto drones (UAVs) or solar towers as conventional cameras without extra cost or efforts. These systems can be used to realize fast (<1 s) and accurate detection (probability of detection: >0.95) of mirror edges and corners to minimize mirror tracking and pointing errors, as well as autonomous detection of mirror defects and soiling conditions to assist system maintenance and minimize the optical loss >5%) in mirror reflection of heliostats. Solar receiver tubes can also be monitored in-situ to avoid failure due to defects formation.

FIGS. 3A and 3B depict front and side views, respectively, of polarimetric imaging drones. The polarimetric imaging drones include a UAV (Unmanned aerial vehicle), visible camera, polarimetric imaging camera, and camera gimbal for stabilization for good image quality. The polarimetric imaging sensor can be a full-stokes polarimetric imaging sensor or a commercially available sensor with similar functions (linear polarization camera by Sony or liquid crystal based polarimetric imaging sensor (e.g., Salsa), or the like).

FIGS. 4A-4E show an overview of a full-stokes polarimetric imaging sensor.

A conventional image with a standard camera (FIG. 5A) and polarimetric images with a polarimetric imaging system (FIGS. 5B and 5C) were taken of a mirror in the background of a blue sky and clouds. Before the test, a deep scratch and a shallow scratch and some dust particles were provided to the mirror surface, as indicated by the arrows on FIG. 5A. From the polarimetric image sensor, one can obtain multiple images showing different polarization parameters, such as intensity (same as conventional images), degree of linear polarization (DOLP, FIG. 5B), angle of polarization (AOP, FIG. 5C), Stokes parameters (S₁, S₂, S₃) and degree of polarization (DOP), etc. One can select the images with the best contrast for various purposes. In one example, it was found that the image of DOLP shows superior contrast than other images for defects and edge detection. Using a standard edge detection algorithm (Canny in Matlab), results from intensity image (FIG. 5D) show missing edges and corner (top left corner of the mirror, indicated by arrows) due to low contrast, while the DOLP image clearly displays all edges and corners (FIG. 5E). Additionally, its high contrast also makes it possible to identify the deep scratch and scatters introduced on the mirror surface. Interestingly, although the AOP image is not ideal for identification of edges, corners, and defects, it has a superior contrast to distinguish the sky, clouds and the mirror. This may serve to provide a good reference for cloud tracking purposes.

Preliminary studies were conducted to analyze the mirror soiling condition. A piece of mirror from a broken CSP collector was obtained, and four soiling conditions were generated by blowing sand onto four parts of the mirror with a hairdryer (FIG. 6A). Microscope images are shown in FIG. 6B. Reflection spectra of mirror surfaces with four different soiling conditions are shown in FIG. 6C. Analysis of the polarimetric imaging results revealed that the DOP image (FIG. 6D) shows the best contrast between clean mirror surface (Region 1, R=100%, average SOP=0.53) and dirty mirror surface (Region 3 and 4 with R=92%, 85%, average SOP=0.42, 0.47, respectively). The DOLP image (FIG. 6D) does not show as much contrast for Regions 3 and 4.

Solar receiver tubes were created by spraying black paint onto metallic tubes to achieve a black surface as shown in FIG. 7A. Conventional visible images of these tubes usually do not show strong contrast for cracks or scratches unless there is a strong color contrast. In comparison, the SOP image taken by a visible polarimetric camera shows strong contrast for all scratches indicated by arrows on the image (FIG. 7B), independent of color contrast. Comparison of polarimetric thermal images and conventional thermal images (FIGS. 7C and 7D, respectively) revealed that the polarimetric thermal images revealed cracks which were not visible in the conventional thermal images.

An autonomous monitoring system can include polarimetric imaging drones, algorithms for autonomous detection, and measurement protocol. Such autonomous monitoring systems allow more frequent and accurate inspection of collector field errors (mirror canting/tracking errors, mirror scratches, mirror cracks, etc), mirror soiling conditions, receiver tube defects (scratches, cracks, etc.), and facilitate more effective maintenance as needed. Thus, these systems can reduce the energy loss (due to collector field errors, mirror soiling) and major failure of receiver tubes in typical CSP plants, ultimately leading to higher energy efficiency (best effort performance >5%). The systems can be scaled up (with a larger number of drones) to provide regular inspection of Mega Watt CSP plants.

Preliminary studies verified that the proposed full-stokes polarimetric imaging can provide better image contrast than conventional imagers for detection of mirror edge, corner, scratches, cracks, soiling condition and receiver defects.

Polarimetric Imaging with High Contrast for Mirror Edges, Corners, Scratches and Cracks

A conventional image with a standard camera and polarimetric images of a mirror in the background of a blue sky and clouds were taken. Before the test, a deep scratch, a shallow scratch, and some dust particles were introduced on the mirror surface. From the polarimetric image sensor, one can obtain multiple images showing different polarization parameters, such as intensity (same as conventional images), degree of linear polarization (DOLP), angle of polarization (AOP), Stokes parameters (S₁, S₂, S₃) and degree of polarization (DOP), etc. One can select the images with the best contrast for various purposes. The image of DOLP showed superior contrast compared to other images for defects and edge detection. Using a standard edge detection algorithm (Canny in Matlab), results from intensity image were shown to miss edges and corners due to low contrast, while the DOLP image clearly displayed all edges and corners. Additionally, its high contrast also makes it possible to identify the deep scratch and scatters introduced on the mirror surface. Thus, the AOP image has a superior contrast to distinguish the sky, clouds and the mirror. This may serve to provide a good reference for cloud tracking purposes. The polarimetric images were taken using a commercial camera with a linear polarizer and a quarter waveplate, with mechanical rotation of the quarter waveplate.

Polarimetric Imaging for Mirror Soiling Condition Monitoring

Preliminary studies were performed to analyze mirror soiling condition. A piece of mirror from a broken CSP collector was provided. Four soiling conditions were generated by blowing sand onto four parts of the mirror with a hairdryer. Microscope images and reflection spectra of mirror surfaces with four different soiling conditions were taken, and polarimetric imaging results were analyzed. It was found that the DOP image shows the best contrast between clean mirror surface (Region 1, R=100%, average SOP=0.53) and dirty mirror surface (Region 3 and 4 with R=92%, 85%, average SOP=0.42, 0.47, respectively). The DOLP image does not show as much contrast for Region 3 and 4.

Polarimetric Images of Solar Receiver Tubes

Solar receiver tubes were prepared by spraying black paint onto metallic tubes to achieve a black surface. Conventional visible images of these tubes usually did not show strong contrast for cracks or scratches unless there is a strong color contrast. In comparison, the SOP image taken by a visible polarimetric camera showed strong contrast for all, independent of color contrast. Polarimetric thermal images and conventional thermal images were also compared, and it was found that the polarimetric thermal images revealed cracks which were not visible in conventional thermal images.

Although this disclosure contains many specific embodiment details, these should not be construed as limitations on the scope of the subject matter or on the scope of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in this disclosure in the context of separate embodiments can also be implemented, in combination, in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments, separately, or in any suitable sub-combination. Moreover, although previously described features may be described as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can, in some cases, be excised from the combination, and the claimed combination may be directed to a sub-combination or variation of a sub-combination.

Particular embodiments of the subject matter have been described. Other embodiments, alterations, and permutations of the described embodiments are within the scope of the following claims as will be apparent to those skilled in the art. While operations are depicted in the drawings or claims in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed (some operations may be considered optional), to achieve desirable results.

Accordingly, the previously described example embodiments do not define or constrain this disclosure. Other changes, substitutions, and alterations are also possible without departing from the spirit and scope of this disclosure.

Claims

1. An imaging system comprising: polarimetric imaging drones;algorithms for autonomous detection; andmeasurement protocols,wherein the imaging system is configured to: initiate optical inspection of a concentrating solar power plant;assess a need to perform maintenance on the concentrating solar power plant.

2. The imaging system of claim 1, wherein the polarimetric imaging drones comprise one or more polarimetric imaging devices for visible wavelengths, infrared wavelengths, or both.

3. The imaging system of claim 2, further comprising a controller.

4. The imaging system of claim 2, wherein the imaging device provides polarization-based detection.

5. The imaging system of claim 1, wherein the imaging system is configured to detect edges and corners of heliostats and facets of the concentrating solar power plant.

6. The imaging system of claim 1, wherein the imaging system is configured to monitor mirror soiling and solar-weighted reflectance of the concentrating solar power plant.

7. The imaging system of claim 1, wherein the imaging system is configured to inspect and monitor receivers of the concentrating solar power plant to detect hot spots, tube defects, and paint degradation.

8. The imaging system of claim 1, wherein the imaging system is coupled to a power tower.

9. A method of imaging a portion of a concentrating solar power plant, the method comprising focusing the polarimetric imaging drones of the imaging system of claim 1 on the portion of the concentrating solar power plant.

10. An imaging system comprising: a plurality of imaging devices configured for imaging visible and infrared wavelengths received from one or more components of a solar power plant; and,a controller operably connected, or connectable, to the plurality of imaging devices, which controller comprises, or capable of accessing, computer readable media comprising non-transitory computer executable instructions which, when executed by at least one electronic processor, perform at least: receiving one or more visible and infrared images of the one or more components of the solar power plant from the plurality of imaging devices; and,identifying one or more errors, defects, and/or soiling conditions of the one or more components of the solar power plant from the one or more visible and infrared images.

11. The imaging system of claim 10, wherein one or more of the plurality of imaging devices are mounted onto unmanned autonomous vehicles (UAVs).

12. The imaging system of claim 10, wherein the solar power plant comprises a concentrating solar power plant (CSP).

13. The imaging system of claim 12, wherein one or more of the plurality of imaging devices are mounted onto one or more power towers of the CSP.

14. The imaging system of claim 10, wherein one or more of the plurality of imaging devices comprise polarimetric imaging devices.

15. The imaging system of claim 10, wherein the polarimetric imaging devices comprise one or more full-stokes polarimetric imaging sensors.

16. The imaging system of claim 10, wherein one or more of the plurality of imaging devices comprise one or more integrated silicon metasurfaces with metallic nanowire polarizers.

17. The imaging system of claim 10, wherein one or more of the plurality of imaging devices comprise one or more CMOS image sensors.

18. The imaging system of claim 10, wherein the controller is configured to identify the one or more errors, defects, and/or soiling conditions in less than about 1 second of receiving the receiving the one or more visible and infrared images and with a probability of detection of more than about 0.95.

19. The imaging system of claim 10, wherein the controller is configured to receive different polarization parameters selected from the group consisting of: intensity, degree of linear polarization (DOLP), angle of polarization (AOP), Stokes parameters (S1, S2, S3), and degree of polarization (DOP).

20. A method of conducting an autonomous solar power field inspection using a plurality of imaging devices, the method comprising: receiving one or more visible and infrared images of one or more components of the solar power plant from the plurality of imaging devices; and,identifying one or more errors, defects, and/or soiling conditions of the one or more components of the solar power plant from the one or more visible and infrared images.