Method for removing wind-induced error in heliostat spot verification and verification system

Abstract

The present application belongs to the technical field of heliostat light spot calibration, and particularly relates to a method for removing wind-induced error in heliostat light spot calibration and a calibration system. The calibration system comprises a heliostat (1), a calibration target (2) installed on a heat absorption tower, a positioning camera (3) installed on the heat absorption tower and arranged below the calibration target (2), a calibration camera (5) arranged on the ground, and a ground marker (4). During calibration, the heliostat (1) reflects sunlight onto the calibration target (2) to form a light spot, the calibration camera (5) photographs the light spot on the calibration target (2), and the positioning camera (3) photographs the corresponding marker on the ground to determine the offset of the calibration target (2). The method for removing wind-induced error in heliostat light spot calibration uses a camera installed on the heat absorption tower and a ground marker to evaluate the deviation of the target, and then combines target image region extraction to process the image, thereby removing the calibration error caused by wind.

Description

Technical Field

[0001] This invention belongs to the field of heliostat spot calibration technology, specifically relating to a method and calibration system for removing wind-induced errors in heliostat spot calibration. Background Technology

[0002] Tower solar thermal power generation systems utilize heliostats that track the sun in real time to reflect sunlight onto the receiver's surface, heating the heat transfer medium within the receiver and thus generating electricity. If the light is not accurately reflected onto the receiver, it will result in light concentration loss and uneven surface temperature distribution, potentially damaging the receiver. Therefore, heliostats need to be periodically calibrated to assess their accuracy. If a significant decrease in accuracy is detected, timely correction is necessary.

[0003] Currently, the most common method for heliostat calibration is to have the heliostat reflect light onto a designated square target, and then calibrate the heliostat based on the degree to which the center of the light spot deviates from the target center. The greater the deviation, the lower the accuracy. For example, the patent "CN102937814B-Dynamic calibration method and system for heliostat accuracy of tower solar thermal power generation system".

[0004] However, current methods do not account for calibration errors caused by wind-induced swaying of the absorber tower and heliostat. Generally, for current commercial heliostat fields, the absorber tower is approximately 200 meters high, and the target is located on top of the absorber tower at a height of about 180 meters. The target's sway due to wind can reach approximately 0.7 meters, and since the side length of a square target is about 20 meters, this swaying amplitude introduces an error of 3.5%. On the other hand, although the calibration camera is not installed very high, the image shift caused by camera sway is still significant. Therefore, calibration errors caused by wind should not be ignored. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for removing wind-induced errors in heliostat spot calibration. The method uses a camera installed on the heat-absorbing tower and ground markers to assess the target deviation, and then combines this with an image processing method that extracts the target image region to remove calibration errors caused by wind.

[0006] This invention is achieved through the following technical solution:

[0007] A calibration system for wind-induced errors in heliostat spot calibration, the calibration system comprising: a heliostat, a calibration target mounted on a heat-absorbing tower, a positioning camera mounted on the heat-absorbing tower and positioned below the calibration target, a calibration camera positioned on the ground, and ground markers;

[0008] During calibration, the heliostat reflects sunlight onto the calibration target to form a light spot;

[0009] The calibration camera captures the light spot on the calibration target;

[0010] The positioning camera captures images of corresponding markers on the ground to determine the offset of the calibration target.

[0011] Furthermore, the verification system includes 1-4 sets of verification units; each verification unit includes one verification target, one verification camera, one positioning camera, and one ground marker; the verification target is installed in the east, south, west, or north direction of the heat absorption tower; the positioning camera is installed below the verification target in the corresponding verification unit and is used to acquire images containing the corresponding ground marker; the ground marker is installed on the ground in front of the positioning camera in the corresponding verification unit; the verification camera is used to capture images of the verification target in the corresponding verification unit.

[0012] When the number of verification units is four;

[0013] The calibration targets include a first calibration target, a second calibration target, a third calibration target, and a fourth calibration target, respectively installed in the east, south, west, and north directions of the heat absorption tower; the first calibration target, the second calibration target, the third calibration target, and the fourth calibration target are squares with the same area;

[0014] The number of verification cameras is four, and the four verification cameras are set on the same horizontal plane; including a first verification camera, a second verification camera, a third verification camera and a fourth verification camera; the first verification camera, the second verification camera, the third verification camera and the fourth verification camera are respectively used to capture images containing the first verification target, the second verification target, the third verification target and the fourth verification target;

[0015] The number of positioning cameras is four, including a first positioning camera, a second positioning camera, a third positioning camera, and a fourth positioning camera respectively installed below the first verification target, the second verification target, the third verification target, and the fourth verification target;

[0016] The number of ground markers is four, including a first marker, a second marker, a third marker, and a fourth marker; the first marker, the second marker, the third marker, and the fourth marker correspond one-to-one with the first positioning camera, the second positioning camera, the third positioning camera, and the fourth positioning camera, and are respectively installed on the ground in front of each positioning camera;

[0017] The first positioning camera, the second positioning camera, the third positioning camera, and the fourth positioning camera are respectively used to acquire images containing the first marker, the second marker, the third marker, and the fourth marker.

[0018] A method for removing wind-induced errors in heliostat spot calibration, using the calibration system described above, the removal method comprising:

[0019] (1) Assess the target deviation: Calculate the pixel coordinates corresponding to the geometric center of the ground marker; calculate the deviation of the positioning camera; calculate the deviation of the verification target based on the deviation of the positioning camera;

[0020] (2) Remove the error caused by the deviation of the verification camera position: Remove the error caused by the deviation of the verification camera position by extracting the target image region;

[0021] (3) Determination of the center of the light spot: In the absence of a heliostat to project the light spot onto the calibration target, the background image of the calibration target is averaged; the heliostat is controlled to project the light spot of the reflected sunlight onto the calibration target, the pixel coordinates of the center of the light spot are obtained, and the relative coordinates of the center of the light spot on the calibration target are calculated.

[0022] (4) Normal distribution fitting of light spot shape: Perform normal distribution fitting of light spot center position and light spot brightness;

[0023] (5) Residual error removal: Calculate the relative coordinates of the spot center on the calibration target under the condition of no offset;

[0024] (6) Determine the beam quality; determine the beam shape and size based on the total energy ratio threshold of the beam.

[0025] Furthermore, step (1) specifically includes:

[0026] S1. Calculate the pixel coordinates corresponding to the geometric center of the ground marker:

[0027] In windless weather, each positioning camera captures images containing the corresponding markers, and the pixel coordinates corresponding to the geometric center of the markers are calculated:

[0028]

[0029]

[0030] Among them, u c v is the x-coordinate of the geometric center of the marker. c It is the ordinate of the geometric center of the marker; u i v is the x-coordinate of pixel i of the marker. i is the ordinate of pixel i of the marker, and n is the number of pixels of the marker;

[0031] S2. Calculation of camera deviation

[0032] In windless weather, the coordinates of the first positioning camera are (x... cam1 ,y cam1 ,z cam1 The attitude parameters of the first positioning camera are (α) cam1 ,β cam1 ,γ cam1 );α cam1 ,β cam1 ,γ cam1 These represent the angles of rotation of the first positioning camera around the x, y, and z axes, respectively; (these three axes are with the camera center as the origin, the east direction as the x-axis, the north direction as the y-axis, and the vertical direction as the z-axis); and the attitude parameters are less affected by wind, and can be considered to remain unchanged regardless of whether there is wind or not.

[0033] In windy weather, let the changes in the x and y coordinates of the first positioning camera be Δx and Δy, respectively; at this time, the coordinates of the first positioning camera are (x... cam1 +Δx,y cam1 +Δy,z cam1 );

[0034] The mirror field coordinates (x) of the geometric center point P1 of the first marker corresponding to the first positioning camera are as follows: m1 ,y m1 ,z m1 The conversion to camera coordinates is as follows:

[0035]

[0036] Among them, R cam It is a rotation matrix, represented as

[0037]

[0038] Then the pixel coordinates of the geometric center point P1 of the first marker in the first positioning camera are:

[0039]

[0040] Among them, u m1 v m1 These are the x and y coordinates of the pixel; f x f y These are the focal lengths in the x and y directions of the first positioning camera, respectively; c x and c y These are the pixel coordinates in the x and y directions corresponding to the center point of the image captured by the first positioning camera;

[0041] (x cam1,y cam1 ,z cam1 ), (α) cam1 ,β cam1 ,γ cam1 ) and (x m1 ,y m1 ,z m1 Calibration should be performed in windless weather.

[0042] Combined (x) cam1 ,y cam1 ,z cam1 ), (α) cam1 ,β cam1 ,γ cam1 ), (x m1 ,y m1 ,z m1 ) and (u m1 ,v m1 ) Perform calculations to determine the changes Δx and Δy in the x and y coordinates corresponding to the first positioning camera;

[0043] S3. Target deviation calculation:

[0044] The height of the first positioning camera is z. cam1 The height of the target center on the verification target is z. tar Using the deviation of the bullseye as the standard for target deviation, the target deviation is calculated as follows.

[0045]

[0046]

[0047] The changes in x and y coordinates Δx and Δy corresponding to the second, third, and fourth positioning cameras are calculated using the method described in step S2. The average of the changes in x and y coordinates corresponding to the four positioning cameras is taken as the final Δx and Δy to improve accuracy.

[0048] Furthermore, step (2) specifically involves:

[0049] The first verification camera captures a verification image containing the first verification target;

[0050] The coordinates of the first verification camera are (x cal1 ,y cal1 ,z cal1 ), attitude parameters are (α) cal1 ,β cal1 ,γ cal1 );α cal1 ,β cal1 ,γ cal1 These represent the angles of rotation of the first verification camera around the x, y, and z axes, respectively.

[0051] The center coordinates of the first verification target corresponding to the first verification camera are (x) tar1 ,y tar1 ,z tar The side length of the first verification target is L; in the coordinate system of the first verification camera, the center coordinates of the first verification target are expressed as...

[0052]

[0053] Among them, R cal It is a rotation matrix, represented as

[0054]

[0055] The center of the first verification target has pixel coordinates in the verification image captured by the first verification camera as follows:

[0056]

[0057] Among them, (u tar1 ,v tar1 ) is the pixel coordinate of the center of the first verification target in the verification image captured by the first verification camera; f cal,x f cal,y These are the focal lengths in the x and y coordinate directions of the first verification camera, respectively; c cal,x and c cal,y These are the pixel coordinates in the x and y directions corresponding to the center point of the image captured by the first verification camera;

[0058] The pixel length of the side length L of the first verification target in the verification image is:

[0059]

[0060] The pixel coordinates of the four vertices of the first verification target in the verification image are (u tar1 -0.5a,v tar1 +0.5a), (u tar1 +0.5a,v tar1 +0.5a), (u tar1 +0.5a,v tar1 -0.5a), (u tar1 -0.5a,v tar1 -0.5a); thus determining the four sides of the first verification target;

[0061] In the case of no offset on the first verification target, the pixel coordinates of the four vertices are calculated. As described earlier, the relationship between the pixel coordinates of the four vertices of the target and the offset is as follows: when the offset is 0, that is, when there is no wind, the pixel coordinates of the four vertices can be calculated directly.

[0062] When there is wind, straight line recognition is performed within the vicinity of the first verification target location in the image. The straight line recognition algorithm calculates four straight line segments that constitute the four edges of the first verification target in the image. The target image region is extracted, and the coordinates of the four vertices of the first verification target in the image are determined by the intersection points of the four straight line segments: (u tar1,vex,1 ,v tar1,vex,1 ), (u tar1,vex,2 ,v tar1,vex,2 ), (u tar1,vex,3 ,v tar1,vex,3 ), (u tar1,vex,4 ,v tar1,vex,4 );

[0063] Therefore, the target center can be determined as:

[0064]

[0065] Furthermore, step (3) specifically involves:

[0066] Without a heliostat projecting a light spot, a short-term background image of the calibration target is acquired. The acquired target background image is then averaged, and the brightness of each pixel is as follows:

[0067]

[0068] In the formula, These are the averaged target background image pixels; T represents the number of short-time background image acquisitions of the test target, t represents the image number, and j and i represent the horizontal and vertical coordinates of the pixel, respectively; p t,i,j This represents the brightness of the pixel at position i,j in the image with sequence number t.

[0069] The designated heliostat projects the reflected sunlight onto the calibration target, and then a target image is acquired, taken simultaneously with the image obtained from step S2 when the positioning camera offset is calculated; the pixel brightness of the target image is recorded as... After performing frame difference processing, we can obtain:

[0070]

[0071] In the formula, It is the brightness of the corresponding pixel after frame difference processing;

[0072] Thresholding is performed on the frame difference image, with the threshold σ ranging from 20 to 80; the specific value can be adjusted according to the implementation effect, with 50 being preferred.

[0073]

[0074] In the formula, This represents the brightness of pixel (i,j) in the image after thresholding.

[0075] The pixel coordinates of the light spot center are calculated using the following formula:

[0076]

[0077]

[0078] In the formula, x c and y c The pixel coordinates represent the center of the light spot, j and i represent the horizontal and vertical coordinates of the pixel, respectively, and N and M represent the resolution of the image in the horizontal and vertical directions, respectively.

[0079] The relative coordinates of the spot center on the target are calculated as follows:

[0080]

[0081] In the formula, (x' c ,y' c ) represents the relative coordinates of the center of the light spot on the target.

[0082] Furthermore, step (4) specifically involves:

[0083] Fit the normal distribution of the light spot center position and the light spot brightness:

[0084]

[0085] Where (x,y) represents the coordinates of each point on the calibration target plane, f(x,y) represents the relative brightness at point (x,y); σ1 represents the standard deviation of the relative brightness in the x direction, σ2 represents the standard deviation of the relative brightness in the y direction, ρ represents the correlation coefficient between x and y; and (μ1,μ2) is the center point of the two-dimensional normal distribution.

[0086] Furthermore, step (5) specifically involves:

[0087] In the world coordinate system of the mirror field, the positive x-axis points due east, and the positive y-axis points due north; the normal directions of the target surfaces of the first, second, third, and fourth verification targets point to the east, south, west, and north, respectively; the height offset of the heat absorption tower from the target center is Δx. tar and Δy tar ;

[0088] If the normal direction of the first verification target points eastward, then the target center offset is Δy. tar Assume Δy tar A value greater than 0 is represented in the image captured by the first verification camera as the target surface shifting to the right, and the center of the light spot shifting to the left relative to the target center. The corresponding shift amount of the light spot center is:

[0089]

[0090] In the formula, Δu1 represents the lateral offset of the pixel coordinates corresponding to the center of the first verification target spot;

[0091] The corresponding relative coordinates of the spot center on the first calibration target under the no-offset condition are:

[0092]

[0093] x" c y" c That is, the relative coordinates of the corrected spot center on the target; for Δy tar If <0, the above two equations also hold true.

[0094] The normal direction of the second calibration target points south, and the target center offset of the second calibration target is Δx. tar Assume Δx tar >0, in the image captured by the second verification camera, this is manifested as the target surface shifting to the right, and the center of the light spot shifting to the left relative to the target center; then

[0095]

[0096] In the formula, Δu2 represents the lateral offset of the pixel coordinates corresponding to the center of the second verification target spot;

[0097] The relative coordinates of the corrected spot center on the second calibration target are:

[0098]

[0099] The normal direction of the third calibration target points westward, and the target center offset of the third calibration target is Δy. tar Assume Δy tar If the value is >0, in the image captured by the third verification camera, this manifests as the target surface shifting to the left, while the center of the light spot shifts to the right relative to the target center; therefore...

[0100]

[0101] In the formula, Δu3 represents the lateral offset of the pixel coordinates corresponding to the center of the third verification target spot;

[0102] The relative coordinates of the corrected spot center on the target are:

[0103]

[0104] The normal direction of the fourth calibration target points north, and the target center offset of the fourth calibration target is Δx. tar Assume Δx tarIf the value is greater than 0, then in the image captured by the fourth verification camera, the target surface will be shifted to the left, and the center of the light spot will be shifted to the right relative to the target center.

[0105]

[0106] In the formula, Δu4 represents the lateral offset of the pixel coordinates corresponding to the center of the fourth verification target spot;

[0107] The relative coordinates of the corrected spot center on the target are:

[0108]

[0109] For cases where the target normal direction is not directly pointing to the four cardinal directions (east, south, west, north), let τ be the counterclockwise angle between the target normal direction and the world coordinate x-direction, and let the target center offset on the image be...

[0110]

[0111] The corresponding relative coordinates of the spot center on the target under the no-offset condition are:

[0112]

[0113] x" c y" c x' represents the relative coordinates of the corrected spot center on the target; c y' c This represents the relative coordinates of the center of the light spot on the target before correction.

[0114] Furthermore, step (6) specifically involves:

[0115] The shape and size of the light spot are determined based on the threshold value q of the total energy of the light spot.

[0116]

[0117] The left side of the above equation represents the integration of f(x,y); where c is a positive number, [-cσ1,cσ1] is the integration range in the x-direction, and [-cσ2,cσ2] is the integration range in the y-direction; when c→∞, the total integral →1; q on the right side represents the threshold of the total energy ratio of the light spot, which is a constant.

[0118] The magnitude of the offset of the corrected spot center relative to the target center corresponds to the pointing angle error, i.e.

[0119]

[0120] Where R is the distance from the center of the heliostat to the target center; L represents the side length of the calibration target; and a represents the pixel length of the calibration target in the calibration image.

[0121] Furthermore, the total energy ratio threshold q is selected to be between 0.8 and 1, preferably 0.97.

[0122] Beneficial technical effects of the present invention:

[0123] The method for removing wind-induced errors in heliostat spot calibration provided by this invention determines the degree of deviation of the heat-absorbing tower by using a positioning camera on the heat-absorbing tower and ground markers to remove the spot calibration error caused by the wind-induced displacement of the heat-absorbing tower; it removes the spot calibration error caused by the displacement of the calibration camera by extracting the position of the target and the relative coordinates of the spot center on the target; and the errors caused by the displacement of the heat-absorbing tower and the displacement of the calibration camera due to other factors can also be removed by this method. Attached Figure Description

[0124] Figure 1 This is a flowchart of a method for removing wind-induced errors in heliostat spot calibration according to an embodiment of the present invention;

[0125] Figure 2 This is a schematic diagram of a heliostat spot calibration system for wind-induced errors in an embodiment of the present invention.

[0126] The attached diagram is labeled as follows: 1. Heliostat; 2. Calibration target; 3. Positioning camera; 4. Ground marker; 5. Calibration camera; 6. Heat absorption tower. Detailed Implementation

[0127] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0128] Conversely, this invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of the invention as defined in the claims. Furthermore, to provide a better understanding of the invention, certain specific details are described in detail below. However, those skilled in the art will fully understand the invention even without these detailed descriptions.

[0129] This invention provides an embodiment of a calibration system for wind-induced errors in heliostat spot calibration, such as... Figure 2 As shown, the calibration system includes: a heliostat 1, a calibration target 2 installed on the heat absorber tower, a positioning camera 3 installed on the heat absorber tower and located below the calibration target 2, a calibration camera 5 located on the ground, and ground markers 4.

[0130] During calibration, the heliostat 1 reflects sunlight onto the calibration target 2 to form a light spot;

[0131] The calibration camera 5 captures the light spot on the calibration target 2;

[0132] The positioning camera 3 captures images of corresponding markers on the ground to determine the offset of the verification target 2.

[0133] The verification system includes 1-4 sets of verification units; each verification unit includes a verification target 2, a verification camera 5, a positioning camera 3, and a ground marker 4; the verification target 2 is installed in the east, south, west, or north direction of the heat absorption tower; the positioning camera 3 is installed below the verification target 2 in the corresponding verification unit and is used to acquire images containing the corresponding ground marker; the ground marker 4 is installed on the ground in front of the positioning camera 3 in the corresponding verification unit; the verification camera 5 is used to capture images of the verification target in the corresponding verification unit.

[0134] In this embodiment, when the number of verification units is four;

[0135] The calibration target 2 includes a first calibration target, a second calibration target, a third calibration target, and a fourth calibration target respectively installed in the east, south, west, and north directions of the heat absorption tower; the first calibration target, the second calibration target, the third calibration target, and the fourth calibration target are squares with the same area;

[0136] The number of verification cameras 5 is four, and the four verification cameras are set on the same horizontal plane; including a first verification camera, a second verification camera, a third verification camera and a fourth verification camera; the first verification camera, the second verification camera, the third verification camera and the fourth verification camera are respectively used to capture images containing the first verification target, the second verification target, the third verification target and the fourth verification target;

[0137] The number of positioning cameras 3 is four, including a first positioning camera, a second positioning camera, a third positioning camera and a fourth positioning camera respectively installed below the first verification target, the second verification target, the third verification target and the fourth verification target;

[0138] The number of ground markers 4 is four, including a first marker, a second marker, a third marker, and a fourth marker; the first marker, the second marker, the third marker, and the fourth marker correspond one-to-one with the first positioning camera, the second positioning camera, the third positioning camera, and the fourth positioning camera, and are respectively installed on the ground in front of each positioning camera;

[0139] The first positioning camera, the second positioning camera, the third positioning camera, and the fourth positioning camera are respectively used to acquire images containing the first marker, the second marker, the third marker, and the fourth marker.

[0140] This invention also provides an embodiment of a method for removing wind-induced errors in heliostat spot calibration, applicable to the removal of wind-induced heliostat calibration errors. The method includes:

[0141] (1) Assess the target deviation: Calculate the pixel coordinates corresponding to the geometric center of the ground marker; calculate the deviation of the positioning camera; calculate the deviation of the verification target based on the deviation of the positioning camera;

[0142] (2) Remove the error caused by the deviation of the verification camera position: Remove the error caused by the deviation of the verification camera position by extracting the target image region;

[0143] (3) Determination of the center of the light spot: In the absence of a heliostat to project the light spot onto the calibration target, the background image of the calibration target is averaged; the heliostat is controlled to project the light spot of the reflected sunlight onto the calibration target, the pixel coordinates of the center of the light spot are obtained, and the relative coordinates of the center of the light spot on the calibration target are calculated.

[0144] (4) Normal distribution fitting of light spot shape: Perform normal distribution fitting of light spot center position and light spot brightness;

[0145] (5) Residual error removal: Calculate the relative coordinates of the spot center on the calibration target under the condition of no offset;

[0146] (6) Determine the beam quality; determine the beam shape and size based on the total energy ratio threshold of the beam.

[0147] In this embodiment, step (1) specifically includes:

[0148] S1. Calculate the pixel coordinates corresponding to the geometric center of the ground marker:

[0149] To assess the degree of target deviation, cameras are installed on the heat-absorbing towers below the targets, shooting at an angle downwards. Markers are installed on the ground or existing buildings within the viewing field are used as markers. One camera is installed on the heat-absorbing tower below each target face, and a corresponding marker is installed on the ground in front of each camera. Generally, there are four target faces, corresponding to four cameras and four markers (see...). Figure 2 ).

[0150] In windless weather, each positioning camera captures images containing the corresponding markers, and the pixel coordinates corresponding to the geometric center of the markers are calculated:

[0151]

[0152]

[0153] Among them, u c v is the x-coordinate of the geometric center of the marker. c It is the ordinate of the geometric center of the marker; u i v is the x-coordinate of pixel i of the marker. i is the ordinate of pixel i of the marker, and n is the number of pixels of the marker;

[0154] Each positioning camera corresponds to one marker. The geometric centers of the pixel coordinates of the first, second, third, and fourth markers are denoted as (u... m1 ,v m1 ), (u m2 ,v m2 ), (u m3 ,v m3 ), (u m4 ,v m4 );

[0155] S2. Calculation of camera deviation

[0156] In windless weather, the coordinates of the first positioning camera are (x... cam1 ,y cam1 ,z cam1 The attitude parameters of the first positioning camera are (α) cam1 ,β cam1 ,γ cam1 );α cam1 ,β cam1 ,γ cam1 These represent the angles of rotation of the first positioning camera around the x, y, and z axes, respectively. These three axes are with the camera center as the origin, the x-axis in the east direction, the y-axis in the north direction, and the z-axis in the vertical direction. Furthermore, the attitude parameters are less affected by wind and can be considered to remain unchanged regardless of whether there is wind or not.

[0157] In windy weather, the first positioning camera oscillates with the heat absorption tower, causing significant changes in its x and y coordinates, while the change in its z coordinate is negligible. Let the changes in the x and y coordinates of the first positioning camera be Δx and Δy, respectively; at this time, the coordinates of the first positioning camera are (x... cam1 +Δx,y cam1 +Δy,z cam1 );

[0158] The mirror field coordinates (x) of the geometric center point P1 of the first marker corresponding to the first positioning camera are as follows: m1 ,y m1 ,z m1 The conversion to camera coordinates is as follows:

[0159]

[0160] Among them, R cam It is a rotation matrix, represented as

[0161]

[0162] Then the pixel coordinates of the geometric center point P1 of the first marker in the first positioning camera are:

[0163]

[0164] Among them, (u m1 ,v m1 ) is the pixel coordinate of the geometric center point P1 of the first marker in the verification image captured by the first verification camera; f x f y These are the focal lengths in the x and y directions of the first positioning camera, typically f. x =f y ;c x and c y These are the pixel coordinates in the x and y directions corresponding to the center point of the image captured by the first positioning camera.

[0165] (x cam1 ,y cam1 ,z cam1 ), (α) cam1 ,β cam1 ,γ cam1 ) and (x m1 ,y m1 ,z m1 Calibration should be performed in windless weather.

[0166] Combined (x) cam1 ,y cam1 ,z cam1 ), (α) cam1 ,β cam1 ,γ cam1 ), (x m1 ,y m1 ,z m1 ) and (u m1 ,v m1 ) Perform calculations to determine the changes Δx and Δy in the x and y coordinates corresponding to the first positioning camera;

[0167] S3. Target deviation calculation:

[0168] The height of the first positioning camera is z. cam1 The height of the target center on the verification target is z. tar Using the deviation of the bullseye as the standard for target deviation, the target deviation is calculated as follows.

[0169]

[0170]

[0171] In this embodiment, the method in step S2 above is used to calculate the changes in the x and y coordinates of the second positioning camera, the third positioning camera, and the fourth positioning camera respectively.

[0172] Because the other three positioning cameras (the second, third, and fourth positioning cameras) are at the same height as the first positioning camera, these four parameters (f) of the other three positioning cameras (the second, third, and fourth positioning cameras) x f y c x and c y The x and y coordinate changes are the same as those of the first positioning camera. In order to improve accuracy, the other three sets of cameras and corresponding markers are solved in the same way. The average of the x and y coordinate changes of the four positioning cameras is taken as the final x and y coordinate changes of the first positioning camera to improve accuracy.

[0173] In this embodiment, step (2) specifically includes:

[0174] The calibration camera may also deviate in position, but this deviation only affects the position of the target image region within the overall image. Therefore, by simply extracting the target image region, the error caused by the deviation in the calibration camera's position can be eliminated.

[0175] The first verification camera captures a verification image containing the first verification target;

[0176] The coordinates of the first verification camera are (x cal1 ,y cal1 ,z cal1 ), attitude parameters are (α) cal1 ,β cal1 ,γ cal1 );α cal1 ,β cal1 ,γ cal1 These represent the angles of rotation of the first verification camera around the x, y, and z axes, respectively.

[0177] The center coordinates of the first verification target corresponding to the first verification camera are (x) tar1 ,y tar1 ,z tar The side length of the first verification target is L; in the coordinate system of the first verification camera, the center coordinates of the first verification target are expressed as...

[0178]

[0179] Among them, R cal It is a rotation matrix, represented as

[0180]

[0181] The pixel coordinates of the center of the first verification target in the image are:

[0182]

[0183] Among them, (u tar1 ,v tar1 ) is the pixel coordinate of the center of the first verification target in the verification image captured by the first verification camera; f cal,x f cal,y These are the focal lengths in the x and y coordinate directions of the first verification camera, respectively; c cal,x and c cal,y The first verification camera uses the pixel coordinates in the x and y directions corresponding to the center point of the image captured by the first camera; the other three verification cameras use these four parameters (f... cal,x f cal,y c cal,x and c cal,y It is the same as the first verification camera.

[0184] The pixel length of the side of the first verification target in the verification image is:

[0185]

[0186] The pixel coordinates of the four vertices of the first verification target in the verification image are (u tar1 -0.5a,v tar1 +0.5a), (u tar1 +0.5a,v tar1 +0.5a), (u tar1 +0.5a,v tar1 -0.5a), (u tar1 -0.5a,v tar1 -0.5a); thus determining the four sides of the first verification target;

[0187] When the first verification target has no offset, Δx = Δy = 0, and the pixel coordinates of the four vertices can be calculated. As previously described, the relationship between the pixel coordinates of the four vertices and the offset has been explained. Therefore, when the offset is 0, i.e., in the absence of wind, the pixel coordinates of the four vertices can be directly calculated. When there is wind, the four sides of the first verification target deviate from their original positions in the verification image, but will remain near their normal positions. Therefore, straight-line recognition can be performed over a relatively large area near the target position in the image, such as in (u... tar1 -0.75a,v tar1+0.75a), (u tar1 +0.75a,v tar1 +0.75a), (u tar1 +0.75a,v tar1 -0.75a), (u tar1 -0.75a,v tar1 -0.75a) Within the square formed by the four points. There are many mature image processing algorithms for line recognition, such as the Canny algorithm, so they will not be elaborated upon here.

[0188] When there is wind, straight line recognition is performed within the vicinity of the first verification target location in the image. The straight line recognition algorithm calculates four straight line segments that constitute the four edges of the first verification target in the image. The target image region is extracted, meaning the four vertices of the first verification target in the image are determined by the intersection points of the four straight line segments: (u tar1,vex,1 ,v tar1,vex,1 ), (u tar1,vex,2 ,v tar1,vex,2 ), (u tar1,vex,3 ,v tar1,vex,3 ), (u tar1,vex,4 ,v tar1,vex,4 Therefore, the target center can be determined as:

[0189]

[0190] In this embodiment, step (3) specifically includes:

[0191] Without a heliostat projecting a light spot, a short-term (e.g., 10 seconds) background image of the calibration target (Lambertian target) is acquired; the acquired target background image is then averaged, and the brightness of each pixel is:

[0192]

[0193] In the formula, This represents the average pixel brightness of the target background image; T represents the number of short-time background image acquisitions of the test target, t represents the image number, and j and i represent the horizontal and vertical coordinates of the pixel, respectively; p t,i,j This represents the brightness of the pixel at position i,j in the image with sequence number t.

[0194] The designated heliostat projects the reflected sunlight onto the calibration target, and then a target image is acquired, taken simultaneously with the image obtained from step S2 when the positioning camera offset is calculated; the pixel brightness of the target image is recorded as... After performing frame difference processing, we can obtain:

[0195]

[0196] In the formula, It is the brightness of the corresponding pixel after frame difference processing;

[0197] Thresholding is performed on the frame difference image, with the threshold σ ranging from 20 to 80; the specific value can be adjusted according to the implementation effect, with 50 being preferred.

[0198]

[0199] In the formula, This represents the brightness of pixel (i,j) in the image after thresholding.

[0200] The pixel coordinates of the light spot center are calculated using the following formula:

[0201]

[0202]

[0203] In the formula, x c and y c The pixel coordinates represent the center of the light spot, j and i represent the horizontal and vertical coordinates of the pixel, respectively, and N and M represent the resolution of the image in the horizontal and vertical directions, respectively.

[0204] The relative coordinates of the spot center on the target are calculated as follows:

[0205]

[0206] In the formula, (x' c ,y' c ) represents the relative coordinates of the center of the light spot on the target.

[0207] In this embodiment, step (4) specifically includes:

[0208] Fit the normal distribution of the light spot center position and the light spot brightness:

[0209]

[0210] Where (x,y) represents the coordinates of each point on the calibration target plane, f(x,y) represents the relative brightness at point (x,y); σ1 represents the standard deviation of the relative brightness in the x direction, σ2 represents the standard deviation of the relative brightness in the y direction, ρ represents the correlation coefficient between x and y; and (μ1,μ2) is the center point of the two-dimensional normal distribution.

[0211] In this embodiment, step (5) specifically includes:

[0212] In the world coordinate system of the mirror field, the positive x-axis points due east, and the positive y-axis points due north; the normal directions of the target surfaces of the first, second, third, and fourth verification targets point to the east, south, west, and north, respectively; the height offset of the heat absorption tower from the target center is Δx. tar and Δy tar ;

[0213] If the normal direction of the first verification target points eastward, then the target center offset is Δy. tar Assume Δy tar A value >0, in the image captured by the first verification camera, is manifested as the target surface shifting to the right, and the center of the light spot shifting to the left relative to the target center. The corresponding shift amount is...

[0214]

[0215] In the formula, Δu1 represents the lateral offset of the pixel coordinates corresponding to the center of the first verification target spot;

[0216] The corresponding relative coordinates of the spot center on the first calibration target under the no-offset condition are:

[0217]

[0218] x" c y" c That is, the relative coordinates of the corrected spot center on the target; for Δy tar If <0, the above two equations also hold true.

[0219] The normal direction of the second calibration target points south, and the target center offset of the second calibration target is Δx. tar Assume Δx tar >0, in the image captured by the second verification camera, this is manifested as the target surface shifting to the right, and the center of the light spot shifting to the left relative to the target center; then

[0220]

[0221] In the formula, Δu2 represents the lateral offset of the pixel coordinates corresponding to the center of the second verification target spot;

[0222] The relative coordinates of the corrected spot center on the second calibration target are:

[0223]

[0224] The normal direction of the third calibration target points westward, and the target center offset of the third calibration target is Δy. tar Assume Δy tar If the value is >0, in the image captured by the third verification camera, this manifests as the target surface shifting to the left, while the center of the light spot shifts to the right relative to the target center; therefore...

[0225]

[0226] In the formula, Δu3 represents the lateral offset of the pixel coordinates corresponding to the center of the third verification target spot;

[0227] The relative coordinates of the corrected spot center on the target are:

[0228]

[0229] The normal direction of the fourth calibration target points north, and the target center offset of the fourth calibration target is Δx. tar Assume Δx tar If the value is greater than 0, then in the image captured by the fourth verification camera, the target surface will be shifted to the left, and the center of the light spot will be shifted to the right relative to the target center.

[0230]

[0231] In the formula, Δu4 represents the lateral offset of the pixel coordinates corresponding to the center of the fourth verification target spot;

[0232] The relative coordinates of the corrected spot center on the target are:

[0233]

[0234]

[0235] For cases where the target normal direction is not directly pointing to the four cardinal directions (east, south, west, north), let τ be the counterclockwise angle between the target normal direction and the world coordinate x-direction, and let the target center offset on the image be...

[0236]

[0237] The corresponding relative coordinates of the spot center on the target under the no-offset condition are:

[0238]

[0239] x" c y" c x' represents the relative coordinates of the corrected spot center on the target; c y' c This represents the relative coordinates of the center of the light spot on the target before correction.

[0240] In this embodiment, step (6) specifically includes:

[0241] The shape and size of the light spot are determined based on the threshold value q of the total energy of the light spot.

[0242]

[0243] The left side of the above equation represents the integration of f(x,y); where c is a positive number, [-cσ1,cσ1] is the integration range in the x-direction, and [-cσ2,cσ2] is the integration range in the y-direction; when c→∞, the total integral →1; q on the right side represents the threshold of the total energy ratio of the light spot, which is a constant.

[0244] By choosing an appropriate value of c to make the above formula true, the shape and size of the light spot can be determined, which corresponds to the beam quality; the total energy ratio threshold q can be a number close to 1 (choose a value between 0.8 and 1), preferably 0.97.

[0245] The magnitude of the offset of the corrected spot center relative to the target center corresponds to the pointing angle error, i.e.

[0246]

[0247] Where R is the distance from the center of the heliostat to the target center; L represents the side length of the calibration target; and a represents the pixel length of the calibration target in the calibration image.

[0248] The method for removing wind-induced errors in heliostat spot calibration provided by this invention uses a camera installed on the heat-absorbing tower and ground markings to assess the degree of target deviation, and then combines it with target image region extraction for processing, in order to remove calibration errors caused by wind.

[0249] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for removing wind-induced errors in heliostat spot calibration, characterized in that, The removal method includes: (1) Assess the degree of target deviation: Calculate the pixel coordinates corresponding to the geometric center of the ground marker; calculate the degree of deviation of the positioning camera; calculate the deviation of the verification target based on the degree of deviation of the positioning camera; Step (1) specifically includes: S1. Calculate the pixel coordinates corresponding to the geometric center of the ground marker: In windless weather, each positioning camera captures images containing the corresponding markers, and the pixel coordinates corresponding to the geometric center of the markers are calculated: ； ； in, It is the x-coordinate of the geometric center of the marker. It is the ordinate of the geometric center of the marker; It is the x-coordinate of pixel i of the marker. is the ordinate of pixel i of the marker, and n is the number of pixels of the marker; S2. Calculation of camera deviation: In windless weather, the coordinates of the first positioning camera are: The attitude parameters of the first positioning camera are: ; These represent the angles of rotation of the first positioning camera around the x, y, and z axes, respectively. In windy weather, let the changes in the x and y coordinates of the first positioning camera be respectively... and At this time, the coordinates of the first positioning camera are: ; The field coordinates of the geometric center point P1 of the first marker corresponding to the first positioning camera Transformed to camera coordinates: ； in, It is a rotation matrix, represented as: ； Then the pixel coordinates of the geometric center point P1 of the first marker in the first positioning camera are: ； in, These are the horizontal and vertical coordinates of the pixel; These are the focal lengths in the x and y directions of the first positioning camera, respectively. and These are the pixel coordinates in the x and y directions corresponding to the center point of the image captured by the first positioning camera; , and Calibration should be performed in windless weather. United , , and Calculations are performed to determine the changes in the x and y coordinates corresponding to the first positioning camera. and ; S3. Target deviation calculation: The height of the first positioning camera is The height of the bullseye on the verification target is... Using the deviation of the bullseye as the standard for target deviation, the target deviation is calculated as follows. ； ； (2) Remove the error caused by the deviation of the verification camera position: remove the error caused by the deviation of the verification camera position by extracting the target image region; (3) Determination of the center of the light spot: In the absence of a heliostat to project the light spot onto the calibration target, the background image of the calibration target is averaged; the heliostat is controlled to project the light spot of the reflected sunlight onto the calibration target, the pixel coordinates of the center of the light spot are obtained, and the relative coordinates of the center of the light spot on the calibration target are calculated. (4) Normal distribution fitting of light spot shape: Perform normal distribution fitting of light spot center position and light spot brightness; (5) Residual error removal: Calculate the relative coordinates of the spot center on the calibration target under the condition of no offset; (6) Determine the beam quality; determine the beam shape and size based on the total energy ratio threshold of the beam.

2. The method for removing wind-induced errors in heliostat spot calibration according to claim 1, characterized in that, Step (2) is as follows: The first verification camera captures a verification image containing the first verification target; The coordinates of the first verification camera are: The attitude parameters are ; These represent the angles of rotation of the first verification camera around the x, y, and z axes, respectively. The center coordinates of the first verification target corresponding to the first verification camera are: The side length of the first verification target is L; in the coordinate system of the first verification camera, the center coordinates of the first verification target are expressed as... ； in, It is a rotation matrix, represented as ； The center of the first verification target has pixel coordinates in the verification image captured by the first verification camera as follows: ； in, It is the pixel coordinate of the center of the first verification target in the verification image captured by the first verification camera; , These are the focal lengths in the x and y coordinate directions of the first verification camera, respectively. and These are the pixel coordinates in the x and y directions corresponding to the center point of the image captured by the first verification camera; The pixel length of the side length L of the first verification target in the verification image is: ； The pixel coordinates of the four vertices of the first verification target in the verification image are respectively , , , ; and then determine the four sides of the first verification target; Given that the first verification target has no offset, calculate the pixel coordinates of the four vertices; When there is wind, straight line recognition is performed within the vicinity of the first verification target position in the image. The straight line recognition algorithm calculates four straight line segments that constitute the four edges of the first verification target in the image. The target image region is then extracted, and the coordinates of the four vertices of the first verification target in the image are determined by the intersection points of the four straight line segments. , , , ; Therefore, the target center can be determined as: 。 3. The method for removing wind-induced errors in heliostat spot calibration according to claim 2, characterized in that, Step (3) is as follows: Without a heliostat projecting a light spot, a short-term background image of the calibration target is acquired. The acquired target background image is then averaged, and the brightness of each pixel is as follows: ； In the formula, It represents the averaged pixel brightness of the target background image; T represents the number of images acquired from the short-term background image of the test target, t represents the image number, and j and i represent the horizontal and vertical coordinates of the pixel, respectively. This represents the brightness of the pixel at position i,j in image number t; The designated heliostat projects the reflected sunlight onto the calibration target, and then a target image is acquired, taken simultaneously with the image obtained from step S2 when the positioning camera offset is calculated; the pixel brightness of the target image is recorded as... After performing frame difference processing, we can obtain: ； In the formula, It is the brightness of the corresponding pixel after frame difference processing; Thresholding is performed on the frame difference image, and the threshold is... Take 20~80; ； In the formula, This represents the brightness of pixel (i,j) in the image after thresholding. The pixel coordinates of the light spot center are calculated using the following formula: ； ； In the formula, and The pixel coordinates represent the center of the light spot, and j and i represent the horizontal and vertical coordinates of the pixel, respectively. and These represent the resolution of the image in the horizontal and vertical directions, respectively. The relative coordinates of the spot center on the target are calculated as follows: ； In the formula, , This represents the relative coordinates of the light spot center on the target.

4. The method for removing wind-induced errors in heliostat spot calibration according to claim 3, characterized in that, Step (4) is as follows: Fit the normal distribution of the light spot center position and the light spot brightness: ； in, This represents the coordinates of each point on the target plane. Indicates at point The relative brightness on; This represents the standard deviation of relative brightness in the x-direction. This represents the standard deviation of relative brightness in the y-direction. This represents the correlation coefficient between x and y; It is the center point of a two-dimensional normal distribution.

5. The method for removing wind-induced errors in heliostat spot calibration according to claim 4, characterized in that, Step (5) is as follows: In the world coordinate system of the mirror field, the positive x-axis points due east, and the positive y-axis points due north; the normal directions of the target surfaces of the first, second, third, and fourth verification targets point to the east, south, west, and north, respectively; the height offset of the heat absorption tower at the target center is... and ; If the normal direction of the first verification target points eastward, then the target center offset is... ; Assumption In the image captured by the first verification camera, this is manifested as the target surface shifting to the right, and the center of the light spot shifting to the left relative to the target center. The corresponding shift amount of the light spot center is: ； In the formula, This represents the lateral offset of the pixel coordinates corresponding to the center of the first verification target spot; The corresponding relative coordinates of the spot center on the first calibration target under the no-offset condition are: ； , That is, the relative coordinates of the corrected spot center on the target; The normal direction of the second calibration target points south, and the target center offset of the second calibration target is... ; Assumption In the image captured by the second verification camera, this is manifested as the target surface shifting to the right, and the center of the light spot shifting to the left relative to the target center; therefore... ； In the formula, This represents the lateral offset of the pixel coordinates corresponding to the center of the second verification target spot; The relative coordinates of the corrected spot center on the second calibration target are: ； The normal direction of the third calibration target points westward, and the target center offset of the third calibration target is... ; Assumption In the image captured by the third verification camera, the target surface shifts to the left, while the center of the light spot shifts to the right relative to the target center; therefore: ； In the formula, This represents the lateral offset of the pixel coordinates corresponding to the center of the third verification target spot; The relative coordinates of the corrected spot center on the target are: ； The normal direction of the fourth calibration target points north, and the target center offset of the fourth calibration target is... ; Assumption In the image captured by the fourth verification camera, this would manifest as the target surface shifting to the left, and the center of the light spot shifting to the right relative to the target center. ； In the formula, This represents the lateral offset of the pixel coordinates corresponding to the center of the fourth verification target spot; The relative coordinates of the corrected spot center on the target are: ； For cases where the target normal direction is not directly pointing to the four cardinal directions (east, south, west, and north), let the counterclockwise deflection angle between the target normal direction and the world coordinate x-direction be . The target offset in the image is ； The corresponding relative coordinates of the spot center on the target under the no-offset condition are: ； , This represents the relative coordinates of the corrected spot center on the target; , This represents the relative coordinates of the center of the light spot on the target before correction.

6. The method for removing wind-induced errors in heliostat spot calibration according to claim 5, characterized in that, Step (6) specifically involves: Based on the threshold of the total energy ratio of the light spot Determine the shape and size of the light spot, that is ； The left side of the above equation represents the pair Integrate; where c is a positive number, [-c , c [] represents the integration range in the x-direction, [-c , c [ ] represents the range of integration in the y-direction; the integral is when At that time, total points 1; Right side The threshold representing the proportion of total energy in the light spot is a constant. The magnitude of the offset of the corrected spot center relative to the target center corresponds to the pointing angle error, i.e. ； in, It is the distance from the center of the heliostat to the target center; This represents the side length of the calibration target; The pixel length representing the side length of the verification target in the verification image.

7. The method for removing wind-induced errors in heliostat spot calibration according to claim 6, characterized in that, Total energy ratio threshold Choose a value between 0.8 and 1.

8. A system for verifying wind-induced errors in heliostat spot calibration, using the method for removing wind-induced errors in heliostat spot calibration as described in any one of claims 1-7, characterized in that, The calibration system includes: a heliostat (1), a calibration target (2) installed on the heat absorption tower, a positioning camera (3) installed on the heat absorption tower and located below the calibration target (2), a calibration camera (5) located on the ground, and ground markers (4). During calibration, the heliostat (1) reflects sunlight onto the calibration target (2) to form a light spot; The calibration camera (5) captures the light spot on the calibration target (2); The positioning camera (3) captures images of corresponding markers on the ground to determine the offset of the verification target (2).

9. The system for verifying wind-induced errors in heliostat spot calibration according to claim 8, characterized in that, The verification system includes 1-4 sets of verification units; each set of verification units includes a verification target (2), a verification camera (5), a positioning camera (3), and a ground marker (4); the verification target (2) is installed in the east, south, west, or north direction of the heat absorption tower; the positioning camera (3) is installed below the verification target (2) in the corresponding verification unit and is used to collect images containing the corresponding ground marker; the ground marker (4) is installed on the ground in front of the positioning camera (3) in the corresponding verification unit; the verification camera (5) is used to capture images of the verification target in the corresponding verification unit; When the number of verification units is four; The calibration target (2) includes a first calibration target, a second calibration target, a third calibration target, and a fourth calibration target respectively installed in the east, south, west, and north directions of the heat absorption tower; the first calibration target, the second calibration target, the third calibration target, and the fourth calibration target are squares with the same area; The number of the verification cameras (5) is four, and the four verification cameras are set on the same horizontal plane; including a first verification camera, a second verification camera, a third verification camera and a fourth verification camera; the first verification camera, the second verification camera, the third verification camera and the fourth verification camera are respectively used to capture images containing the first verification target, the second verification target, the third verification target and the fourth verification target; The number of positioning cameras (3) is four, including a first positioning camera, a second positioning camera, a third positioning camera and a fourth positioning camera respectively installed under the first verification target, the second verification target, the third verification target and the fourth verification target; The number of ground markers (4) is four, including a first marker, a second marker, a third marker and a fourth marker; the first marker, the second marker, the third marker and the fourth marker correspond one-to-one with the first positioning camera, the second positioning camera, the third positioning camera and the fourth positioning camera, and are respectively installed on the ground in front of each positioning camera; The first positioning camera, the second positioning camera, the third positioning camera, and the fourth positioning camera are respectively used to acquire images containing the first marker, the second marker, the third marker, and the fourth marker.

Images