A heliostat calibration method with self-correction of calibration equipment

Abstract

The present invention discloses a heliostat calibration method using a self-correcting calibration device, comprising the following steps: setting initial parameters of the calibration device and initial parameters of the heliostat motion model, calculating a theoretical heliostat posture value in which the center of the heliostat light spot falls exactly on the center of a camera; calibrating each heliostat using the calibration device to obtain a heliostat posture calibration value; constructing a normal / abnormal calibration result statistical table to determine which calibration devices are abnormal; correcting abnormal calibration devices; recalibrating each heliostat using the corrected calibration device, reconstructing a normal / abnormal calibration result statistical table to determine which heliostats are abnormal; and calibrating heliostats determined to be abnormal. The calibration method disclosed by the present invention can eliminate the influence of errors in the calibration system itself, reduce the period of regular manual calibration of the calibration system, and improve the accuracy of heliostat calibration.

Description

Technical Field

[0001] The present invention relates to the technical field of heliostat calibration, and in particular to a heliostat calibration method with self-correction of calibration equipment. Background Art

[0002] Tower-type solar thermal power generation systems use heliostats that track the sun in real time to reflect sunlight onto the tower's absorber screen, heating the heat medium within the absorber and generating electricity. Failure to accurately reflect sunlight onto the absorber results in loss of concentration. Uneven distribution of light across the absorber can also lead to uneven surface temperature distribution, potentially damaging the absorber. Therefore, regular calibration of the heliostats is necessary to assess their accuracy. Any significant decrease in accuracy requires prompt correction.

[0003] An effective heliostat calibration method involves having the heliostat reflect light from an artificial light source directly onto a camera. However, the calibration system itself introduces errors, and existing solutions lack a solution for eliminating these errors. Failure to eliminate these errors significantly impacts calibration accuracy. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a heliostat calibration method with self-correction of the calibration equipment, which can eliminate the influence of the error of the inspection system itself, reduce the period of regular manual calibration of the calibration system, and improve the accuracy of heliostat calibration.

[0005] To achieve the above object, the technical solution of the present invention is as follows:

[0006] A heliostat calibration method with self-correction of a calibration device comprises the following steps:

[0007] Step 1: Set the initial parameters of the calibration equipment, including the position and attitude parameters of the camera and the position of the artificial light source;

[0008] Step 2: setting the initial parameters of the heliostat motion model, and calculating the theoretical value of the heliostat posture when the center of the heliostat spot falls exactly on the center of the camera based on the initial parameters of the heliostat motion model;

[0009] Step 3: Use the calibration equipment to calibrate each heliostat to obtain the heliostat posture calibration value so that the center of the heliostat spot falls exactly on the center of the camera;

[0010] Step 4: Based on the deviation between the calibration value and the theoretical value, a statistical table of normal / abnormal calibration results is constructed;

[0011] Step 5: According to the normal / abnormal statistical table of the calibration results, determine which calibration equipment is abnormal by setting a normal ratio threshold value;

[0012] Step 6: Correct the calibration equipment that is determined to be abnormal;

[0013] Step 7: Recalibrate each heliostat using the corrected calibration equipment and reconstruct a statistical table of normal / abnormal calibration results;

[0014] Step 8: According to the reconstructed normal / abnormal statistical table of the calibration results, determine which heliostats are abnormal by setting a normal ratio threshold 2;

[0015] Step nine: calibrate the heliostat that is determined to be abnormal.

[0016] In the above solution, in step 1, the camera's attitude parameters include three angles: azimuth, pitch, and rotation.

[0017] In the above scheme, in step 2, the initial parameters of the heliostat motion model include the coordinates of the intersection of the centerlines of the heliostat's azimuth-elevation axes, the heliostat's initial posture, the distance from the heliostat's center to the axis, the heliostat's surface deviation, and the angle between the azimuth-elevation axes. The distance between the camera and the artificial light source and the heliostat must be much greater than the heliostat's size.

[0018] In the above scheme, in step 3,

[0019] The heliostat position calculated in step 2, where the center of the heliostat spot falls exactly on the center of the camera, is used as the starting position, including the pitch angle θ h,theory and azimuth Perform a grid scan in the two-dimensional space composed of azimuth and pitch angles or a two-dimensional space after linear transformation with this posture as the center; keep the specific azimuth angle unchanged and scan the pitch angle to determine the two pitch angle boundary values ​​at this azimuth angle; keep the specific pitch angle unchanged and scan the azimuth angle to determine the two azimuth angle boundary values ​​at this pitch angle; after determining all the boundary values, calculate the center of the geometric figure composed of these boundary values, that is:

[0020]

[0021] Where n is the number of boundary points, θ i and is the pitch angle and azimuth angle corresponding to the boundary point; θ h,test and It is the pitch angle and azimuth angle corresponding to the heliostat posture obtained by verification that the center of the heliostat spot falls exactly on the center of the camera.

[0022] In the above solution, in step 4, the deviation between each calibration value and the theoretical value is calculated:

[0023]

[0024] Among them, θh,theory and They represent the pitch angle and azimuth angle of the heliostat corresponding to the theoretically calculated heliostat position where the center of the heliostat spot falls exactly on the center of the camera; θ h,test and To verify the pitch and azimuth angles of the heliostat posture corresponding to the obtained heliostat spot center exactly falling on the camera center;

[0025] The deviation is compared with the set deviation threshold to determine whether the current result is abnormal. The calibration results of all calibration combinations consisting of cameras, artificial light sources, and heliostats are summarized to construct a statistical table of normal / abnormal calibration results.

[0026] In the above solution, in step 6, if the artificial light source is determined to be abnormal, the corrected coordinates of the artificial light source are calculated by constructing the following equation 1:

[0027] θ h,theory (x l,new ,y l,new ,z l,new )=θ h,test

[0028]

[0029] Among them, x l,new ,y l,new ,z l,new Respectively represent the coordinates of the corrected artificial light source in the x, y, and z directions; θ h,theory (x l,new ,y l,new ,z l,new )and represents the pitch angle and azimuth angle of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected artificial light source parameters; θ h,test and The pitch angle and azimuth angle corresponding to the heliostat posture obtained by step 3 verification that the center of the heliostat spot falls exactly on the center of the camera;

[0030] Use the gradient descent method to solve the above equation 1 and obtain the corrected coordinates of the artificial light source, that is, x l,new ,y l,new ,z l,new .

[0031] In the above solution, in step 6, if the camera is determined to be abnormal, the corrected camera parameters are calculated by constructing the following equation 2:

[0032]

[0033] Among them, x c,new ,yc,new ,z c,new Represent the coordinates of the camera in the x, y, and z directions after correction, θ c,new , Represents the corrected camera pitch and azimuth angles; and represents the pitch and azimuth angles of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected camera parameters; θ h,test and The pitch angle and azimuth angle corresponding to the heliostat posture obtained by step 3 verification that the center of the heliostat spot falls exactly on the center of the camera;

[0034] Use the gradient descent method to solve the above equation 2 and obtain the corrected camera parameters, namely x c,new ,y c,new ,z c,new ,θ c,new ,

[0035] In a further technical solution, solving equation 1 requires at least two samples, which are the heliostat calibration for the artificial light source with the abnormality. The other two devices in the calibration combination, the camera and the heliostat, cannot be exactly the same.

[0036] The loss function is constructed as follows:

[0037]

[0038] Where p is the number of samples, which is the number of camera and heliostat groups, and p ≥ 2; k is the sampling sequence number;

[0039] The iterative process is as follows:

[0040]

[0041]

[0042] Where N represents the number of iterations, is the derivative of the loss function with respect to the artificial light source parameters at the N-1th iteration, γ is the step size of the gradient descent; x l,new,N is the x-coordinate of the artificial light source at the Nth iteration, and so on;

[0043] When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified artificial light source parameters.

[0044] In a further technical solution, solving equation 2 requires at least three samples, wherein the samples refer to the heliostat calibration for the camera with the abnormality, and the other two devices in the calibration combination, the artificial light source and the heliostat, cannot be exactly the same;

[0045] The loss function is constructed as follows:

[0046]

[0047] Where p is the number of samples, which is the number of heliostats and artificial light sources, and p ≥ 3; k is the sampling sequence number;

[0048] The iterative process is as follows:

[0049]

[0050] Among them, j represents the parameter number, N represents the number of iterations, is the derivative of the loss function with respect to the camera parameters at the N-1th iteration, γ is the step size of the gradient descent; x c,new,N is the x-coordinate of the camera at the Nth iteration, and so on;

[0051] When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified camera parameters.

[0052] In the above scheme, in step nine, the heliostat is calibrated by constructing the following equation three:

[0053] θ h,theory (a1,a2,…,a m )=θ h,test

[0054]

[0055] Among them, a1, a2, …, a m Represent the corrected parameters of the heliostat, and m represents the number of parameters in the heliostat motion model. Solving Equation 3 requires at least m / 2 samples, which are the samples used for the calibration of the abnormal heliostat. The other two devices in the calibration combination, the artificial light source and the camera, cannot be exactly the same. h,theory (a1,a2,…,a m )and is the pitch angle and azimuth angle of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected heliostat parameters; θ h,test and The pitch angle and azimuth angle of the heliostat are obtained when the center of the heliostat spot falls exactly on the center of the camera.

[0056] The gradient descent method is used to solve the parameters of the heliostat, and the loss function is constructed as follows:

[0057]

[0058] Where p is the number of samples, here the number of heliostats and lamps, and k is the sampling number;

[0059] The iterative process is as follows:

[0060]

[0061] Among them, j represents the parameter number, N represents the number of iterations, a j,N represents the value of the jth parameter of the heliostat motion model at the Nth iteration, is the derivative of the loss function with respect to the heliostat parameters at the N-1th iteration, and γ is the step size of the gradient descent. When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified heliostat parameters.

[0062] Through the above technical solution, the heliostat calibration method with self-correction of the calibration device provided by the present invention has the following beneficial effects:

[0063] The method of the present invention uses the measurement results of the calibration sample to reversely evaluate the errors of the calibration system itself, thereby removing the influence of the calibration system's own errors. This method can reduce the period of regular manual calibration of the calibration system and improve the accuracy of heliostat calibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for describing the embodiments or the prior art.

[0065] Figure 1 A schematic diagram of a calibration device comprising a camera and an artificial light source disclosed in an embodiment of the present invention;

[0066] Figure 2 Schematic diagram of the maximum position of the heliostat when rotating counterclockwise;

[0067] Figure 3 Schematic diagram of the maximum position of the heliostat when rotating clockwise;

[0068] Figure 4 The present invention discloses a flow chart of a heliostat calibration method with self-correction of the calibration equipment.

[0069] In the picture, 1. Heliostat; 2. Light; 3. Camera. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0071] The present invention proposes a calibration system consisting of an artificial light source, a heliostat and a camera. In this embodiment, the artificial light source is replaced by a lamp.

[0072] First, let's take a two-dimensional simplified model as an example to introduce the verification principle. Here, it is assumed that the heliostat 1 is a line segment and rotates around its center point; the camera 3 lens is also simplified to a line segment; and the lamp 2 is simplified to a point. Figure 1 As shown, the current heliostat posture can reflect light 2 onto camera 3. Figure 1 The middle heliostat rotates counterclockwise around the center of heliostat 1 to Figure 2 In the heliostat posture shown, the right edge of heliostat 1 reflects the light from lamp 2 to the left edge of camera 3. If heliostat 1 continues to rotate counterclockwise, camera 3 cannot capture the light reflected by heliostat 1. Figure 1 The middle heliostat position rotates clockwise around the center of heliostat 1 to Figure 3 In the heliostat posture shown, the left edge of heliostat 1 reflects the light from lamp 2 to the right edge of camera 3. If heliostat 1 continues to rotate clockwise, camera 3 cannot capture the light reflected by heliostat 1.

[0073] Considering the influence of scattering, when heliostat 1 moves from Figure 3 Rotate the shown position counterclockwise to Figure 2 In the posture shown, the imaging spot at the position of heliostat 1 in camera 3 changes from dark to bright and then back to dark. In this simplified two-dimensional structure, the posture of heliostat 1 with the brightest spot corresponds to the situation where the center of heliostat 1 reflects light from lamp 2 to the center of camera 3. Therefore, the posture of heliostat 1 can be determined by the change in the brightness of the spot in camera 3. The motion control device of heliostat 1 also generates the heliostat posture. By examining the deviation between the two, the heliostat motion model can be calibrated.

[0074] In practice, the principle of calculating the heliostat pose that can be reflected to the center of the pose range of camera 3 is the same as that of the 2D simplified model. Compared with the 2D simplified model, the actual practice has an additional z dimension and uses the actual heliostat motion model.

[0075] The heliostat motion model in the present invention refers to an actual, real heliostat motion model. The heliostat motion model depends on the type of heliostat, such as spin-pitch, azimuth-pitch, etc. Taking an azimuth-pitch rectangular heliostat as an example, the model is established by considering the following parameters, including the coordinates of the intersection point (axis center) of the center lines of the heliostat's azimuth-pitch axes, the heliostat's initial posture (the three angles of azimuth, pitch, and rotation), the distance from the heliostat center to the axis center, the heliostat's surface deviation, the angle between the azimuth-pitch axes, etc. A detailed discussion of the heliostat motion model has been published in scientific and technological literature and will not be repeated here. For details, please refer to "A General Heliostat Accurate Azimuth-Pitch Tracking Formula and Its Application Research, Electric Power and Energy Progress, 2015, 3(5): 123-138".

[0076] Based on the above discussion on the principle, the present invention provides a heliostat calibration method with self-correction of the calibration equipment, such as Figure 4 As shown, the following steps are included:

[0077] Step 1: Set the initial parameters of the calibration equipment, including the position and attitude parameters of the camera and the position of the light; the attitude parameters of the camera include the three angles of azimuth, pitch and rotation; Figure 1 As shown, the distances between the camera and the artificial light source and the heliostat are much larger than the size of the heliostat.

[0078] Step 2: Set the initial parameters of the heliostat motion model; and calculate the theoretical value of the heliostat posture based on the initial parameters of the heliostat motion model, including the pitch angle θ, so that the center of the heliostat spot falls exactly on the center of the camera. h,theory and azimuth

[0079] Taking the azimuth-elevation rectangular heliostat as an example, the initial parameters of the heliostat motion model include the coordinates of the intersection of the centerlines of the heliostat's azimuth-elevation axes, the heliostat's initial posture, the distance from the heliostat's center to the axis, the heliostat's surface deviation, and the angle between the azimuth-elevation axes.

[0080] Step 3: Use the calibration equipment to calibrate each heliostat to obtain the heliostat posture calibration value so that the center of the heliostat spot falls exactly on the center of the camera.

[0081] The heliostat posture calculated in step 2 is used as the starting position. A grid scan is performed in the two-dimensional space composed of azimuth and elevation angles or the two-dimensional space after linear transformation, with this posture as the center. The specific azimuth angle is kept constant and the elevation angle is scanned to determine the two elevation angle boundary values ​​at this azimuth angle. The specific elevation angle is kept constant and the azimuth angle is scanned to determine the two elevation angle boundary values ​​at this pitch angle. After all boundary values ​​are determined, the center of the geometric figure composed of these boundary values ​​is calculated, that is:

[0082]

[0083] Where n is the number of boundary points, θ i and is the pitch angle and azimuth angle corresponding to the boundary point; θ h,test and It is the pitch angle and azimuth angle corresponding to the heliostat posture obtained by verification that the center of the heliostat spot falls exactly on the center of the camera.

[0084] Step 4: Based on the deviation between the calibration value and the theoretical value, a statistical table of normal / abnormal calibration results is constructed;

[0085] Calculate the deviation of each calibration value from the theoretical value:

[0086]

[0087] The deviation is compared with the set deviation threshold to determine whether the current result is abnormal. The calibration results of all calibration combinations consisting of cameras, lights and heliostats are summarized to construct a statistical table of normal / abnormal calibration results.

[0088] The following table is an example of a normal / abnormal statistics table for the verification results.

[0089] Table 1. Statistics of normal / abnormal verification results

[0090] <![CDATA[L1,C1]]> <![CDATA[L1,C2]]> <![CDATA[L1,C3]]> <![CDATA[L2,C1]]> <![CDATA[L2,C2]]> <![CDATA[L2,C3]]> <![CDATA[L3,C1]]> <![CDATA[L3,C2]]> <![CDATA[L3,C3]]> <![CDATA[H1]]> x x x √ √ √ √ √ √ <![CDATA[H2]]> x x x √ √ √ √ √ √ <![CDATA[H3]]> x x x √ √ √ √ √ √ <![CDATA[H4]]> x x x √ √ √ √ √ √ <![CDATA[H5]]> x x x √ √ √ √ √ √ <![CDATA[H6]]> x x x √ √ √ √ √ √ <![CDATA[H7]]> x x x √ √ √ √ √ √ <![CDATA[H8]]> x x x √ √ √ √ √ √ <![CDATA[H9]]> x x x √ √ √ √ √ √ <![CDATA[H1]]> x x x x x x x x x

[0091] In the table, H1~H 10 represents heliostats 1-10, L1, L2, L3 represent lights 1, 2, 3, C1, C2, C3 represent cameras 1, 2, 3, √ indicates that the calibration result is normal, and x indicates that the calibration result is abnormal.

[0092] Step 5: According to the normal / abnormal statistical table of the calibration results, determine which calibration equipment is abnormal by setting a normal ratio threshold value;

[0093] The calibration equipment consists of two types: a camera and a light. These two devices are paired together to calibrate different heliostats. This combination is not fixed; a single camera can be paired with multiple lights, and a single light can be paired with multiple cameras. Because the calibration equipment combinations are not fixed, and the initial equipment is calibrated, only a few devices will exhibit significant errors. If the percentage of abnormal results from a particular device exceeds a set threshold (e.g., 90%), the device is considered abnormal.

[0094] For example, in Table 1, L1 is clearly an abnormal calibration device.

[0095] Step 6: Correct the calibration equipment that is determined to be abnormal;

[0096] (1) If the lamp is judged to be abnormal, the corrected coordinates of the lamp are calculated by constructing the following equation 1:

[0097] θ h,theory (x l,new ,y l,new ,z l,new )=θ h,test

[0098]

[0099] Among them, x l,new ,y l,new ,z l,new Respectively represent the coordinates of the corrected light in the x, y, and z directions; θ h,theory (x l,new ,y l,new ,z l,new )and represents the pitch and azimuth angles of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected lamp parameters; θ h,test and The pitch angle and azimuth angle corresponding to the heliostat posture obtained by step 3 verification that the center of the heliostat spot falls exactly on the center of the camera;

[0100] Use the gradient descent method to solve the above equation 1 and get the corrected coordinates of the lamp, that is, x l,new ,y l,new ,z l,new .

[0101] Solving Equation 1 requires at least two samples. The samples refer to the heliostat calibration for the abnormal lamp. The other two devices in the calibration combination, the camera and the heliostat, cannot be exactly the same.

[0102] The loss function is constructed as follows:

[0103]

[0104] Where p is the number of samples, which is the number of camera and heliostat groups, and p ≥ 2; k is the sampling sequence number;

[0105] The iterative process is as follows:

[0106]

[0107] Where N represents the number of iterations, is the derivative of the loss function with respect to the lamp parameter at the N-1th iteration, γ is the step size of the gradient descent; x l,new,N is the x-coordinate of the light at the Nth iteration, and so on;

[0108] When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified lamp parameters.

[0109] (2) If the camera is judged to be abnormal, the corrected camera parameters are calculated by constructing the following equation 2:

[0110]

[0111] Among them, x c,new ,y c,new ,z c,new Represent the coordinates of the camera in the x, y, and z directions after correction, θ c,new , Represents the corrected camera pitch and azimuth angles; and represents the pitch and azimuth angles of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected camera parameters; θ h,test and The pitch angle and azimuth angle corresponding to the heliostat posture obtained by step 3 verification that the center of the heliostat spot falls exactly on the center of the camera;

[0112] Use the gradient descent method to solve the above equation 2 and obtain the corrected camera parameters, namely x c,new ,y c,new ,z c,new ,θ c,new ,

[0113] Solving Equation 2 requires at least three samples. These samples refer to the heliostat calibration for the camera with the anomaly. The other two devices in the calibration combination, the lamp and the heliostat, cannot be exactly the same.

[0114] The loss function is constructed as follows:

[0115]

[0116] Where p is the number of samples, which is the number of heliostats and lamps, and p ≥ 3; k is the sampling sequence number;

[0117] The iterative process is as follows:

[0118]

[0119] Where N represents the number of iterations, is the derivative of the loss function with respect to the camera parameters at the N-1th iteration, γ is the step size of the gradient descent; x c,new,j,N is the x-coordinate of the camera at the Nth iteration, and so on;

[0120] When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified camera parameters.

[0121] Step 7: Recalibrate each heliostat using the corrected calibration equipment and reconstruct a statistical table of normal / abnormal calibration results;

[0122] For example, after calibration of the calibration equipment in Table 1, reconstructing the normal / abnormal statistics table of the new calibration results will produce Table 2.

[0123] Table 2. Statistics of normal / abnormal calibration results (after equipment calibration)

[0124] <![CDATA[L1,C1]]> <![CDATA[L1,C2]]> <![CDATA[L1,C3]]> <![CDATA[L2,C1]]> <![CDATA[L2,C2]]> <![CDATA[L2,C3]]> <![CDATA[L3,C1]]> <![CDATA[L3,C2]]> <![CDATA[L3,C3]]> <![CDATA[H1]]> √ √ √ √ √ √ √ √ √ <![CDATA[H2]]> √ √ √ √ √ √ √ √ √ <![CDATA[H3]]> √ √ √ √ √ √ √ √ √ <![CDATA[H4]]> √ √ √ √ √ √ √ √ √ <![CDATA[H5]]> √ √ √ √ √ √ √ √ √ <![CDATA[H6]]> √ √ √ √ √ √ √ √ √ <![CDATA[H7]]> √ √ √ √ √ √ √ √ √ <![CDATA[H8]]> √ √ √ √ √ √ √ √ √ <![CDATA[H9]]> √ √ √ √ √ √ √ √ √ <![CDATA[H1]]> x x x x x x x x x

[0125] Step 8: According to the reconstructed normal / abnormal statistical table of the calibration results, determine which heliostats are abnormal by setting a normal ratio threshold 2;

[0126] For example, it is obvious from Table 2 that the heliostat H 10 abnormal.

[0127] Step nine: calibrate the heliostat that is determined to be abnormal.

[0128] The heliostat is calibrated by constructing the following equation:

[0129] θ h,theory (a1,a2,…,a m )=θ h,test

[0130]

[0131] Among them, a1, a2, …, a m Represent the corrected parameters of the heliostat, and m represents the number of parameters in the heliostat motion model. Solving Equation 3 requires at least m / 2 samples, where the sample refers to the calibration of the abnormal heliostat. The other two devices in the calibration combination, the light and the camera, cannot be exactly the same. h,theory (a1,a2,…,a m )and is the pitch angle and azimuth angle of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected heliostat parameters; θ h,test and The pitch angle and azimuth angle of the heliostat are obtained when the center of the heliostat spot falls exactly on the center of the camera.

[0132] The gradient descent method is used to solve the parameters of the heliostat, and the loss function is constructed as follows:

[0133]

[0134] Where p is the number of samples, here the number of heliostats and lamps, and k is the sampling number;

[0135] The iterative process is as follows:

[0136]

[0137] Among them, j represents the parameter number, N represents the number of iterations, a j,N represents the value of the jth parameter of the heliostat motion model at the Nth iteration, is the derivative of the loss function with respect to the heliostat parameters at the N-1th iteration, and γ is the step size of the gradient descent. When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified heliostat parameters.

[0138] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A heliostat calibration method with self-correction of calibration equipment, characterized in that: The steps include: Step 1: Set the initial parameters of the calibration equipment, including the position and attitude parameters of the camera and the position of the artificial light source; Step 2: setting the initial parameters of the heliostat motion model, and calculating the theoretical value of the heliostat posture when the center of the heliostat spot falls exactly on the center of the camera based on the initial parameters of the heliostat motion model; Step 3: Use the calibration equipment to calibrate each heliostat to obtain the heliostat posture calibration value so that the center of the heliostat spot falls exactly on the center of the camera; Step 4: Based on the deviation between the calibration value and the theoretical value, the deviation is compared with the set deviation threshold to determine whether the current result is abnormal. The calibration results of all calibration combinations consisting of cameras, artificial light sources, and heliostats are summarized to construct a statistical table of normal / abnormal calibration results. Step 5: According to the normal / abnormal statistical table of the calibration results, determine which calibration equipment is abnormal by setting a normal ratio threshold value; Step 6: Correct the calibration equipment that is determined to be abnormal; Step 7: Recalibrate each heliostat using the corrected calibration equipment and reconstruct a statistical table of normal / abnormal calibration results; Step 8: According to the reconstructed normal / abnormal statistical table of the calibration results, determine which heliostats are abnormal by setting a normal ratio threshold 2; Step nine: calibrate the heliostat that is determined to be abnormal.

2. The heliostat calibration method with self-correction of calibration equipment according to claim 1, characterized in that: In step 1, the camera's attitude parameters include azimuth, pitch, and rotation; the distance between the camera and the artificial light source and the heliostat must be much larger than the size of the heliostat.

3. The heliostat calibration method with self-correction of calibration equipment according to claim 1, characterized in that: In step 2, the initial parameters of the heliostat motion model include the coordinates of the intersection of the heliostat's azimuth-elevation axis centerlines, the heliostat's initial posture, the distance from the heliostat's center to the axis center, the heliostat's surface deviation, and the angle between the azimuth-elevation axis.

4. The heliostat calibration method with self-correction of calibration equipment according to claim 1, characterized in that: In step three, The heliostat position calculated in step 2, where the center of the heliostat spot falls exactly on the center of the camera, is used as the starting position, including the pitch angle θ h,theory and azimuth Perform a grid scan in the two-dimensional space composed of azimuth and pitch angles or a two-dimensional space after linear transformation with this posture as the center; keep the specific azimuth angle unchanged and scan the pitch angle to determine the two pitch angle boundary values ​​at this azimuth angle; keep the specific pitch angle unchanged and scan the azimuth angle to determine the two azimuth angle boundary values ​​at this pitch angle; after determining all the boundary values, calculate the center of the geometric figure composed of these boundary values, that is: Where n is the number of boundary points, θ i and is the pitch angle and azimuth angle corresponding to the boundary point; θ h,test and It is the pitch angle and azimuth angle corresponding to the heliostat posture obtained by verification that the center of the heliostat spot falls exactly on the center of the camera.

5. The heliostat calibration method with self-correction of calibration equipment according to claim 1, characterized in that: In step 4, the deviation between each calibration value and the theoretical value is calculated: Among them, θ h,theory and They represent the pitch angle and azimuth angle of the heliostat corresponding to the theoretically calculated heliostat position where the center of the heliostat spot falls exactly on the center of the camera; θ h,test and To verify the pitch angle and azimuth angle of the heliostat, the center of the heliostat spot falls exactly on the center of the camera.

6. The heliostat calibration method with self-correction of calibration equipment according to claim 1, characterized in that: In step 6, if the artificial light source is determined to be abnormal, the corrected coordinates of the artificial light source are calculated by constructing the following equation 1: θ h,theory (x l,new ,y l,new ,z l,new )=θ h,test Among them, x l,new ,y l,new ,z l,new Respectively represent the coordinates of the corrected artificial light source in the x, y, and z directions; θ h,theory (x l,new ,y l,new ,z l,new )and represents the pitch angle and azimuth angle of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected artificial light source parameters; θ h,test and The pitch angle and azimuth angle corresponding to the heliostat posture obtained by step 3 verification that the center of the heliostat spot falls exactly on the center of the camera; Use the gradient descent method to solve the above equation 1 and obtain the corrected coordinates of the artificial light source, that is, x l,new ,y l,new ,z l,new .

7. The heliostat calibration method with self-correction of calibration equipment according to claim 1, characterized in that: In step 6, if the camera is judged to be abnormal, the corrected camera parameters are calculated by constructing the following equation 2: Among them, x c,new ,y c,new ,z c,new Represent the coordinates of the corrected camera in the x, y, and z directions, respectively. Represents the corrected camera pitch and azimuth angles; and represents the pitch and azimuth angles of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected camera parameters; θ h,test and The pitch angle and azimuth angle corresponding to the heliostat posture obtained by step 3 verification that the center of the heliostat spot falls exactly on the center of the camera; Use the gradient descent method to solve the above equation 2 and obtain the corrected camera parameters, which are 8. The heliostat calibration method with self-correction of calibration equipment according to claim 6, characterized in that: Solving Equation 1 requires at least two samples. These samples refer to the heliostat calibration for the artificial light source that has an abnormality. The other two devices in the calibration combination, the camera and the heliostat, cannot be exactly the same. The loss function is constructed as follows: Where p is the number of samples, which is the number of camera and heliostat groups, and p ≥ 2; k is the sampling sequence number; The iterative process is as follows: Where N represents the number of iterations, is the derivative of the loss function with respect to the artificial light source parameters at the N-1th iteration, γ is the step size of the gradient descent; x l,new,N is the x-coordinate of the artificial light source at the Nth iteration, and so on; When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified artificial light source parameters.

9. The heliostat calibration method with self-correction of calibration equipment according to claim 7, characterized in that: Solving Equation 2 requires at least three samples. These samples refer to the heliostat calibration for the camera with the anomaly. The other two devices in the calibration combination, the artificial light source and the heliostat, cannot be exactly the same. The loss function is constructed as follows: Where p is the number of samples, which is the number of heliostats and artificial light sources, and p ≥ 3; k is the sampling sequence number; The iterative process is as follows: Where N represents the number of iterations, is the derivative of the loss function with respect to the camera parameters at the N-1th iteration, γ is the step size of the gradient descent; x c,new,N is the x coordinate of the camera at the Nth iteration, and so on; When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified camera parameters.

10. The heliostat calibration method with self-correction of calibration equipment according to claim 1, characterized in that: In step nine, the heliostat is calibrated by constructing the following equation three: i h,theory (a1,a2,…,a m )=θ h,test Among them, a1, a2, …, a m Represent the corrected parameters of the heliostat, and m represents the number of parameters in the heliostat motion model. Solving Equation 3 requires at least m / 2 samples, which are the samples used for the calibration of the abnormal heliostat. The other two devices in the calibration combination, the artificial light source and the camera, cannot be exactly the same. h,theory (a1,a2,…,a m )and is the pitch angle and azimuth angle of the heliostat when the center of the heliostat spot falls exactly on the center of the camera, calculated based on the initial parameters of the motion model under the corrected heliostat parameters; θ h,test and The pitch angle and azimuth angle of the heliostat are obtained when the center of the heliostat spot falls exactly on the center of the camera. The gradient descent method is used to solve the parameters of the heliostat, and the loss function is constructed as follows: Where p is the number of samples, here the number of heliostats and lamps, and k is the sampling number; The iterative process is as follows: Among them, j represents the parameter number, N represents the number of iterations, a j,N represents the value of the jth parameter of the heliostat motion model at the Nth iteration, is the derivative of the loss function with respect to the heliostat parameters at the N-1th iteration, and γ is the step size of the gradient descent. When the loss function is less than the set threshold, the iteration ends and the obtained parameters are the verified heliostat parameters.

Images