Dish concentrator azimuth axis tilt measurement and correction method based on light spot timing

Abstract

A method for measuring and correcting the azimuth axis tilt of a dish condenser based on light spot time sequence includes the following steps: establishing a dish condenser tracking model with azimuth axis tilt; calculating the tracking error of the dish condenser at any time; determining the centroid coordinates of the focal plane focusing energy flow distribution under the above tracking error using a ray tracing method; repeating the above steps to calculate the centroid coordinates of the focusing energy flow distribution during the tracking process every day of the year under different azimuth axis tilt errors, forming a database, and using a neural network to train and obtain a prediction model; selecting a certain day, measuring the time sequence data of the centroid coordinates of the focal plane focusing energy flow distribution during the tracking operation of the dish condenser, and inputting it into the prediction model to predict the actual azimuth axis tilt error value, and using this to establish a solar tracking correction model for the dish condenser, thereby improving the tracking accuracy of the dish condenser; this invention can perform rapid measurement and tracking correction in service, and has the advantages of high efficiency, flexibility, and low cost.

Description

Technical Field

[0001] This invention belongs to the field of solar concentrated thermal power generation, and in particular relates to a method for measuring and tracking the tilt of the azimuth axis of a dish concentrator based on light spot timing measurement. Background Technology

[0002] Solar energy is a clean, environmentally friendly, abundant, and widely distributed renewable energy source. Developing and utilizing solar energy resources for concentrated solar power (CSP) is one of the important ways to upgrade the future energy structure. A dish concentrator is an optical device used in CSP. It concentrates low-density solar energy onto a small-area receiver to form high-density energy, allowing the working fluid to reach a higher temperature, thereby improving the photoelectric energy conversion efficiency of the entire CSP system. For example, dish concentrators are used in dish / Stirling CSP systems.

[0003] Because the sun's position changes constantly throughout the day, dish concentrators need to accurately track the sun's position—that is, achieve high tracking accuracy—to concentrate sunlight efficiently and effectively into the receiver. Altitude-azimuth dual-axis tracking is currently the most widely used dual-axis sun tracking method in heliostats and dish concentrators. The fixed axis is the azimuth rotation axis, vertically upward and perpendicular to the ground plane, while the driven axis is the altitude rotation axis, horizontal in direction. To reduce commercialization costs, altitude-azimuth dual-axis tracking in both dish concentrators and heliostats typically employs an open-loop control strategy based on the sun's position information. The sun's position can be determined using high-precision astronomical algorithms, with a calculated error of up to 0.003°. Theoretically, its tracking accuracy is already very high and fully meets the requirements of engineering applications. However, installation errors in the support columns of the dish concentrator and over 20 years of service can lead to ground settlement, both of which cause the support columns to tilt. This disrupts the perpendicularity between the support column axis (i.e., the azimuth axis) and the ground plane, resulting in tracking errors. Since the tilt error of the column will gradually change over long-term service, its measurement and correction become more difficult, and current technologies have not yet offered a solution. Therefore, there is a need to invent a method that can quickly measure and determine the azimuth axis tilt error and perform tracking compensation and correction, without affecting normal solar power generation. This would enable the dish concentrator to track the sun's position with high precision even when the azimuth axis is tilted, which is particularly important and urgent for engineering practice. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for measuring and correcting the azimuth axis tilt of a dish concentrator based on beam spot time-series measurement. This method utilizes the one-to-one correspondence between the azimuth axis tilt error and the centroid position of the focused beam spot at different dates and times. A neural network prediction algorithm is used to establish a prediction model. Using the time-series data of the centroid coordinates measured on a specific day as input, the azimuth axis tilt error value of the dish concentrator is predicted. This error is then used to correct the tracking model, enabling the dish concentrator to track the sun's position with high precision even when the azimuth axis is tilted, thereby improving the heat collection efficiency and safe operation performance of the dish concentrator.

[0005] The technical solution adopted in this invention is: a method for measuring and correcting the azimuth axis tilt of a dish concentrator based on spot timing, comprising the following implementation steps:

[0006] Step 1: Establish a sun-tracking model for a dish concentrator that includes azimuth axis tilt error, and determine the actual focal axis vector at any given time: Establish a global coordinate system O- at the intersection point O of the ideal azimuth axis and the ground plane. xyz , where + y The axis points due south. x The axis points due east, + z The axis points to the zenith and coincides with the ideal azimuth axis; astronomical algorithms are used to determine any date of the year in the area where the disc concentrator is to be constructed. t The true position of the sun at any given moment is determined by the solar altitude angle. β With solar azimuth α Composition; using radial error angle θ 1 and circumferential error angle Describe the magnitude and direction of the tilt angle of the azimuth axis, respectively; t The actual focal axis vector N of the disc condenser at that moment ft for:

[0007] ;

[0008] In the formula, the function This is a rotation function matrix used to implement arbitrary vectors. Rotate by any angle around any unit vector The functions are as follows: N a1 It is the actual azimuth axis vector. N et It is the actual height axis vector. N f =[-1,0,0]、N a =[0,0,-1] and N e=[0,1,0] represent the focal axis vector, azimuth axis vector, and height axis vector of the disc concentrator under ideal operating conditions without column tilt error, respectively, and are all values ​​when the focal axis vector points due east at the start of daily tracking operation of the disc concentrator; vector Vector ;

[0009] Step 2: Calculate the results from Step 1 t The tracking error value of the disc condenser at any given time, including the height tracking error angle of the disc condenser. and azimuth tracking error angle The two tracking error components are as follows:

[0010] ;

[0011] In the formula, and These are the actual elevation and azimuth angles tracked by the disc concentrator, respectively. Vector azimuth It can be regarded as the actual focal axis vector N ft Projection vector N1 along the ground plane = [ x 1, y [1, 0] and - x The included angle of the axis, ;

[0012] Step 3: Set the aperture radius of the parabolic reflector of the disc condenser. R and focal length f Based on the two tracking error component values ​​determined in step 2, the Monte Carlo ray tracing method is used to calculate the focusing energy flux density distribution of the planar receiver located on the focal plane of the dish condenser, and the centroid coordinates of the focusing energy flux distribution are calculated. x c and y c ;

[0013] Step 4: Repeat steps 1 to 3 to calculate the centroid coordinates of the focusing energy flow distribution on the focal plane at any time during the tracking process throughout the year under different azimuth axis tilt errors, forming a database of the changes in centroid coordinate values ​​with the time of day for different dates and column tilt error values; then, use a neural network prediction algorithm to train this database and obtain a prediction model, so that by inputting the time series data of the changes in centroid coordinates with the time of day for a certain day, the azimuth axis tilt error value of the disc concentrator can be predicted;

[0014] Step 5: Select a day M in the whole year, use a CCD camera to measure the focusing energy flow distribution on the focal plane during the open-loop tracking operation of the disc condenser, and determine the centroid coordinates of the focusing energy flow distribution at different times to obtain time-series data of the centroid coordinates changing over time.

[0015] Step 6: Input the centroid coordinate time series data obtained in Step 5 into the prediction model established in Step 4, and set the same date in the prediction model to predict the tilt error value of the actual azimuth axis of the disc concentrator, including the error angles describing the magnitude and direction of the tilt angle of the azimuth axis. θ 2 and error angle ;

[0016] Step 7: Based on the azimuth tilt error value predicted in Step 6 θ 2 and The apparent solar tracking correction model for the dish concentrator is established as follows:

[0017] ;

[0018] In the formula, and N s =[-1,0,0]; ; The rotation angle is obtained by solving the problem. β 1 and α 1, these are any of the steps in step 1 after tracking and correction. t The rotation angles of the height axis and azimuth axis in the motor-driven dual-axis tracking mechanism at any given time.

[0019] In the above-mentioned method for measuring and correcting the azimuth axis tilt of a dish condenser based on spot timing, the specific process of step 3 is as follows: The planar receiver located on the focal plane is a square with a side length of L, and its center point p coincides with the focal point of the dish condenser. A p-... is established on the receiver plane with point p as the origin. x p y p coordinate system x p The axis is parallel to the bottom edge of the planar receiver, pointing to the right and parallel to the ground plane; the planar receiver is divided into two equal parts along the perpendicular directions. H Shares and U portions, and H = U Discretized into H × U A small square grid was used to calculate the centroid coordinates of the focused energy flow distribution. x c and y c They are respectively:

[0020] ;

[0021] In the formula, h and u These are the horizontal and vertical numbers of the discrete small square grid; The horizontal and vertical numbering are respectively h and u Energy flux density of a discrete grid.

[0022] In the above-mentioned method for measuring and correcting the azimuth axis tilt of a dish condenser based on the time sequence of light spots, the date selected in step 5 is either the summer solstice or the winter solstice.

[0023] In the above-mentioned method for measuring and correcting the azimuth axis tilt of a disc concentrator based on spot timing, steps 4 and 5 both obtain the focusing energy flow distribution and centroid coordinates at the same starting point and time interval T1.

[0024] In the aforementioned method for measuring and correcting the azimuth axis tilt of a dish condenser based on spot timing, step 5 involves installing a flip-up, water-cooled square Lambertian target on the absorber of the dish condenser. When measuring the focusing energy flow distribution, the Lambertian target is flipped to the focal plane position, with its center coinciding with the focal point. A local coordinate system q- is established within the Lambertian target plane with the center point q. x q y q To determine the centroid coordinates, where x q The axis is parallel to the bottom edge of the plane receiver, pointing to the right and parallel to the ground plane; when not measuring, it is in a retracted state and does not obstruct the normal operation of the absorber receiving and concentrating solar energy; the CCD camera is mounted on the disc concentrator and perpendicular to the focal plane, acquiring an image of the focused spot on the surface of the Lambertian target located on the focal plane; the focused spot image is transmitted to the processor for grayscale processing, and then the centroid coordinates of the focused spot weighted by grayscale values ​​are calculated; the focused spot is measured once at intervals T1 according to the above process and the centroid coordinates are determined, finally obtaining the time series data of the centroid coordinates changing with time.

[0025] Compared with existing technologies, the advantages of this invention are as follows: This invention establishes a one-to-one correspondence database between the tilt error of the azimuth axis and the centroid position of the focused spot at different dates and times throughout the year through theoretical tracking models and Monte Carlo ray tracking methods. It establishes an efficient prediction model using a neural network prediction algorithm. Using the time-series data of the centroid coordinates measured on-site on a certain day as input, it can predict the azimuth axis tilt error value of the dish concentrator with high accuracy. Then, it uses this to correct the tracking model, thereby improving the tracking accuracy of the dish concentrator. The on-site measured focused spot uses a flip-out Lambertian target in conjunction with a CCD camera to intermittently collect the spot, which does not affect the normal operation of the dish concentrating solar thermal system or the concentrating solar power generation system. It can perform autonomous measurement and tracking correction while in service, and has the advantages of high efficiency, flexibility and low cost. Attached Figure Description

[0026] Picture 1 This is the coordinate system and related variable representation of the apparent solar tracking model of the dish concentrator in this invention.

[0027] Picture 2 This is a schematic diagram of the centroid coordinates of the focusing energy flow distribution at the focal plane position in this invention. Detailed Implementation

[0028] The invention will now be further described with reference to the accompanying drawings.

[0029] The present invention provides a method for measuring and correcting the azimuth axis tilt of a dish concentrator based on spot timing, the implementation steps of which are as follows:

[0030] Step 1: Establish a sun-tracking model for the disc condenser that includes azimuth axis tilt error, and determine the actual focal axis vector at any given time: (e.g., ...) Picture 1 As shown, a global coordinate system O- is established at the intersection point O of the ideal azimuth axis and the ground plane. xyz , where + y The axis points due south. x The axis points due east, + z The axis points to the zenith and coincides with the ideal azimuth axis; astronomical algorithms are used to determine any date of the year in the area where the disc concentrator is to be constructed. t The true position of the sun at any given moment is determined by the solar altitude angle. β With solar azimuth α Composition; using radial error angle θ 1 and circumferential error angle Describe the magnitude and direction of the tilt angle of the azimuth axis, respectively; t The actual focal axis vector N of the disc condenser at that moment ft for:

[0031] ;

[0032] In the formula, the function This is a rotation function matrix used to implement arbitrary vectors. Rotate by any angle around any unit vector The functions are as follows: N a1 It is the actual azimuth axis vector. N et It is the actual height axis vector. N f =[-1,0,0]、N a =[0,0,-1] and N e =[0,1,0] represent the focal axis vector, azimuth axis vector, and height axis vector of the disc concentrator under ideal operating conditions without column tilt error, respectively, and are all values ​​when the focal axis vector points due east at the start of daily tracking operation of the disc concentrator; vector Vector ;

[0033] Step 2: Calculate the results from Step 1 t The tracking error value of the disc condenser at any given time, including the height tracking error angle of the disc condenser. and azimuth tracking error angle The two tracking error components are as follows:

[0034] ;

[0035] In the formula, and These are the actual elevation and azimuth angles tracked by the disc concentrator, respectively. Vector azimuth It can be regarded as the actual focal axis vector N ft Projection vector N1 along the ground plane = [ x 1, y [1, 0] and - x The included angle of the axis, ;

[0036] Step 3: Set the aperture radius of the parabolic reflector of the disc condenser. R and focal length f Based on the two tracking error component values ​​determined in step 2, the Monte Carlo ray tracing method is used to calculate the focusing energy flux density distribution of the planar receiver located on the focal plane of the dish condenser, and the centroid coordinates of the focusing energy flux distribution are calculated. x c and y c ,like Picture 2As shown; the main operation process includes: the planar receiver located on the focal plane is a square with a side length of L, its center point p coincides with the focal point of the disc condenser, and a p- is established on the receiver plane with point p as the origin. x p y p coordinate system x p The axis is parallel to the bottom edge of the planar receiver, pointing to the right and parallel to the ground plane; the planar receiver is divided into two equal parts along the perpendicular directions. H Shares and U portions, and H = U Discretized into H × U A small square grid was used to calculate the centroid coordinates of the focused energy flow distribution. x c and y c They are respectively:

[0037] ;

[0038] In the formula, h and u These are the horizontal and vertical numbers of the discrete small square grid; The horizontal and vertical numbering are respectively h and u Energy flux density of a discrete grid.

[0039] Step 4: Repeat steps 1 to 3 to calculate the centroid coordinates of the focusing energy flow distribution on the focal plane at any time during the tracking process throughout the year under different azimuth axis tilt errors, forming a database of the changes in centroid coordinate values ​​with the time of day for different dates and column tilt error values; then, use a neural network prediction algorithm to train this database and obtain a prediction model, so that by inputting the time series data of the changes in centroid coordinates with the time of day for a certain day, the azimuth axis tilt error value of the disc concentrator can be predicted;

[0040] Step 5: Select a day M in the whole year, use a CCD camera to measure the focusing energy flow distribution on the focal plane during the open-loop tracking operation of the disc condenser, and determine the centroid coordinates of the focusing energy flow distribution at different times to obtain time-series data of the centroid coordinates changing over time.

[0041] Step 6: Input the centroid coordinate time series data obtained in Step 5 into the prediction model established in Step 4, and set the same date in the prediction model to predict the tilt error value of the actual azimuth axis of the disc concentrator, including the error angles describing the magnitude and direction of the tilt angle of the azimuth axis. θ 2 and error angle ;

[0042] Step 7: Based on the azimuth tilt error value predicted in Step 6 θ 2 and The apparent solar tracking correction model for the dish concentrator is established as follows:

[0043] ;

[0044] In the formula, and N s =[-1,0,0]; ; The rotation angle is obtained by solving the problem. β 1 and α 1, these are any of the steps in step 1 after tracking and correction. t The motor drives the rotation angle of the altitude and azimuth axes in the dual-axis tracking mechanism at all times, so that the disc concentrator can accurately track the sun's position even when there is an azimuth axis tilt error.

[0045] Preferably, the date selected in step 5 is the summer solstice or the winter solstice, which makes it easier to measure the effect of the azimuth axis tilt error on the position of the centroid of the focused spot.

[0046] Preferably, steps 4 and 5 are both performed at the same time starting point and at a time interval T1 to obtain the focused energy flow distribution and centroid coordinates, without the need for continuous time energy flow distribution results, thus avoiding long-term focused spot measurement.

[0047] Preferably, step 5 involves installing a flip-up, water-cooled square Lambertian target on the absorber of the disc concentrator. When measuring the focusing energy flow distribution, the Lambertian target is flipped to the focal plane position, with its center coinciding with the focal point. A local coordinate system q- is established within the Lambertian target plane with the center point q. x q y q To determine the centroid coordinates, where x qThe axis is parallel to the bottom edge of the plane receiver, pointing to the right and parallel to the ground plane; when not measuring, it is in a retracted state and does not obstruct the normal operation of the receiver receiving and concentrating solar energy; the CCD camera is mounted on the dish concentrator and perpendicular to the focal plane, acquiring an image of the focused spot on the surface of the Lambertian target located on the focal plane; the focused spot image is transmitted to the processor for grayscale processing, and then the centroid coordinates of the focused spot are calculated by weighting the grayscale values; the focused spot is measured once at intervals T1 according to the above process to determine the centroid coordinates, and finally the time-series data of the centroid coordinates changing over time is obtained. In the field measurement of the focused spot, a flip-up Lambertian target is used in conjunction with the CCD camera to intermittently acquire the spot, which does not affect the normal operation of the dish concentrating solar collector system or the concentrating solar power generation system. It can perform autonomous measurement and tracking correction in service, and has the advantages of high efficiency, flexibility and low cost.

Claims

1. A method for measuring and correcting the azimuth axis tilt of a dish concentrator based on spot timing, characterized in that: The implementation steps include the following: Step 1: Establish a sun-tracking model for a dish concentrator that includes azimuth axis tilt error, and determine the actual focal axis vector at any given time: Establish a global coordinate system O- at the intersection point O of the ideal azimuth axis and the ground plane. xyz , where + y The axis points due south. x The axis points due east, + z The axis points to the zenith and coincides with the ideal azimuth axis; astronomical algorithms are used to determine any date of the year in the area where the disc concentrator is to be constructed. t The true position of the sun at any given moment is determined by the solar altitude angle. β With solar azimuth α Composition; using radial error angle θ 1 and the circumferential error angle φ1 describe the magnitude and direction of the tilt angle of the azimuth axis, respectively; t The actual focal axis vector N of the disc condenser at that moment ft for: N ft =N f ⋅Rot(N a1 , α )⋅Rot(N et , β ); In the formula, the function Rot(e, β 2 ) is a rotation function matrix used to realize any vector e = [e x ,e y ,e z Rotate around any unit vector by any angle β 2 The functions are as follows: N a1 It is the actual azimuth axis vector, N a1 =N a ⋅Rot(E −x , θ 1)⋅Rot(E −z ,φ1); N et It is the actual height axis vector, N et =N e ⋅Rot(N a1 , α );N f =[-1,0,0]、N a =[0,0,-1] and N e =[0,1,0] represent the focal axis vector, azimuth axis vector, and height axis vector of the disc concentrator under ideal operating conditions without column tilt error, and are all values ​​when the focal axis vector points due east at the start of daily tracking operation of the disc concentrator; vector E −x =[-1,0,0], Vector E −z =[0,0,-1]; Step 2: Calculate the results from Step 1 t The tracking error value of the disc condenser at any given time, including the height tracking error angle of the disc condenser. β error and azimuth tracking error angle α error The two tracking error components are as follows: β error = β ′− β α error = α ′− α ； In the formula, β 'and α ′ represents the actual elevation and azimuth angles tracked by the disc concentrator, respectively. β ′= -arccos(N) ft E z Vector E z =[0,0,1]; Azimuth α ′ can be regarded as the actual focal axis vector N ft Projection vector N1 along the ground plane = [ x 1, y [1, 0] and - x The included angle of the axis, ; Step 3: Set the aperture radius of the parabolic reflector of the disc condenser. R and focal length f Based on the two tracking error component values ​​determined in step 2, the Monte Carlo ray tracing method is used to calculate the focusing energy flux density distribution of the planar receiver located on the focal plane of the dish condenser, and the centroid coordinates of the focusing energy flux distribution are calculated. x c and y c ; Step 4: Repeat steps 1 to 3 to calculate the centroid coordinates of the focusing energy flow distribution on the focal plane at any time during the tracking process throughout the year under different azimuth axis tilt errors, forming a database of the changes in centroid coordinate values ​​with the time of day for different dates and column tilt error values; then, use a neural network prediction algorithm to train this database and obtain a prediction model, so that by inputting the time series data of the changes in centroid coordinates with the time of day for a certain day, the azimuth axis tilt error value of the disc concentrator can be predicted; Step 5: Select a day M in the whole year, use a CCD camera to measure the focusing energy flow distribution of the focal plane during the open-loop tracking operation of the disc condenser, and determine the centroid coordinates of the focusing energy flow distribution at different times to obtain time series data of the centroid coordinates changing over time. Step 6: Input the centroid coordinate time series data obtained in Step 5 into the prediction model established in Step 4, and set the same date in the prediction model to predict the tilt error value of the actual azimuth axis of the disc concentrator, including the error angles describing the magnitude and direction of the tilt angle of the azimuth axis. θ 2 and error angle ; Step 7: Based on the azimuth tilt error value predicted in Step 6 θ 2 and The apparent solar tracking correction model for the dish concentrator is established as follows: N f ⋅Rot(N a2 , α 1)⋅Rot(N et1 , β 1)=N s ⋅Rot(E −z , α )⋅Rot(E y , β ); In the formula, E y =[0,1,0] and N s =[-1,0,0];N a2 =N a ⋅Rot(E −x ,θ2)⋅Rot(E −z , );N et1 =N e ⋅Rot(N a2 , α The rotation angle is obtained by solving the problem. β 1 and α 1, these are any of the steps in step 1 after tracking and correction. t The rotation angles of the height axis and azimuth axis in the motor-driven dual-axis tracking mechanism at any given time.

2. The method for measuring and correcting the azimuth axis tilt of a dish concentrator based on spot timing according to claim 1, characterized in that: The specific process of step 3 is as follows: The planar receiver located on the focal plane is a square with a side length of L. Its center point p coincides with the focal point of the disc condenser, and a p-... is established on the receiver plane with point p as the origin. x p y p coordinate system x p The axis is parallel to the bottom edge of the planar receiver, pointing to the right and parallel to the ground plane; the planar receiver is divided into two equal parts along the perpendicular directions. H Shares and U portions, and H = U Discretized into H × U A small square grid; The centroid coordinates of the focused energy flux distribution were calculated. x c and y c They are respectively: ； In the formula, h and u These are the horizontal and vertical numbers of the discrete small square grid; The horizontal and vertical numbering are respectively h and u Energy flux density of a discrete grid.

3. The method for measuring and correcting the azimuth axis tilt of a dish concentrator based on spot timing as described in claim 1, characterized in that: The date selected in step 5 is either the summer solstice or the winter solstice.

4. The method for measuring and correcting the azimuth axis tilt of a dish concentrator based on spot timing as described in claim 1, characterized in that: Both steps 4 and 5 involve obtaining the focused energy flow distribution and centroid coordinates at the same starting point and time interval T1.

5. The method for measuring and correcting the azimuth axis tilt of a dish concentrator based on spot timing according to claim 1, characterized in that: Step 5 involves installing a rotatable, water-cooled square Lambertian target on the absorber of the disc concentrator. When measuring the focusing energy flow distribution, the Lambertian target is rotated to the focal plane position, with its center coinciding with the focal point. A local coordinate system q- is established within the Lambertian target plane with the center point q. x q y q To determine the centroid coordinates, where x q The axis is parallel to the bottom edge of the plane receiver, pointing to the right and parallel to the ground plane; when not measuring, it is in a retracted state and does not obstruct the normal operation of the absorber receiving and concentrating solar energy; the CCD camera is mounted on the disc concentrator and perpendicular to the focal plane, acquiring an image of the focused spot on the surface of the Lambertian target located on the focal plane; the focused spot image is transmitted to the processor for grayscale processing, and then the centroid coordinates of the focused spot weighted by grayscale values ​​are calculated; the focused spot is measured once at intervals T1 according to the above process and the centroid coordinates are determined, finally obtaining the time series data of the centroid coordinates changing with time.

Images