CN104729430B - A kind of tower type solar energy thermal power generation heliostat surface testing method - Google Patents

Abstract

A method for detecting the surface shape of a heliostat for tower-type solar thermal power generation. The industrial computer (4) controls the heliostat (5) to be measured, so that the CCD camera (2) can collect the composite fringe image screen reflected by the heliostat (5) (3) of the deformed image, the computer (1) controls the CCD camera (2) to collect the deformed composite fringe image reflected by the heliostat (5), and combines the constructed virtual reference plane phase distribution and the mirror slope and phase mapping model to obtain the mirror surface slope distribution. Based on the generalized Hermite interpolation method, the surface shape of the heliotrope (5) to be measured can be obtained.

Description

A method for detecting the surface shape of a heliostat for tower-type solar thermal power generation

technical field

The invention relates to a method for detecting the surface shape of a heliostat.

Background technique

Tower-type solar thermal power generation has the advantages of high concentration multiple, high working temperature, short heat transfer distance, less heat loss, and high overall system efficiency. Its output characteristics are the closest to conventional thermal power generation, and it is the most promising development of solar thermal power generation. one of the routes. Tower-type solar thermal power generation is mainly composed of three parts: heliostat concentrating system, heat absorption and heat transfer system, and power generation system. The heliostat concentrating system consists of a large number of heliostats that condense and reflect sunlight to the heat absorber located on the tower, through which the light energy is converted into heat energy, and then power is generated through a thermal cycle. The heliostat is the core component of the tower solar thermal power station, accounting for 40-50% of the investment cost of the station, and is the carrier to realize the efficient light-to-heat conversion of the system. The main characteristics of heliostats used in tower solar thermal power plants are: 1) The opening area is relatively large, ranging from tens of square meters to hundreds of square meters; 2) The number of heliostats in the mirror field is huge (tens of thousands, even tens of 10,000 surfaces), and far away from the heat sink, the focal length of the heliostat is relatively large (from tens of meters to thousands of meters away, which requires high concentration accuracy); 3) The heliostat is formed by splicing multi-faceted unit mirrors The complex curved surface gathers and reflects sunlight to the heat sink, and the overall surface shape requires high requirements. Therefore, the high-precision mirror surface shape is the premise to ensure that the sunlight collected by the heliostat is accurately reflected and reaches the heat sink. However, on the one hand, there is a certain deviation between the actual overall surface shape formed by splicing unit mirrors during the installation process of the heliostat and the theoretical design surface shape; Due to gravity, temperature and other factors, the mirror surface is deformed, so that the mirror surface of the heliostat cannot form an ideal curved surface. The resulting slight deviation of the mirror surface shape makes the reflected solar radiation deviate from the daylight opening of the heat absorber, reducing the concentration efficiency of the mirror field. Due to the above error factors, the focused and reflected light spot is shifted, or even deviated from the heat absorber, which affects the safety of the tower and reduces the light concentration efficiency. Therefore, fast heliostat mirror surface shape detection is the key to ensure the concentrating effect, improve the installation efficiency of the mirror field, reduce the cost, improve the concentrating efficiency of the mirror field and the operation efficiency of the power station. At present, this kind of specular surface shape detection usually adopts three-coordinate machine, laser radar, photogrammetry and fringe reflection method. Compared with the first three measurement methods, the fringe reflection method has the advantages of high sensitivity, fast detection, and simple system, and is suitable for detecting various mirror surfaces. The document "S.Ulmer, et.al, Solar Energy, 2011, 85(4) 681-687" used the fringe reflection method to detect the surface shape of the 39.6 square meter heliostat in the Spanish PSA power station, and obtained good measurement results. However, this method uses a four-step phase-shift method to project horizontal and vertical fringe images to the white screen on the tower with a projector arranged in the mirror field, at least sixteen fringe images are projected, and the real-time performance of detection is low. , while the entire measurement process can only be carried out at night and is easily affected by external light sources. Therefore, an efficient, fast, accurate and simple detection method is needed to detect the surface shape of the heliostat.

Contents of the invention

The object of the present invention is to overcome the deficiencies in the prior art and provide an efficient, fast, accurate and simple detection method for the surface shape of the tower-type solar thermal power generation heliostat.

The method for detecting the surface shape of a heliostat for tower-type solar thermal power generation of the present invention comprises the following steps:

1. Install the composite fringe image screen under the heat absorber port on the heat absorbing tower, adjust the positions of the heliostat to be measured and the CCD camera on the heat absorbing tower, so that the CCD camera can observe the deformation composite of the heliotrope to be measured. Stripe image; the composite stripe image screen is a diffuse reflection screen, and the composite stripe image presented has the same stripe frequency in the horizontal direction and the vertical direction;

2. Use the CCD camera to collect a deformed composite fringe image reflected by the heliostat to be measured, and transmit the image signal to the computer;

3. Based on the window Fourier filter and mass-guided phase unwrapping method, the phase distribution in the horizontal and vertical directions of the deformed composite fringe image reflected by the heliotrope to be measured is obtained. Use the Zernike polynomial to obtain the phase distribution of the virtual reference plane with the same area as the heliodon to be measured, and obtain the phase offset caused by the surface shape of the heliodon to be measured according to the phase distribution of the virtual reference plane and the heliodon to be measured;

4. Based on the spatial relative positions of the CCD camera, heliostat, and composite fringe image screen, establish a mathematical relationship model between the phase offset and the mirror slope caused by the heliostat mirror, and combine the phase caused by the heliostat mirror in step 3 The slope distribution of the heliostat mirror surface in both horizontal and vertical directions is obtained by offset;

5. Use the generalized Hermite interpolation algorithm to reconstruct the 3D surface shape of the heliostat.

The detection system of the present invention includes a CCD camera, a composite fringe image screen, a heliostat to be measured, a computer and the like. The composite stripe image screen is placed under the heat-absorbing port, the heliostat to be measured in the mirror field is controlled by the industrial computer, and the CCD camera installed on the top of the heat-absorbing tower of the tower solar thermal power generation system is controlled by the computer to collect the reflection of the heliodon to be measured Deformed composite fringe image of . Compared with existing detection methods, the present invention has the following advantages:

1. The present invention uses a composite fringe image screen, and obtains the phase distribution of the heliostat in the horizontal and vertical directions based on the window Fourier filter and the mass-guided phase unwrapping method. Only one image can be collected to obtain the heliostat Compared with the existing detection method based on the four-step phase shift, the phase offset of the mirror surface has the characteristics of high real-time performance, strong anti-interference, and low environmental influence;

2. The present invention uses Zernike polynomials to obtain the phase distribution of the virtual reference plane. Due to the large mirror area of the heliostat to be measured, it is difficult to obtain the phase distribution of a standard plane mirror with the same area as the heliostat to be measured. Therefore, the phase distribution of the composite fringe image screen is obtained first, and the phase distribution of the heliostat with the same area as the heliostat is obtained by combining the Zernike polynomial. The phase distribution of the virtual reference plane; compared with the existing fringe reflection detection method, the flexibility of the present invention is higher.

3. The present invention uses window Fourier filtering to extract the phase of the deformed composite fringe image reflected by the mirror surface. Compared with the traditional Fourier transform method, it has stronger anti-interference, and the acquisition is in both horizontal and vertical directions. phase distribution.

4. The present invention carries out three-dimensional reconstruction on the surface shape of the heliostat based on the generalized Hermite interpolation algorithm. The reconstruction process from gradient or normal vector to 3D surface shape is mainly based on integral calculation. Ideally, the integration has nothing to do with the path, but in the actual measurement process, due to the influence of detection noise and system deviation, the measured gradient is a non-conservative field, and the selection of the integration path and integration algorithm will directly lead to different reconstruction results. Therefore, for such large amount of data, noise, and under-sampled gradient discrete data, the generalized Hermite interpolation algorithm is used for 3D surface reconstruction, which can ensure both local and global reconstruction accuracy.

Description of drawings

Fig. 1 is a schematic structural diagram of a detection system applying the heliostat surface shape detection method for solar thermal power generation of the present invention;

Fig. 2 is the composite fringe image of the method for detecting the surface shape of a heliostat for solar thermal power generation according to the present invention.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

As shown in FIG. 1 , the detection system based on the method for detecting the surface shape of a heliostat for solar thermal power generation in the present invention includes a CCD camera 2 , a composite fringe image screen 3 , an industrial computer 4 and a computer 1 . The computer 1 controls the CCD camera 2 to collect images, and processes the acquired images; the CCD camera 2 is placed on the top of the heat absorption tower; the composite stripe image screen 3 is installed on the heat absorption tower; the industrial computer 4. Control the operation of the heliostat 5 to be measured, so that the CCD camera 2 can collect the deformed image of the composite fringe image screen 3 reflected by the heliostat 5 .

The method for detecting the surface shape of a heliostat for solar thermal power generation of the present invention comprises the following steps:

1. Install the composite fringe image screen 3 below the heat absorber port on the heat absorbing tower, adjust the positions of the heliostat 5 to be measured and the CCD camera 2, so that the CCD camera 2 can observe the deformation compounding reflected by the heliodon 5 to be measured. stripe image;

2. Based on the window Fourier filter and mass-guided phase unwrapping method, the phase distribution in the horizontal and vertical directions of the deformed composite fringe image reflected by the heliotrope to be measured is obtained. Using the Zernike polynomial to obtain the phase distribution of the virtual reference plane with the same area as the heliodon to be measured, and according to the phase distribution of the virtual reference plane and the heliodon to be measured, the phase offset caused by the surface shape of the heliodon to be measured is obtained.

The composite fringe image expression presented by the composite fringe image screen 3 is:

Where (x, y) is the coordinates of the composite fringe image screen 3, p _x and p _y are the fringe period in the horizontal direction and the vertical direction respectively, I ₀ (x, y) is the fringe intensity, and A(x, y) is Background light intensity, B ₁ (x,y) and B ₂ (x,y) are the amplitudes of the stripes in the horizontal direction and vertical direction, respectively.

Obtain the wrapping phase of the composite fringe image presented by this composite fringe image screen 3 based on the window Fourier filtering method:

First, window Fourier transform is performed on expression (1), and the transformation is as follows:

in Sf(u,v,ξ,η) is the Fourier transform of I ₀ (x,y), (u,v) is the space coordinate, (ξ,η) is the frequency domain coordinate, the window function g(x,y ) is a Gaussian function:

σ _x and σ _y are the standard deviations of the Gaussian function in the x and y directions, respectively.

Secondly, the formula (2) is filtered:

where θ is the threshold.

At the same time, the fundamental frequency components in the horizontal and vertical directions of the filtered spectrum are respectively selected. Therefore, the wrapping phase of the composite fringe image in the horizontal and vertical directions They are:

in, with are respectively the wrapping phases of x and y upwards, (x, y) are space coordinates, with respectively the fundamental frequency components in the x and y directions, with respectively with The inverse Fourier transform of , g(x, y) is the Gaussian function in formula (2).

Finally, the wrapped phase is obtained based on the mass-guided phase unwrapping algorithm with The phase distributions φ _0x (x,y) and φ _0y (x,y) of .

Express the above continuous phase distributions φ _0x (x, y) and φ _0y (x, y) as linear combination expressions of Zernike polynomials:

Where z _n is the nth Zernike polynomial, is the nth Zernike polynomial coefficient.

Solve the coefficients based on the least squares algorithm Use this coefficient to construct the phase distribution φ ₀ ' _x (x,y) and φ ₀ ' _y (x,y) of the virtual plane.

3. Based on the spatial relative positions of the CCD camera, heliostat, and composite fringe image screen, establish a mathematical relationship model between the phase offset caused by the heliostat mirror and the slope of the mirror, and combine the phase caused by the heliostat mirror in step 3 Offset obtains the slope distribution of the heliostat mirror surface in both horizontal and vertical directions.

The industrial computer 4 controls the heliostat 5 to be measured, so that the CCD camera, 2 can collect the deformed image of the composite fringe image screen 3 reflected by the heliostat 5 . The computer 1 controls the CCD camera to collect the deformed composite fringe image reflected by the heliostat 5, and the expression of the deformed composite fringe image is as follows:

I ₁ (x,y)＝a(x,y)+b ₁ (x,y)cos[φ _x (x,y)]+b ₂ (x,y)cos[φ _y (x,y)] (7)

Where I ₁ (x, y) is the light intensity distribution of the fringe image, a (x, y) is the background light intensity, b ₁ (x, y) and b ₂ (x, y) represent the horizontal and vertical directions respectively The degree of modulation of , φ _x (x, y) and φ _y (x, y) represent the phases modulated by the surface shape of the heliostat 5 to be measured in the horizontal and vertical directions, respectively.

Using formula (7) to use the calculation process of formula (2) to formula (5) and the mass-guided phase unwrapping algorithm to obtain the phase φ _1x (x, y) and φ _1y ( x,y).

The detection system shown in Figure 1 shows that the relationship between the slope of the heliostat mirror and the phase is:

Where α is the acquisition angle of the CCD camera, Slope _x is the slope distribution of the mirror surface in the horizontal direction, and Slope _y is the slope distribution of the mirror surface in the vertical direction.

Based on the obtained mirror phases φ _1x (x,y) and φ _1y (x,y) and the phases of the virtual plane φ ₀ ' _x (x,y) and φ ₀ ' _y (x,y), combined with formula (8) The slope distributions Slope _x and Slope _y of the mirror surface in the horizontal and vertical directions can be obtained.

4. Using the generalized Hermite interpolation algorithm to reconstruct the three-dimensional surface shape of the heliostat.

The generalized Hermite interpolation algorithm based on radial basis function is used to reconstruct the slope of the acquired heliotrope to be measured in three dimensions. Define the interpolation method as follows:

Where X=(x,y) ^T , α _i and β _i (1≤i≤N) are unknown coefficients, ψ:R ² →R is radial basis function, ψ _x and ψ _y are radial basis function ψ Partial derivatives in the x and y directions, s(X) is the mirror shape function of the day to be determined. Establish the matching relationship between the analytical derivative of this interpolation method and the measured slope data Slope _x and Slope _y :

Where j is the jth sampling point, that is, to solve the following linear equation:

Among them, ψ _xx , ψ _xy , and ψ _yy are the second-order partial derivatives of ψ with respect to x, the second-order mixed partial derivatives with respect to x, y, and the second-order partial derivatives with respect to y; the coefficient α calculated based on formula (11) _i and β _i , apply them to the interpolation of formula (9) to reconstruct the surface shape, and obtain the three-dimensional surface shape s(X) of the heliotropic mirror surface to be measured.

Claims (1)

1. a heliostat surface shape detection method for tower type solar thermal power generation, described detection method may further comprise the steps: (1) Use the CCD camera to collect the deformed composite fringe image of the heliotrope to be measured, and store it in the computer, and then combine the window Fourier filter method and the mass-guided phase unwrapping method to obtain the deformed composite fringe image of the heliotrope to be measured. Phase distribution in horizontal and vertical directions; (2) Use the Zernike polynomial to obtain the phase distribution of the virtual reference plane with the same area as the heliodon to be measured, and obtain the phase offset caused by the surface shape of the heliodon to be measured according to the phase distribution of the virtual reference plane and the heliodon to be measured; (3) According to the spatial relative positions of the CCD camera in the detection system, the heliodon to be measured and the composite fringe image screen, establish the mathematical relationship model between the phase shift and the mirror slope caused by the heliotrope to be measured, and combine steps (1) and The phase distribution in (2) obtains the slope distribution of the heliostat mirror surface in the horizontal direction and the vertical direction; (4) Based on the generalized Hermite interpolation method and the slope distribution of the heliostat in step (3), obtain the surface shape reconstruction of the heliostat to be measured; In described step (2), utilize Zernike polynomial to obtain and the method for the phase distribution of the virtual reference plane of equal area of the heliodon to be measured is as follows: Based on the window Fourier filtering method, obtain the package phase distribution φ _0x (x, y) and φ _0y (x, y) of the composite fringe image presented by the composite fringe image screen (3), and express this phase distribution as a Zernike polynomial The linear combination expression of : <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <msub> <mi>&amp;phi;</mi> <mrow> <mn>0</mn> <mi>x</mi> </mrow> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>n</mi> <mo>=</mo> <mn>0</mn> </mrow> <mi>&amp;infin;</mi> </munderover> <msubsup> <mi>k</mi> <mi>n</mi> <mn>1</mn> </msubsup> <msub> <mi>z</mi> <mi>n</mi> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>=</mo> <msubsup> <mi>k</mi> <mn>0</mn> <mn>1</mn> </msubsup> <mo>+</mo> <msubsup> <mi>k</mi> <mn>1</mn> <mn>1</mn> </msubsup> <msub> <mi>z</mi> <mn>1</mn> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>+</mo> <mn>...</mn> <mo>+</mo> <msubsup> <mi>k</mi> <mi>n</mi> <mn>1</mn> </msubsup> <msub> <mi>z</mi> <mi>n</mi> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mtd> </mtr> <mtr> <mtd> <msub> <mi>&amp;phi;</mi> <mrow> <mn>0</mn> <mi>y</mi> </mrow> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>n</mi> <mo>=</mo> <mn>0</mn> </mrow> <mi>&amp;infin;</mi> </munderover> <msubsup> <mi>k</mi> <mi>n</mi> <mn>2</mn> </msubsup> <msub> <mi>z</mi> <mi>n</mi> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>=</mo> <msubsup> <mi>k</mi> <mn>0</mn> <mn>2</mn> </msubsup> <mo>+</mo> <msubsup> <mi>k</mi> <mn>1</mn> <mn>2</mn> </msubsup> <msub> <mi>z</mi> <mn>1</mn> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>+</mo> <mn>...</mn> <mo>+</mo> <msubsup> <mi>k</mi> <mi>n</mi> <mn>2</mn> </msubsup> <msub> <mi>z</mi> <mi>n</mi> </msub> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mtd> </mtr> </mtable> </mfenced> Where z _n is the nth Zernike polynomial, Be the nth Zernike polynomial coefficient; (x, y) are the coordinates of the composite fringe image screen; Solve the coefficients based on the least squares algorithm Use this coefficient to construct the phase distribution φ' _0x (x,y) and φ' _0y (x,y) of the virtual plane; Described step (4) is based on generalized Hermite interpolation and heliostat slope distribution and reconstructs the method for the three-dimensional surface shape of the heliostat mirror surface to be measured as follows: Obtained by described step (3) the slope of the heliostat mirror surface to be measured in the horizontal direction and the vertical direction is respectively Slope _x and Slope _y , and the Hermite interpolation function is defined as: <mrow> <mi>s</mi> <mrow> <mo>(</mo> <mi>X</mi> <mo>)</mo> </mrow> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <msub> <mi>&amp;alpha;</mi> <mi>i</mi> </msub> <msub> <mi>&amp;psi;</mi> <mi>x</mi> </msub> <mrow> <mo>(</mo> <mi>X</mi> <mo>-</mo> <msub> <mi>X</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>+</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <msub> <mi>&amp;beta;</mi> <mi>i</mi> </msub> <msub> <mi>&amp;psi;</mi> <mi>y</mi> </msub> <mrow> <mo>(</mo> <mi>X</mi> <mo>-</mo> <msub> <mi>X</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> </mrow> Where X=(x,y) ^T , α _i and β _i (1≤i≤N) are unknown coefficients, ψ:R ² →R is radial basis function, ψ _x and ψ _y are the partial derivatives of the radial basis function ψ in the x and y directions, respectively, s(X) is the specular surface function to be measured; (x, y) is the coordinates of the compound fringe image screen; Establish the matching relationship between the analytical derivative of this interpolation method and the measured slope data Slope _x and Slope _y : <mrow> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <msub> <mi>s</mi> <mi>x</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>X</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>Slope</mi> <mi>x</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>X</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>s</mi> <mi>y</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>X</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>Slope</mi> <mi>y</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>X</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> <mn>1</mn> <mo>&amp;le;</mo> <mi>j</mi> <mo>&amp;le;</mo> <mi>N</mi> </mrow> Where j is the jth sampling point; That is, solve the following linear equation: Among them, ψ _xx , ψ _xy , and ψ _yy are the second-order partial derivatives of ψ to x, the second-order mixed partial derivatives to x, y, and the second-order partial derivatives to y; Based on the obtained coefficients α _i and β _i , the three-dimensional surface shape s(X) of the heliotrope to be measured is obtained.