CN102930160B - Determine the method for tower type solar heat and power system mirror field optics efficiency - Google Patents

Abstract

The invention discloses a kind of computing method of tower type solar heat and power system mirror field optics efficiency, implementation step is as follows: 1) generate Jing Chang, determine incident ray; 2) determine that scope is spread in throwing according to the coordinate of described Jing Chang, incident ray and heliostat; 3) in the heliostat field that scope is spread in described throwing, random throwing spreads luminous point; 4) utilize CUDA computing platform to adopt multithreading to carry out ray tracing, the shade being calculated Jing Chang by Ray-tracing Method is blocked efficiency and overflows efficiency; 5) cosine efficiency is calculated; 6) shade blocked efficiency, overflow efficiency, cosine efficiency three computing and obtain mirror field optics efficiency and export.The present invention can utilize the high-speed computation ability of CUDA computing platform to realize the computing of GPU high performance parallel, multiple thread is allowed to calculate simultaneously and to judge, the spended time of whole program computation mirror field efficiency is few, be conducive to optimization and the design of Jing Chang, there is the advantage that counting yield is high, computing velocity fast, mirror field is applied widely.

Description

Determine the method for tower type solar heat and power system mirror field optics efficiency

Technical field

The present invention relates to tower type solar heat and power system field, be specifically related to a kind of method determining tower type solar heat and power system mirror field optics efficiency.

Background technology

Tower type solar heat and power system is the large-scale concentrating to generate power mode of a kind of large area, tower type solar heat and power system forms primarily of heliostat, tower, heat dump, heat-exchanger rig, heat-storing device and thermal electric generator, its principle of work is that heat dump sunshine being converged to tower top by the heliostat of some produces high temperature, to the dielectric heating in heat dump be flowed through again, and produce high-temperature steam pushing turbine and generate electricity.Owing to there is the advantage of focusing ratio large (generally can reach 300 ~ 1500), running temperature high (can at 500 DEG C ~ 1500 DEG C), therefore tower type solar heat and power system obtains at present and applies more widely.Condenser system is the key components of tower-type solar thermal power generating system, be conducive to optimization and the design of Jing Chang by calculating the optical efficiency (cosine efficiency, shade block efficiency, overflow efficiency) obtaining Jing Chang, thus the efficiency of overall cost and raising generating can be reduced.

The index weighing tower type solar heat and power system mirror field optics efficiency mainly comprises: cosine efficiency, shade block efficiency, overflow efficiency, atmospheric transmission efficiency.Due to relative position between atmospheric transmission efficiency and heliostat have nothing to do, more easily ask for, therefore general calculate weigh tower type solar heat and power system mirror field optics efficiency time first can not consider atmospheric transmission efficiency.Cosine efficiency, shade block efficiency, it is as follows to overflow the concrete meaning of efficiency:

1) cosine efficiency: when solar irradiation is mapped to heliostat surface, meeting and mirror surfaces produce certain angle, then the cosine value of the angle of the normal vector of the incident vector sum heliostat of incident ray is defined as cosine efficiency.As shown in Figure 1, θ angle is cosine angle.In general, cosine efficiency causes the inevitable main cause of heliostat field loss in efficiency.

2) shade blocks efficiency: as shown in Figure 2, and when incident ray is irradiated to target heliostat Mir1, is blocked, cause shadow loss by heliostat Mir2, is shade efficiency by the amount of light of shade and the ratio of the total amount of light shining this heliostat; In like manner, then cause eclipsing loss when reflection ray is blocked by heliostat Mir3, the amount of light be blocked is eclipsing loss with the ratio of the total amount of light shining this heliostat.Mirror may occur by shade and situation about being blocked simultaneously.

3) overflow efficiency: when reflection ray does not shine in heat dump owing to being subject to the restriction of heat dump size, claim these light to be the light overflowed, by the light that overflows and the ratio of total light number for overflow efficiency.

In prior art, the computing method of mirror field optics efficiency have by heliostat is carried out gridding, by judging each small grid successively whether by the shade of heliostat generation around with block; In addition, also have and the apex coordinate of mirror field heliostat is projected by the plane calculating heliostat along incident or reflection ray, tried to achieve by coordinate transform and calculated area percentage that heliostat is blocked thus try to achieve efficiency.But the said method computation process versus busy of prior art, efficiency are low, when calculation overflow efficiency, difficulty is large, is especially doubled and redoubled along with the expansion of mirror field scale computing time.Prior art also favourable Ray-tracing Method calculates the method for mirror field optics efficiency, the advantage of this method be by follow the trail of throw the light be sprinkling upon in mirror field can be visual and clear judge the position relationship of each root light and each heliostat, and finally can calculation overflow efficiency easily, obtain the schematic diagram that is heated of heat dump, but the shortcoming that prior art utilizes Ray-tracing Method to calculate the method for mirror field optics efficiency is then the factors such as calculated amount also can increase along with the mirror quantity of Jing Chang, the enlarged areas of Jing Chang and increases simultaneously.

Summary of the invention

The technical problem to be solved in the present invention is to provide the method for the determination tower-type solar thermal electric system mirror field optics efficiency that a kind of counting yield is high, computing velocity fast, mirror field is applied widely.

For solving the problems of the technologies described above, the technical solution used in the present invention is:

Determine a method for tower type solar heat and power system mirror field optics efficiency, implementation step is as follows:

1) generate Jing Chang, determine incident ray;

2) determine that scope is spread in throwing according to the coordinate of described Jing Chang, incident ray and heliostat;

3) in the heliostat field that scope is spread in described throwing, random throwing spreads luminous point;

4) CUDA computing platform is utilized to adopt the form of multithreading to carry out ray tracing for each root light, four summits being sprinkling upon ground luminous point and heliostat to determine incident ray and each heliostat successively crossing situation along incident ray projection coordinate is on the ground thrown by random, if the light of spot projection not in any heliostat corresponding to this luminous point is invalid incident ray, then consider next luminous point; Otherwise judge this incident ray not by the heliostat that shade is also finally irradiated to, and ask for the intersection point with this heliostat; Then enter and block decision stage; First calculate this reflection ray corresponding to root incident ray, judge reflection ray whether block by other mirror, if be not blocked, ask for the intersection point of the final and absorber of reflection ray, judge this intersection point whether in absorber, if intersection point is not in absorber, judge that light overflows; Finally blocked efficiency according in all incident raies by the shade that shade and the incident ray quantity be blocked calculate Jing Chang, the incident ray quantity that in all incident raies, light overflows calculates the spilling efficiency of Jing Chang;

5) cosine efficiency is calculated;

6) described shade blocked efficiency, overflow efficiency, cosine efficiency three and carry out multiplying, obtain final mirror field optics efficiency and export.

Further improvements in methods as determination tower type solar heat and power system mirror field optics efficiency of the present invention:

Described step 1) in generate the detailed step of Jing Chang as follows: determine mirror field parameters, heat dump size, sun altitude and solar azimuth, described mirror field parameters comprises heliostat quantity and size, mirror field type of arrangement; The centre coordinate O (x, y, z) calculating heliostat in mirror field generates Jing Chang.

Described step 1) in determine that the detailed step of incident ray is as follows:

Assuming that incident ray is directional light, minute surface is pointed to by the sun in direction, and incident ray in the projection components of X-axis, Y-axis, Z axis respectively such as formula shown in (1):

a＝-cos(h _s)*sin(p _s)

b＝cos(h _s)*cos(p _s) (1)

c＝-sin(h _s)

In formula (1), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis, h _sfor sun altitude, p _sfor solar azimuth.

Described step 2) in calculate the detailed step of coordinate of heliostat as follows:

2.1) the reflection vector of heliostat and planar process vector is calculated;

2.2) apex coordinate of heliostat is calculated;

2.3) heliostat center and summit is calculated along incident ray projection coordinate on the ground;

2.4) dynamically follow the tracks of heliostat summit and centre coordinate along incident light view field on the ground, determine that scope is spread in throwing.

Described step 2.1) specifically refer to that through type (2) ~ (3) calculate the reflection vector of heliostat and planar process vector;

The reflection vector (rox, roy, roz) of any heliostat is such as formula shown in (2):

$\left\{\begin{array}{c}R=\sqrt{x^2+y^2+{(z-H)}^2}\\ rox=x/R\\ roy=y/R\\ roz=(z-H)/R\end{array}\right.---\left(2\right)$

In formula (2), R is the mould of heliostat center to heat dump center vector; H is the centre-height of heat dump, i.e. (0,0, the H) centre coordinate that is heat dump; The centre coordinate that (x, y, z) is heliostat;

The normal vector (nx, ny, nz) of any heliostat is such as formula shown in (3):

$\left\{\begin{array}{c}NR=\sqrt{{(a+rox)}^2+{(b+roy)}^2+{(c+roz)}^2}\\ nx=(a+rox)/NR\\ ny=(b+roy)/NR\\ nz=(c+roz)/NR\end{array}\right.---\left(3\right)$

In formula (3), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; NR is the mould of minute surface normal vector; H is the centre-height of heat dump, i.e. (0,0, the H) centre coordinate that is heat dump; (rox, roy, roz) is reflection vector;

Described step 2.2) specifically refer to that through type (4) ~ (7) calculate the apex coordinate of heliostat;

$P_1\left\{\begin{array}{c}{x}_{p1}=x+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p1}=y+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p1}=z-\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(4\right)$

$P_2\left\{\begin{array}{c}{x}_{p2}=x-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p2}=y-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p2}=z+\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(5\right)$

$P_3\left\{\begin{array}{c}{x}_{p3}=x-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p3}=y-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p3}=z+\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(6\right)$

$P_4\left\{\begin{array}{c}{x}_{p4}=x+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p4}=y+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p4}=z-\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(7\right)$

In formula (4) ~ (7), the centre coordinate that (x, y, z) is heliostat, (nx, ny, nz) is the unit normal vector of this face heliostat, l is the length of side of heliostat and transverse axis parallel edges, and w is the length of side of heliostat and transverse axis vertical edges, (x _pi, y _pi, z _pi) be i-th apex coordinate, i represents four summit sequence numbers respectively, i ∈ [Isosorbide-5-Nitrae].

Described step 2.3) specifically refer to and calculate heliostat center and summit along incident ray projection coordinate on the ground according to formula (8) ~ (9);

$\left\{\begin{array}{c}{x}_{os}=-\left(\frac{a}{b}\right)*z+x\\ y_{os}=-\left(\frac{b}{c}\right)*z+y\end{array}\right.---\left(8\right)$

$\left\{\begin{array}{c}{x}_{psi}=-\left(\frac{a}{b}\right)*z_{pi}+x_{pi}\\ y_{psi}=-\left(\frac{b}{c}\right)*z_{pi}+y_{pi}\end{array}\right.---\left(9\right)$

In formula (8) ~ (9), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; The centre coordinate that (x, y, z) is heliostat, (x _os, y _os) be the center projection coordinate on the ground of heliostat, (x _psi, y _psi) be the summit projection coordinate on the ground of heliostat, (x _pi, y _pi, z _pi) be the coordinate on heliostat i-th summit, i represents four summit sequence numbers respectively, i ∈ [Isosorbide-5-Nitrae].

Described step 3) specifically refer to that utilizing Monte Carlo method to spread random throwing in the heliostat field of scope in described throwing spreads luminous point, and throw and spread luminous point quantity and should try one's best greatly, large to the view field that can be formed completely on the ground with all minute surfaces covering Jing Chang, the expression formula of the coordinate (Xis, Yis) of projected light is on the ground such as formula shown in (10);

Xis＝x _min+(x _max-x _min)*rand(N,1)

(10)

Yis＝y _min+(y _max-y _min)*rand(N,1)

In formula (10), x _min, x _maxbe respectively the minimum and maximum X-coordinate value that heliostat apex coordinate projects on the ground along incident light, y _min, y _maxbe respectively the minimum and maximum Y-coordinate value that heliostat apex coordinate projects on the ground along incident light, N is that spot number is spread in throwing.

Described step 4) detailed step as follows:

4.1) utilize CUDA computing platform to adopt the form of multithreading to carry out each parameter of initialization for each root light, described parameter comprises:

Flags initial value is 0: shade marker for determination, value be 0 explanation light by shade, be that 1 explanation light is not by shade;

Flagb initial value is 1: block marker for determination, and value is that 0 explanation light is blocked, and is that 1 explanation light is not blocked;

M initial value is 0: incident ray impinges upon the mirror position of target heliostat;

Temps initial value is 0: the intersection point Z coordinate (higher value) of incident ray and heliostat;

Xm, Ym, Zm initial value is 0: the intersecting point coordinate of light and heliostat;

Xrec, Zrec initial value is 0: the intersecting point coordinate of light and heat dump;

4.2) according to whether satisfying condition judge that impact point E projects in the parallelogram ABCD of ground formation at mirror, wherein for the vector formed between the summit A of parallelogram ABCD and impact point E, for the vector formed between summit A, summit D of parallelogram ABCD, for the vector formed between summit A, summit B of parallelogram ABCD; If impact point E is in the parallelogram that mirror is formed, redirect performs step 4.3), otherwise redirect performs step 4.6); Wherein, coefficient of determination u and v meets 0≤u, v≤1, and the expression formula of coefficient of determination u, v is such as formula shown in (11) ~ (13);

$\left\{\begin{array}{c}ax=x_B-x_A\\ ay=y_B-y_A\\ bx=x_D-x_A\\ by=y_D-y_A\\ cx=x_E-x_A\\ cy=y_E-y_A\end{array}\right.---\left(11\right)$

$\left\{\begin{array}{c}aa=ax*ax+ay*ay\\ ab=ax*bx+ay*by\\ ac=ax*cx+ay*cy\\ bb=bx*bx+by*by\\ bc=bx*cx+by*cy\end{array}\right.---\left(12\right)$

$\left\{\begin{array}{c}u=\frac{aa*bc-ab*ac}{aa*bb-ab*ab}\\ v=\frac{bb*ac-bc*ab}{aa*bb-ab*ab}\end{array}\right.---\left(13\right)$

In formula (11) ~ (13), x _afor the X-coordinate of ABCD point A, y _afor the Y-coordinate of an A, x _bfor the X-coordinate of ABCD point B, y _bfor the Y-coordinate of a B, x _dfor the X-coordinate of ABCD point D, y _dfor the Y-coordinate of a D, x _efor the X-coordinate of impact point E, y _efor the Y-coordinate of impact point E;

4.3) make the value of Flags be 1, and calculate the intersection point Z coordinate Zm of this incident ray and this heliostat;

4.4) judge whether Zm is greater than Temps, if then redirect performs step 4.5), otherwise redirect performs step 4.6);

4.5) make the value of Temps equal Zm, and record the position of this heliostat, be called target heliostat M; Target heliostat M is the mirror that incident ray is finally irradiated to; The final value of Temps is the intersection point Z coordinate of incident ray and target heliostat M, for making comparisons blocking decision stage;

4.6) judge whether circulation terminates; If then turn redirect to perform step 4.7), otherwise turn redirect execution step 4.2);

4.7) judge whether this root light impinges upon in mirror or, even the value of Flags is that 1 redirect performs step 4.8 on the ground), enter and block decision stage; Otherwise turn redirect and perform step 4.17), start after terminating to judge next luminous point;

4.8) the intersecting point coordinate Xm of the reflection ray corresponding to incident ray and other heliostat except target heliostat M is calculated, Ym, Zm; Blocking decision stage, by the intersecting point coordinate Zm of computational reflect light and surrounding heliostat, compare the size of Z coordinate, if once judged result is Zm be less than Temps, then illustrate that reflection ray is blocked, thus quantitatively record increase by 1 in shutting out the light of target heliostat; When calculating the intersecting point coordinate with other heliostat except target heliostat M herein, reduction judgement scope is carried out to all heliostats, and reduction principle for: line number is reduced to 2 row up and down of being expert at target heliostat, and columns is reduced to and arranges in the left and right 2 of target heliostat column;

4.9) cosine of an angle value formed by the vector at four vectors that rectangle four summits calculating described intersection point and heliostat according to formula (14) are linked to be and rectangle four edges place, formed by the vector judging four vectors that rectangle four summits of intersection point and heliostat are linked to be and rectangle four edges place according to described cosine value, whether angle is all as acute angle, if be all acute angle, judge that this intersection point is in this face heliostat, then redirect performs step 4.10), otherwise redirect performs step 4.12);

cosEAD＝(x _E-x _A)*(x _D-x _A)+(y _E-y _A)*(y _D-y _A)+(z _E-z _A)*(z _D-z _A)

cosEDC＝(x _E-x _D)*(x _C-x _D)+(y _E-y _D)*(y _C-y _D)+(z _E-z _D)*(z _C-z _D)

(14)

cosECB＝(x _E-x _C)*(x _B-x _C)+(y _E-y _C)*(y _B-y _C)+(z _E-z _C)*(z _B-z _C)

cosEBA＝(x _E-x _B)*(x _A-x _B)+(y _E-y _B)*(y _A-y _B)+(z _E-z _B)*(z _A-z _B)

In formula (14), cosine of an angle value formed by the vector at four vectors that rectangle four summits that cosEAD, cosEDC, cosECB, cosEBA are respectively described intersection point and heliostat are linked to be and rectangle four edges place; x _efor the X-coordinate of impact point E, x _afor the X-coordinate of rectangle summit A, x _bfor the X-coordinate of rectangle summit B, x _cfor the X-coordinate of rectangle summit C, x _dfor the X-coordinate of rectangle summit D; y _efor the Y-coordinate of punctuate E, y _afor the Y-coordinate of rectangle summit A, y _bfor the Y-coordinate of rectangle summit B, y _cfor the Y-coordinate of rectangle summit C, y _dfor the Y-coordinate of rectangle summit D; z _efor the Z coordinate of impact point E, z _afor the Z coordinate of rectangle summit A, z _bfor the Z coordinate of rectangle summit B, z _cfor the Z coordinate of rectangle summit C, z _dfor the Z coordinate of rectangle summit D;

4.10) judge whether this intersection point Zm is greater than Temps, if then redirect performs step 4.11), otherwise redirect performs step 4.12);

4.11) illustrate that this root light is blocked, make the value of Flagb be 0;

4.12) judge whether circulation terminates; If then redirect performs step 4.13), otherwise redirect performs step 4.8);

4.13) judge whether this root light is blocked, even the value of Flagb is that 1 redirect performs step 4.14), enter spilling decision stage; Otherwise redirect performs step 4.17);

4.14) intersection point of this root reflection ray and heat dump is calculated;

4.15) this intersection point is judged whether in heat dump, if then redirect performs step 4.16), otherwise redirect performs step 4.17); Meanwhile, if this intersection point is not in heat dump, then the spilling amount of light of this heliostat corresponding to root light increases by 1, otherwise this root light is effective sunlight;

4.16) then record X-coordinate and Z coordinate Xrec, the Zrec of this intersection point, obtained the schematic diagram that is heated of heat dump by the intersecting point coordinate recording all effective sunlights and heat dump;

4.17) terminate, if all light completes as calculated, then blocked efficiency according in all incident raies by the shade that shade and the incident ray quantity be blocked calculate Jing Chang, the incident ray quantity that in all incident raies, light overflows calculates the spilling efficiency of Jing Chang.

Described step 5) in specifically refer to and calculate cosine efficiency according to formula (15);

eCos＝a*nx+b*ny+c*nz (15)

In formula (15), eCos is cosine efficiency, and a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; (nx, ny, nz) is the unit normal vector of this face heliostat, and * is multiplication operator.

The present invention has following advantage: the invention provides the algorithm flow calculating tower type solar heat and power system mirror field optics efficiency based on Ray-tracing Method, set forth in detail simultaneously and how to determine position of sun, generate mirror field coordinate, ask for the apex coordinate of heliostat, unit reflection vector sum normal vector, and dynamically determine that the scope of luminous point is spread in random throwing, first Ray-tracing Method is determined the situation of light and heliostat by luminous point and heliostat summit projection coordinate on the ground, if ray cast is not in any heliostat projection, then consider next light, otherwise ask for the intersection point of this incident ray and this heliostat, then judge whether the reflection ray corresponding with this incident ray is blocked by other mirror, if be not blocked, then ask for the intersection point of the final and absorber of reflection ray, judge this intersection point whether in absorber, after being disposed etc. whole light, the optical efficiency of Jing Chang can be calculated, in order to speed-up computation saves time, invention introduces CUDA computing platform, GPU is utilized to achieve high performance parallel computing.In sum, the present invention can utilize the high-speed computation ability of CUDA computing platform, achieve the computing of GPU high performance parallel, multiple thread is allowed to calculate according to Ray-tracing Method simultaneously and to judge, the spended time of whole program computation mirror field efficiency is few, be conducive to optimization and the design of Jing Chang, there is the advantage that counting yield is high, computing velocity fast, mirror field is applied widely.

Accompanying drawing explanation

Fig. 1 is prior art tower type solar mirror field coordinate schematic diagram, and wherein-X-axis points to due east, and Y-axis level points to due south, and Z axis points to zenith.

Fig. 2 is shade and blocks schematic diagram.

Fig. 3 is the basic procedure schematic diagram of the embodiment of the present invention.

Fig. 4 is that the heliostat twin shaft of the application embodiment of the present invention fixes schematic diagram.

Fig. 5 is embodiment of the present invention step 4) detailed process schematic diagram.

Fig. 6 is the principle schematic of embodiment of the present invention judging point and parallelogram relative position.

Embodiment

As shown in Figure 3, the implementation step of the method for the present embodiment determination tower type solar heat and power system mirror field optics efficiency is as follows:

1) generate Jing Chang, determine incident ray;

2) determine that scope is spread in throwing according to the coordinate of Jing Chang, incident ray and heliostat;

3) in the heliostat field throwing scope of spreading, random throwing spreads luminous point;

4) CUDA computing platform is utilized to adopt the form of multithreading to carry out ray tracing for each root light, four summits being sprinkling upon ground luminous point and heliostat to determine incident ray and each heliostat successively crossing situation along incident ray projection coordinate is on the ground thrown by random, if the light of spot projection not in any heliostat corresponding to this luminous point is invalid incident ray, then consider next luminous point; Otherwise judge this incident ray not by the heliostat that shade is also finally irradiated to, and ask for the intersection point with this heliostat; Then enter and block decision stage; First calculate this reflection ray corresponding to root incident ray, judge reflection ray whether block by other mirror, if be not blocked, ask for the intersection point of the final and absorber of reflection ray, judge this intersection point whether in absorber, if intersection point is not in absorber, judge that light overflows; Finally blocked efficiency according in all incident raies by the shade that shade and the incident ray quantity be blocked calculate Jing Chang, the incident ray quantity that in all incident raies, light overflows calculates the spilling efficiency of Jing Chang;

5) cosine efficiency is calculated;

6) shade blocked efficiency, overflow efficiency, cosine efficiency three and carry out multiplying, obtain final mirror field optics efficiency and export.

In the present embodiment, each thread of CUDA computing platform completes the tracking to a light, record the number of times of this root light and heliostat generation shade, namely which heliostat this root light can to shine in but by shade, record blocking and spilling situation of this root light simultaneously, if block and spilling, then in corresponding statistical variable, numerical value adds one, the high performance parallel computation ability of CUDA computing platform can be made full use of, and the present embodiment can be tried to achieve the comprehensive shade of Jing Chang simultaneously and blocked efficiency and overflow efficiency visual and clearly, and the schematic diagram that is heated of thermoreceptor can be generated according to the intersection point of light and thermoreceptor, be conducive to optimization and the design of Jing Chang.Utilize Monte Carlo method in the present embodiment, after dynamically determining that scope is spread in throwing, throw and spread a large amount of random luminous points.And in order to quick calculating can be completed, under VS2008, introduce CUDA computing platform, realize the computing of GPU high performance parallel, allow multiple thread calculate according to Ray-tracing Method simultaneously and to judge.

In the present embodiment, step 1) in generate the detailed step of Jing Chang as follows: determine mirror field parameters, heat dump size, sun altitude and solar azimuth, mirror field parameters comprises heliostat quantity and size, mirror field type of arrangement; The centre coordinate O (x, y, z) calculating heliostat in mirror field generates Jing Chang.

In the present embodiment, step 1) in determine that the detailed step of incident ray is as follows:

Assuming that incident ray is directional light, minute surface is pointed to by the sun in direction, and incident ray in the projection components of X-axis, Y-axis, Z axis respectively such as formula shown in (1):

a＝-cos(h _s)*sin(p _s)

b＝cos(h _s)*cos(p _s) (1)

c＝-sin(h _s)

In formula (1), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis, h _sfor sun altitude, p _sfor solar azimuth.The vector of unit length of incident ray can according to the elevation angle h of the sun _swith position angle p _suniquely determine, consider in the present embodiment that the position of sun distance earth is very far away, therefore supposition incident ray is directional light, direction is downward, namely points to minute surface by the sun.

In the present embodiment, step 2) in calculate the detailed step of the coordinate of heliostat as follows:

2.1) the reflection vector of heliostat and planar process vector is calculated;

2.2) apex coordinate of heliostat is calculated;

2.3) heliostat center and summit is calculated along incident ray projection coordinate on the ground;

2.4) dynamically follow the tracks of heliostat summit and centre coordinate along incident light view field on the ground, determine that scope is spread in throwing.

The reflection vector of heliostat is by the direction vector of heat dump center to heliostat center, the reflection of heliostat vector is changed into vector of unit length with planar process vector by unification in the present embodiment, and minute surface unit normal vector can be represented by the vector sum of incident ray and reflection ray.In the present embodiment, step 2.1) specifically refer to that through type (2) ~ (3) calculate the reflection vector of heliostat and planar process vector;

The reflection vector (rox, roy, roz) of any heliostat is such as formula shown in (2):

$\left\{\begin{array}{c}R=\sqrt{x^2+y^2+{(z-H)}^2}\\ rox=x/R\\ roy=y/R\\ roz=(z-H)/R\end{array}\right.---\left(2\right)$

In formula (2), R is the mould of heliostat center to heat dump center vector; H is the centre-height of heat dump, i.e. (0,0, the H) centre coordinate that is heat dump; The centre coordinate that (x, y, z) is heliostat;

The normal vector (nx, ny, nz) of any heliostat is such as formula shown in (3):

$\left\{\begin{array}{c}NR=\sqrt{{(a+rox)}^2+{(b+roy)}^2+{(c+roz)}^2}\\ nx=(a+rox)/NR\\ ny=(b+roy)/NR\\ nz=(c+roz)/NR\end{array}\right.---\left(3\right)$

In formula (3), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; NR is the mould of minute surface normal vector; H is the centre-height of heat dump, i.e. (0,0, the H) centre coordinate that is heat dump; (rox, roy, roz) is reflection vector;

Heliostat is generally twin shaft heliostat.In twin shaft heliostat as shown in Figure 4, a longitudinal bracing axle axle A, connects the fixed position on heliostat minute surface center and ground, keeps center, ground and minute surface center to fix when heliostat is spinned; Have a cross-brace axle axle B at the heliostat back side, axle B crosses minute surface center and vertical with minute surface normal vector, keeps minute surface center fix and minute surface is rotated around axle B during spin.Constructed can be obtained by the twin shaft of heliostat, two limits being parallel to the heliostat of axle B all the time with ground keeping parallelism.

Therefore, step 2.2 in the present embodiment) specifically refer to that through type (4) ~ (7) calculate the apex coordinate of heliostat;

$P_1\left\{\begin{array}{c}{x}_{p1}=x+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p1}=y+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p1}=z-\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(4\right)$

$P_2\left\{\begin{array}{c}{x}_{p1}=x-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p2}=y-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p2}=z+\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(5\right)$

$P_3\left\{\begin{array}{c}{x}_{p3}=x-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p3}=y-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p3}=z+\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(6\right)$

$P_4\left\{\begin{array}{c}{x}_{p4}=x+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p4}=y+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p4}=z-\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(7\right)$

In formula (4) ~ (7), (x, y, z) be the centre coordinate of heliostat, (nx, ny, nz) be the unit normal vector of this face heliostat, l is the length of side of heliostat and transverse axis parallel edges, and w is the length of side of heliostat and transverse axis (axle B) vertical edges, (x _pi, y _pi, z _pi) be i-th apex coordinate, i represents four summit sequence numbers respectively, i ∈ [Isosorbide-5-Nitrae].

In three dimensions, heliostat center and summit are equivalent to calculated line and plane intersection point along incident ray projection coordinate is on the ground calculated.Therefore, step 2.3 in the present embodiment) specifically refer to and calculate heliostat center and summit along incident ray projection coordinate on the ground according to formula (8) ~ (9);

$\left\{\begin{array}{c}{x}_{os}=-\left(\frac{a}{b}\right)*z+x\\ y_{os}=-\left(\frac{b}{c}\right)*z+y\end{array}\right.---\left(8\right)$

$\left\{\begin{array}{c}{x}_{psi}=-\left(\frac{a}{b}\right)*z_{pi}+x_{pi}\\ y_{psi}=-\left(\frac{b}{c}\right)*z_{pi}+x_{pi}\end{array}\right.---\left(9\right)$

In formula (8) ~ (9), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; The centre coordinate that (x, y, z) is heliostat, (x _os, y _os) be the center projection coordinate on the ground of heliostat, (x _psi, y _psi) be the summit projection coordinate on the ground of heliostat, (x _pi, y _pi, z _pi) be the coordinate on heliostat i-th summit, i represents four summit sequence numbers respectively, i ∈ [Isosorbide-5-Nitrae].

In the present embodiment, step 3) specifically refer to that utilizing Monte Carlo method to throw at random in the heliostat field throwing scope of spreading spreads luminous point, and throw and spread luminous point quantity and should try one's best greatly, large to the view field that can be formed completely on the ground with all minute surfaces covering Jing Chang, the expression formula of the coordinate (Xis, Yis) of projected light is on the ground such as formula shown in (10);

Xis＝x _min+(x _max-x _min)*rand(N,1)

(10)

Yis＝y _min+(y _max-y _min)*rand(N,1)

In formula (10), x _min, x _maxbe respectively the minimum and maximum X-coordinate value that heliostat apex coordinate projects on the ground along incident light, y _min, y _maxbe respectively the minimum and maximum Y-coordinate value that heliostat apex coordinate projects on the ground along incident light, N is that spot number is spread in throwing.The present embodiment, by determining that scope is spread in throwing, can reduce the workload that luminous point is spread in random throwing, reduces and is carrying out throwing the non-productive work spreading luminous point at random, thus can improve the counting yield when calculating mirror field optics efficiency.

As shown in Figure 5, in the present embodiment, step 4) detailed step as follows:

4.1) utilize CUDA computing platform to adopt the form of multithreading to carry out each parameter of initialization for each root light, initialization parameter comprises:

Flags initial value is 0: shade marker for determination, value be 0 explanation light by shade, be that 1 explanation light is not by shade;

Flagb initial value is 1: block marker for determination, and value is that 0 explanation light is blocked, and is that 1 explanation light is not blocked;

M initial value is 0: incident ray impinges upon the mirror position of target heliostat;

Temps initial value is 0: the intersection point Z coordinate (higher value) of incident ray and heliostat;

Xm, Ym, Zm initial value is 0: the intersecting point coordinate of light and heliostat;

Xrec, Zrec initial value is 0: the intersecting point coordinate of light and heat dump;

4.2) according to whether satisfying condition judge that impact point E projects in the parallelogram ABCD of ground formation at mirror, wherein for the vector formed between the summit A of parallelogram ABCD and impact point E, for the vector formed between summit A, summit D of parallelogram ABCD, for the vector formed between summit A, summit B of parallelogram ABCD; If impact point E is in the parallelogram that mirror is formed, redirect performs step 4.3), otherwise redirect performs step 4.6); Wherein, coefficient of determination u and v meets 0≤u, v≤1, and the expression formula of coefficient of determination u, v is such as formula shown in (11) ~ (13);

$\left\{\begin{array}{c}ax=x_B-x_A\\ ay=y_B-y_A\\ bx=x_D-x_A\\ by=y_D-y_A\\ cx=x_E-x_A\\ cy=y_E-y_A\end{array}\right.---\left(11\right)$

$\left\{\begin{array}{c}aa=ax*ax+ay*ay\\ ab=ax*bx+ay*by\\ ac=ax*cx+ay*cy\\ bb=bx*bx+by*by\\ bc=bx*cx+by*cy\end{array}\right.---\left(12\right)$

$\left\{\begin{array}{c}u=\frac{aa*bc-ab*ac}{aa*bb-ab*ab}\\ v=\frac{bb*ac-bc*ab}{aa*bb-ab*ab}\end{array}\right.---\left(13\right)$

In formula (11) ~ (13), x _afor the X-coordinate of ABCD point A, y _afor the Y-coordinate of an A, x _bfor the X-coordinate of ABCD point B, y _bfor the Y-coordinate of a B, x _dfor the X-coordinate of ABCD point D, y _dfor the Y-coordinate of a D, x _efor the X-coordinate of impact point E, y _efor the Y-coordinate of impact point E;

4.3) make the value of Flags be 1, and calculate the intersection point Z coordinate Zm of this incident ray and this heliostat;

4.4) judge whether Zm is greater than Temps, if then redirect performs step 4.5), otherwise redirect performs step 4.6);

4.5) make the value of Temps equal Zm, and record the position of this heliostat, be called target heliostat M; Target heliostat M is the mirror that incident ray is finally irradiated to; The final value of Temps is the intersection point Z coordinate of incident ray and target heliostat M, for making comparisons blocking decision stage;

4.6) judge whether circulation terminates; If then turn redirect to perform step 4.7), otherwise turn redirect execution step 4.2);

4.7) judge whether this root light impinges upon in mirror or, even the value of Flags is that 1 redirect performs step 4.8 on the ground), enter and block decision stage; Otherwise turn redirect and perform step 4.17), start after terminating to judge next luminous point;

4.8) the intersecting point coordinate Xm of the reflection ray corresponding to incident ray and other heliostat except target heliostat M is calculated, Ym, Zm; Blocking decision stage, by the intersecting point coordinate Zm of computational reflect light and surrounding heliostat, compare the size of Z coordinate, if once judged result is Zm be less than Temps, then illustrate that reflection ray is blocked, thus quantitatively record increase by 1 in shutting out the light of target heliostat; When calculating the intersecting point coordinate with other heliostat except target heliostat M herein, reduction judgement scope is carried out to all heliostats, and reduction principle for: line number is reduced to 2 row up and down of being expert at target heliostat, and columns is reduced to and arranges in the left and right 2 of target heliostat column;

4.9) cosine of an angle value formed by the vector at four vectors that rectangle four summits calculating intersection point and heliostat according to formula (14) are linked to be and rectangle four edges place, according to cosine value judge four vectors that rectangle four summits of intersection point and heliostat are linked to be and rectangle four edges place vector formed by angle whether all as acute angle, if be all acute angle, judge that this intersection point is in this face heliostat, then redirect performs step 4.10), otherwise redirect performs step 4.12);

cosEAD＝(x _E-x _A)*(x _D-x _A)+(y _E-y _A)*(y _D-y _A)+(z _E-z _A)*(z _D-z _A)

cosEDC＝(x _E-x _D)*(x _C-x _D)+(y _E-y _D)*(y _C-y _D)+(z _E-z _D)*(z _C-z _D)

(14)

cosECB＝(x _E-x _C)*(x _B-x _C)+(y _E-y _C)*(y _B-y _C)+(z _E-z _C)*(z _B-z _C)

cosEBA＝(x _E-x _B)*(x _A-x _B)+(y _E-y _B)*(y _A-y _B)+(z _E-z _B)*(z _A-z _B)

In formula (14), cosine of an angle value formed by the vector at four vectors that rectangle four summits that cosEAD, cosEDC, cosECB, cosEBA are respectively intersection point and heliostat are linked to be and rectangle four edges place; x _efor the X-coordinate of impact point E, x _afor the X-coordinate of rectangle summit A, x _bfor the X-coordinate of rectangle summit B, x _cfor the X-coordinate of rectangle summit C, x _dfor the X-coordinate of rectangle summit D; y _efor the Y-coordinate of punctuate E, y _afor the Y-coordinate of rectangle summit A, y _bfor the Y-coordinate of rectangle summit B, y _cfor the Y-coordinate of rectangle summit C, y _dfor the Y-coordinate of rectangle summit D; z _efor the Z coordinate of impact point E, z _afor the Z coordinate of rectangle summit A, z _bfor the Z coordinate of rectangle summit B, z _cfor the Z coordinate of rectangle summit C, z _dfor the Z coordinate of rectangle summit D;

4.10) judge whether this intersection point Zm is greater than Temps, if then redirect performs step 4.11), otherwise redirect performs step 4.12);

4.11) illustrate that this root light is blocked, make the value of Flagb be 0;

4.12) judge whether circulation terminates; If then redirect performs step 4.13), otherwise redirect performs step 4.8);

4.13) judge whether this root light is blocked, even the value of Flagb is that 1 redirect performs step 4.14), enter spilling decision stage; Otherwise redirect performs step 4.17);

4.14) intersection point of this root reflection ray and heat dump is calculated;

4.15) this intersection point is judged whether in heat dump, if then redirect performs step 4.16), otherwise redirect performs step 4.17); Meanwhile, if this intersection point is not in heat dump, then the spilling amount of light of this heliostat corresponding to root light increases by 1, otherwise this root light is effective sunlight;

4.16) then record X-coordinate and Z coordinate Xrec, the Zrec of this intersection point, obtained the schematic diagram that is heated of heat dump by the intersecting point coordinate recording all effective sunlights and heat dump, thus be conducive to scheduling and the optimization of analyzing prism field;

4.17) terminate, if all light completes as calculated, then blocked efficiency according in all incident raies by the shade that shade and the incident ray quantity be blocked calculate Jing Chang, the incident ray quantity that in all incident raies, light overflows calculates the spilling efficiency of Jing Chang.

The present embodiment is by step 4.1) ~ 4.17) Ray-tracing Method realize based on CUDA computing platform, step 4.1) ~ 4.17) by carrying out trace to any one random luminous point, comprise that shade judges, shadowing and spilling judgement.The situation of every mirror in any light and mirror field can be added up by the track following the tracks of light, comprise and whether intersect, shade, to block, finally whether overflow again, according to the shade amount of light of each heliostat, the quantity that shuts out the light, overflow the optical efficiency that amount of light and the ratio of the total amount of light shining into each heliostat can calculate Jing Chang, obtain the efficiency situation of each heliostat and the integrated optical efficiency of Jing Chang in mirror field.

The step 4.2 of the present embodiment) mainly determine whether throw arbitrarily the light be sprinkling upon corresponding to ground aiming pip impinges upon on the ground or other places (invalid light), or be effective sunlight.Effective sunlight in this way, judge the position relationship with each heliostat successively, if intersect, the amount of light that statistics shines into this face heliostat adds 1, and calculate the intersection point Z coordinate with this heliostat, by judging that the size of Z coordinate determines that light finally throws the position of the heliostat spread, referred to herein as target heliostat, all the other intersect but the less then explanation of Z coordinate by shade, the shade amount of light of its corresponding heliostat increases by 1 respectively.

And spread on the ground because random luminous point is all thrown, therefore in shade decision stage, the present embodiment step 4.2) take 2 dimension judging rules by whether unified in heliostat for judgement point, first by four of heliostat summits respectively along incident vector projection on ground, in the ordinary course of things, heliostat is rectangle, assuming that incident light is directional light, then heliostat on the ground be projected as a parallelogram, so judge whether incident ray impinges upon in heliostat just to convert to and judge a luminous point whether in a parallelogram.In the present embodiment, step 4.2) shadow stage adopt 2 dimension dicision ruleses, 2 dimension dicision ruleses judge simple efficient, and computing cost is little, and calculated performance is good, as shown in Figure 6, if some E is in parallelogram, then have wherein coefficient of determination meets 0≤u, v≤1, and the expression formula of coefficient of determination u, v is such as formula shown in (11) ~ (13).Step 4.9) stage of blocking adopt 3 dimension dicision ruleses, 3 dimension dicision ruleses computing cost little, calculated performance is good.3 dimension dicision ruleses judge a point in three dimensions whether in a rectangle, if four vectors that four summits of this point and rectangle are linked to be (this point is pointed to by summit in direction) are all acute angle with vector (direction is unified for clockwise or counterclockwise) the formed angle at rectangle four edges place.Above-mentioned four angle ∠ DAE, the expression formula in three dimensions of ∠ EDC, ∠ ECB, ∠ EBA is such as formula shown in (14), and as shown in Figure 6, if ∠ is DAE, ∠ EDC, ∠ ECB, the cosine value of ∠ EBA is all greater than 0, then this point is in rectangle.Therefore only need calculate this four cosine of an angle values, and two vectorial angle cosine expression formulas are denominator part is that modulus value is greater than 0 all the time, as long as therefore meet molecular moiety and be greater than 0.

In the present embodiment, step 5) in specifically refer to and calculate cosine efficiency according to formula (15);

eCos＝a*nx+b*ny+c*nz (15)

In formula (15), eCos is cosine efficiency, and a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; (nx, ny, nz) is the unit normal vector of this face heliostat, and * is multiplication operator.Because the relative position asked between heliostat of cosine efficiency has nothing to do, therefore the expression formula of the cosine efficiency of mirror can calculate by through type (15) arbitrarily.

The embodiment of the present invention is applied to the eSolar north and south Jing Chang of American South California, and eSolar north and south Jing Chang comprises 10000 mirrors altogether.The present embodiment has spread 10 to being total to throwing within the scope of whole mirror field ⁷spot number.The Geforce GTX 260 that the computer operating system used is windows7, video card calculates for support CUDA, by introducing CUDA computing platform under VS2008 platform, the time only needing 1728.08ms to be about 1.73s just can complete the calculating of whole mirror field efficiency, can obtain thus, the spended time of whole program computation mirror field efficiency is few, is conducive to optimization and the design of Jing Chang.In the present embodiment, can be calculated effective incident ray is 5966970, is 99207 by the light number of shade, and the light number be blocked is 89781, and the amount of light of overflowing heat dump is 454888, thus each efficiency that can calculate Jing Chang is as follows:

Cosine efficiency is 82.84%.

It is 96.04% that shade blocks efficiency ((shade amount of light+shut out the light quantity)/total effectively incident ray).

Overflowing efficiency (overflowing amount of light/(total effective sunlight quantity-shut out the light quantity)) is 92.26%.

Final mirror field optics efficiency (cosine efficiency * shade blocks efficiency * and overflows efficiency) is 73.40%.

The foregoing is only the preferred embodiment of the present invention, protection scope of the present invention is not limited in above-mentioned embodiment, and every technical scheme belonging to the principle of the invention all belongs to protection scope of the present invention.For a person skilled in the art, some improvements and modifications of carrying out under the prerequisite not departing from principle of the present invention, these improvements and modifications also should be considered as protection scope of the present invention.

Claims (5)

1. determine a method for tower type solar heat and power system mirror field optics efficiency, it is characterized in that implementation step is as follows:

1) generate Jing Chang, determine incident ray;

Described step 1) in generate the detailed step of Jing Chang as follows: determine mirror field parameters, heat dump size, sun altitude and solar azimuth, described mirror field parameters comprises heliostat quantity and size, mirror field type of arrangement; The centre coordinate O (x, y, z) calculating heliostat in mirror field generates Jing Chang;

Described step 1) in determine that the detailed step of incident ray is as follows:

Assuming that incident ray is directional light, minute surface is pointed to by the sun in direction, and incident ray in the projection components of X-axis, Y-axis, Z axis respectively such as formula shown in (1):

a＝-cos(h _s)*sin(p _s)

b＝cos(h _s)*cos(p _s) (1)

c＝-sin(h _s)

In formula (1), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis, h _sfor sun altitude, p _sfor solar azimuth; 2) determine that scope is spread in throwing according to the coordinate of described Jing Chang, incident ray and heliostat;

Described step 2) in calculate the detailed step of coordinate of heliostat as follows:

2.1) the reflection vector of heliostat and planar process vector is calculated;

2.2) apex coordinate of heliostat is calculated;

2.3) heliostat center and summit is calculated along incident ray projection coordinate on the ground;

2.4) dynamically follow the tracks of heliostat summit and centre coordinate along incident light view field on the ground, determine that scope is spread in throwing;

3) in the heliostat field that scope is spread in described throwing, random throwing spreads luminous point;

Described step 3) specifically refer to that utilizing Monte Carlo method to spread random throwing in the heliostat field of scope in described throwing spreads luminous point, and throw and spread luminous point quantity and should try one's best greatly, large to the view field that can be formed completely on the ground with all minute surfaces covering Jing Chang, the expression formula of the coordinate (Xis, Yis) of projected light is on the ground such as formula shown in (10);

Xis＝x _min+(x _max-x _min)*rand(N,1)

(10)

Yis＝y _min+(y _max-y _min)*rand(N,1)

In formula (10), x _min, x _maxbe respectively the minimum and maximum X-coordinate value that heliostat apex coordinate projects on the ground along incident light, y _min, y _maxbe respectively the minimum and maximum Y-coordinate value that heliostat apex coordinate projects on the ground along incident light, N is that spot number is spread in throwing;

4) CUDA computing platform is utilized to adopt the form of multithreading to carry out ray tracing for each root light, four summits being sprinkling upon ground luminous point and heliostat to determine incident ray and each heliostat successively crossing situation along incident ray projection coordinate is on the ground thrown by random, if the light of spot projection not in any heliostat corresponding to this luminous point is invalid incident ray, then consider next luminous point; Otherwise judge this incident ray not by the heliostat that shade is also finally irradiated to, and ask for the intersection point with this heliostat; Then enter and block decision stage; First calculate this reflection ray corresponding to root incident ray, judge reflection ray whether block by other mirror, if be not blocked, ask for the intersection point of the final and absorber of reflection ray, judge this intersection point whether in absorber, if intersection point is not in absorber, judge that light overflows; Finally blocked efficiency according in all incident raies by the shade that shade and the incident ray quantity be blocked calculate Jing Chang, the incident ray quantity that in all incident raies, light overflows calculates the spilling efficiency of Jing Chang;

Described step 4) detailed step as follows:

4.1) utilize CUDA computing platform to adopt the form of multithreading to carry out each parameter of initialization for each root light, described parameter comprises:

Flags initial value is 0: shade marker for determination, value be 0 explanation light by shade, be that 1 explanation light is not by shade;

Flagb initial value is 1: block marker for determination, and value is that 0 explanation light is blocked, and is that 1 explanation light is not blocked;

M initial value is 0: incident ray impinges upon the mirror position of target heliostat;

Temps initial value is 0: the intersection point Z coordinate of incident ray and heliostat;

Xm, Ym, Zm initial value is 0: the intersecting point coordinate of light and heliostat;

Xrec, Zrec initial value is 0: the intersecting point coordinate of light and heat dump;

4.2) according to whether satisfying condition judge that impact point E projects in the parallelogram ABCD of ground formation at mirror, wherein for the vector formed between the summit A of parallelogram ABCD and impact point E, for the vector formed between summit A, summit D of parallelogram ABCD, for the vector formed between summit A, summit B of parallelogram ABCD; If impact point E is in the parallelogram that mirror is formed, redirect performs step 4.3), otherwise redirect performs step 4.6); Wherein, coefficient of determination u and v meets 0≤u, v≤1, and the expression formula of coefficient of determination u, v is such as formula shown in (11) ~ (13);

$\left\{\begin{array}{c}ax=x_B-x_A\\ ay=y_B-y_A\\ bx=x_D-x_A\\ by=y_D-y_A\\ cx=x_E-x_A\\ cy=y_E-y_A\end{array}\right.---\left(11\right)$

$\left\{\begin{array}{c}aa=ax*ax+ay*ay\\ ab=ax*bx+ay*by\\ ac=ax*cx+ay*cy\\ bb=bx*bx+by*by\\ bc=bx*cx+by*cy\end{array}\right.---\left(12\right)$

$\left\{\begin{array}{c}u=\frac{aa*bc-ab*ac}{aa*bb-ab*ab}\\ v=\frac{bb*ac-bc*ab}{aa*bb-ab*ab}\end{array}\right.---\left(13\right)$

In formula (11) ~ (13), x _afor the X-coordinate of ABCD point A, y _afor the Y-coordinate of an A, x _bfor the X-coordinate of ABCD point B, y _bfor the Y-coordinate of a B, x _dfor the X-coordinate of ABCD point D, y _dfor the Y-coordinate of a D, x _efor the X-coordinate of impact point E, y _efor the Y-coordinate of impact point E;

4.3) make the value of Flags be 1, and calculate the intersection point Z coordinate Zm of this incident ray and this heliostat;

4.4) judge whether Zm is greater than Temps, if then redirect performs step 4.5), otherwise redirect performs step 4.6);

4.5) make the value of Temps equal Zm, and record the position of this heliostat, be called target heliostat M; Target heliostat M is the mirror that incident ray is finally irradiated to; The final value of Temps is the intersection point Z coordinate of incident ray and target heliostat M, for making comparisons blocking decision stage;

4.6) judge whether circulation terminates; If then turn redirect to perform step 4.7), otherwise turn redirect execution step 4.2);

4.7) judge whether this root light impinges upon in mirror or, even the value of Flags is that 1 redirect performs step 4.8 on the ground), enter and block decision stage; Otherwise turn redirect and perform step 4.17), start after terminating to judge next luminous point;

4.8) the intersecting point coordinate Xm of the reflection ray corresponding to incident ray and other heliostat except target heliostat M is calculated, Ym, Zm; Blocking decision stage, by the intersecting point coordinate Zm of computational reflect light and surrounding heliostat, compare the size of Z coordinate, if once judged result is Zm be less than Temps, then illustrate that reflection ray is blocked, thus quantitatively record increase by 1 in shutting out the light of target heliostat; When calculating the intersecting point coordinate with other heliostat except target heliostat M herein, reduction judgement scope is carried out to all heliostats, and reduction principle for: line number is reduced to 2 row up and down of being expert at target heliostat, and columns is reduced to and arranges in the left and right 2 of target heliostat column;

4.9) cosine of an angle value formed by the vector at four vectors that rectangle four summits calculating described intersection point and heliostat according to formula (14) are linked to be and rectangle four edges place, formed by the vector judging four vectors that rectangle four summits of intersection point and heliostat are linked to be and rectangle four edges place according to described cosine value, whether angle is all as acute angle, if be all acute angle, judge that this intersection point is in this face heliostat, then redirect performs step 4.10), otherwise redirect performs step 4.12);

cosEAD＝(x _E-x _A)*(x _D-x _A)+(y _E-y _A)*(y _D-y _A)+(z _E-z _A)*(z _D-z _A)

cosEDC＝(x _E-x _D)*(x _C-x _D)+(y _E-y _D)*(y _C-y _D)+(z _E-z _D)*(z _C-z _D)

(14)

cosECB＝(x _E-x _C)*(x _B-x _C)+(y _E-y _C)*(y _B-y _C)+(z _E-z _C)*(z _B-z _C)

cosEBA＝(x _E-x _B)*(x _A-x _B)+(y _E-y _B)*(y _A-y _B)+(z _E-z _B)*(z _A-z _B)

In formula (14), cosine of an angle value formed by the vector at four vectors that rectangle four summits that cosEAD, cosEDC, cosECB, cosEBA are respectively described intersection point and heliostat are linked to be and rectangle four edges place; x _efor the X-coordinate of impact point E, x _afor the X-coordinate of rectangle summit A, x _bfor the X-coordinate of rectangle summit B, x _cfor the X-coordinate of rectangle summit C, x _dfor the X-coordinate of rectangle summit D; y _efor the Y-coordinate of punctuate E, y _afor the Y-coordinate of rectangle summit A, y _bfor the Y-coordinate of rectangle summit B, y _cfor the Y-coordinate of rectangle summit C, y _dfor the Y-coordinate of rectangle summit D; z _efor the Z coordinate of impact point E, z _afor the Z coordinate of rectangle summit A, z _bfor the Z coordinate of rectangle summit B, z _cfor the Z coordinate of rectangle summit C, z _dfor the Z coordinate of rectangle summit D;

4.10) judge whether this intersection point Zm is greater than Temps, if then redirect performs step 4.11), otherwise redirect performs step 4.12);

4.11) illustrate that this root light is blocked, make the value of Flagb be 0;

4.12) judge whether circulation terminates; If then redirect performs step 4.13), otherwise redirect performs step 4.8);

4.13) judge whether this root light is blocked, even the value of Flagb is that 1 redirect performs step 4.14), enter spilling decision stage; Otherwise redirect performs step 4.17);

4.14) intersection point of this root reflection ray and heat dump is calculated;

4.15) this intersection point is judged whether in heat dump, if then redirect performs step 4.16), otherwise redirect performs step 4.17); Meanwhile, if this intersection point is not in heat dump, then the spilling amount of light of this heliostat corresponding to root light increases by 1, otherwise this root light is effective sunlight;

4.16) then record X-coordinate and Z coordinate Xrec, the Zrec of this intersection point, obtained the schematic diagram that is heated of heat dump by the intersecting point coordinate recording all effective sunlights and heat dump;

4.17) terminate, if all light completes as calculated, then blocked efficiency according in all incident raies by the shade that shade and the incident ray quantity be blocked calculate Jing Chang, the incident ray quantity that in all incident raies, light overflows calculates the spilling efficiency of Jing Chang;

5) cosine efficiency is calculated;

6) described shade blocked efficiency, overflow efficiency, cosine efficiency three and carry out multiplying, obtain final mirror field optics efficiency and export.

2. the method determining tower type solar heat and power system mirror field optics efficiency according to claim 1, is characterized in that: described step 2.1) specifically refer to that through type (2) ~ (3) calculate the reflection vector of heliostat and planar process vector;

The reflection vector (rox, roy, roz) of any heliostat is such as formula shown in (2):

$\left\{\begin{array}{c}R=\sqrt{x^2+y^2+{(z-H)}^2}\\ rox=x/R\\ roy=y/R\\ roz=(z-H)/R\end{array}\right.---\left(2\right)$

In formula (2), R is the mould of heliostat center to heat dump center vector; H is the centre-height of heat dump, i.e. (0,0, the H) centre coordinate that is heat dump; The centre coordinate that (x, y, z) is heliostat;

The normal vector (nx, ny, nz) of any heliostat is such as formula shown in (3):

$\left\{\begin{array}{c}NR=\sqrt{{(a+rox)}^2+{(b+roy)}^2+{(c+roz)}^2}\\ nx=(a+rox)/NR\\ ny=(b+roy)/NR\\ nz=(c+roz)/NR\end{array}\right.---\left(3\right)$

In formula (3), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; NR is the mould of minute surface normal vector; H is the centre-height of heat dump, i.e. (0,0, the H) centre coordinate that is heat dump; (rox, roy, roz) is reflection vector.

3. the method determining tower type solar heat and power system mirror field optics efficiency according to claim 1, is characterized in that: described step 2.2) specifically refer to that through type (4) ~ (7) calculate the apex coordinate of heliostat;

$P_1\left\{\begin{array}{c}{x}_{p1}=x+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p1}=y+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p1}=z-\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(4\right)$

$P_2\left\{\begin{array}{c}{x}_{p2}=x-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p2}=y-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p2}=z+\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(5\right)$

$P_3\left\{\begin{array}{c}{x}_{p3}=x-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p3}=y-\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p3}=z+\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(6\right)$

$P_4\left\{\begin{array}{c}{x}_{p4}=x+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}-\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ y_{p4}=y+\frac{1}{2}*w*nz*\frac{nx}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}+\frac{1}{2}*l*\frac{ny}{\sqrt{{\mathrm{nx}}^2+{\mathrm{ny}}^2}}\\ z_{p4}=z-\frac{1}{2}*w*\sqrt{1-{\mathrm{nz}}^2}\end{array}\right.---\left(7\right)$

In formula (4) ~ (7), the centre coordinate that (x, y, z) is heliostat, (nx, ny, nz) is the unit normal vector of this face heliostat, l is the length of side of heliostat and transverse axis parallel edges, and w is the length of side of heliostat and transverse axis vertical edges, (x _pi, y _pi, z _pi) be i-th apex coordinate, i represents four summit sequence numbers respectively, i ∈ [Isosorbide-5-Nitrae].

4. the method determining tower type solar heat and power system mirror field optics efficiency according to claim 1, is characterized in that: described step 2.3) specifically refer to and calculate heliostat center and summit along incident ray projection coordinate on the ground according to formula (8) ~ (9);

$\left\{\begin{array}{c}{x}_{os}=-\left(\frac{a}{b}\right)*z+x\\ y_{os}=-\left(\frac{b}{c}\right)*z+y\end{array}\right.---\left(8\right)$

$\left\{\begin{array}{c}{x}_{psi}=-\left(\frac{a}{b}\right)*z_{pi}+x_{pi}\\ y_{psi}=-\left(\frac{b}{c}\right)*z_{pi}+y_{pi}\end{array}\right.---\left(9\right)$

In formula (8) ~ (9), a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; The centre coordinate that (x, y, z) is heliostat, (x _os, y _os) be the center projection coordinate on the ground of heliostat, (x _psi, y _psi) be the summit projection coordinate on the ground of heliostat, (x _pi, y _pi, z _pi) be the coordinate on heliostat i-th summit, i represents four summit sequence numbers respectively, i ∈ [Isosorbide-5-Nitrae].

5. the method determining tower type solar heat and power system mirror field optics efficiency according to claim 1, is characterized in that: described step 5) in specifically refer to and calculate cosine efficiency according to formula (15);

eCos＝a*nx+b*ny+c*nz (15)

In formula (15), eCos is cosine efficiency, and a is the projection components of incident ray in X-axis, and b is the projection components of incident ray in Y-axis, and c is the projection components of incident ray at Z axis; (nx, ny, nz) is the unit normal vector of this face heliostat, and * is multiplication operator.