下载文档原格式
优化函数1、像差SPHA(球差)：surf表面编号/wave波长/target设定目标值/weight权重指定表面产生的球差贡献值，以波长表示。
如果表面编号值为零，则为整个系统的总和COMA(彗差) ：surf表面编号/wave波长/target设定目标值/weight权重指定表面产生的贡献值，以波长表示。
如果表面编号值为0，则是针对整个系统。
这是由塞得和数计算得到的第三级彗差，对非近轴系统无效.ASTI(像散)：指定表面产生像散的贡献值，以波长表示。
如果表面编号值为0，则是针对整个系统。
这是由塞得和数计算得到的第三级色散，对非近轴系统无效FCUR(场曲)：指定表面产生的场曲贡献值，以波长表示。
如果表面编号值为0，则是计算整个系统的场曲。
这是由塞得系数计算出的第三级场曲，对非近轴系统无效.DIST(畸变)：指定表面产生的畸变贡献值，以波长表示。
如果表面编号值为0，则使用整个系统。
同样，如果表面编号值为0，则畸变以百分数形式给出。
这是由塞得系数计算出的第三级畸变，对与非近轴系统无效.DIMX(最大畸变值)：它与DIST 相似，只不过它仅规定了畸变的绝对值的上限。
视场的整数编号可以是0，这说明使用最大的视场坐标，也可以是任何有效的视场编号。
注意，最大的畸变不一定总是在最大视场处产生。
得到的值总是以百分数为单位，以系统作为一个整体。
这个操作数对于非旋转对称系统可能无效。
AXCL(轴向色差)：以镜头长度单位为单位的轴向色差。
这是两种定义的最边缘的波长的理想焦面的间隔。
这个距离是沿着Z 轴测量的。
对非近轴系统无效.LACL(垂轴色差)：这是定义的两种极端波长的主光线截点的y方向的距离。
对于非近轴系统无效TRAR(垂轴像差)：在像面半径方向测定的相对于主光线的垂轴像差.TRAX(x方向垂轴像差)：在像面x方向测定的相对于主光线的垂轴像差TRAY(Y方向垂轴像差)：在像面Y方向测定的相对于主光线的垂轴像差TRAI(垂轴像差)：在指定表面半口径方向测定的相对于主光线的垂轴像差.类似于TRAR，只不过是针对一个表面，而不是指定的像面.OPDC(光程差)：指定波长的主光线的光程差.PETZ(匹兹伐曲率半径)：以镜头长度单位表示,对非近轴系统无效PETC(匹兹伐曲率)：以镜头长度单位的倒数表示,对非近轴系统无效RSCH：相对于主光线的RMS 斑点尺寸（光线像差）。
zemax主要优化函数（ZEMAX main optimization function）ZEMAX main optimization function table, 2008 28 Monday 07 00:53 optimization functionAberration 1SPHA (spherical aberration):surf surface number /wave, wavelength /target, set target value, /weight weightThe contribution of the spherical aberration to the specified surface is represented by wavelengths. If the surface number is zero, it is the sum of the entire systemCOMA (coma):surf surface number /wave wavelength /target target value /weight weightSpecify the contribution value of the surface, expressed in wavelength. If the surface number is 0, it is for the entire system. This isAnd as calculated by the number of third level coma of non paraxial invalid system.ASTI (astigmatism): Specifies the contribution of a surface to astigmatism, represented by wavelengths. If the surface number is 0, it is for the entire system. This is the third level was calculated by dispersion and number of non paraxial, invalid systemFCUR (field curvature): specify the contribution of the surface to the field, expressed at wavelengths. If the surface numberis 0, then the whole system is calculated. This is the third level was calculated by the field curvature coefficient, the non paraxial invalid system.DIST (distortion): Specifies the contribution of a surface to distortion, expressed as a wavelength. If the surface number is 0, the entire system is used. Similarly, if the surface number is 0, the distortion is given in percentage. This is the third level was calculated by the distortion coefficient of non paraxial, and invalid system.DIMX (maximum distortion): it is similar to DIST, but it only specifies the upper limit of the absolute value of distortion. The integer number of the field of view can be 0, indicating the use of the maximum field of view coordinates, or any valid field number. Note that the greatest distortion does not always occur at the maximum field of view. The resulting value is always in percentage, and the system as a whole. This operand may be invalid for a non rotationally symmetric system.AXCL (axial chromatic aberration): axial chromatic aberration in units of lens length. This is the two definition of the most marginal wavelength of the ideal focal plane interval. This distance is measured along the Z axis. Is not valid for non paraxial systemsLACL (vertical axis chromatic aberration): This is the distance of the Y direction of the main light point of the two extreme wavelengths defined. Invalid for non paraxial systemsTRAR (vertical axis aberrations): vertical aberrationsrelative to the main light measured in the direction of the image plane radiusTRAX (x direction vertical axis aberrations): vertical aberrations relative to the main light measured in the image plane xTRAY (Y direction vertical axis aberrations): vertical aberrations relative to the main light measured in the image plane YTRAI (vertical axis aberration): vertical aberrations in the specified surface half aperture direction relative to the main light. Similar to TRAR, only for a surface, not for a specified image planeOPDC (Guang Chengcha): the optical path difference of the primary ray of the specified wavelengthPETZ (Petzval curvature radius): the length of the lens unit, non paraxial system of invalidPETC (Petzval curvature): the reciprocal of length of the lens unit, the non paraxial invalid systemRSCH: the RMS spot size (light aberration) relative to the main light.RSCE: Hx, Hy, measured by the length of a lens, relative to the geometric centroid of the RMS spot (ray aberration).This operand is similar to RSCH, except that the reference point is like the centroid, not the primary ray. For details, see RSCH. R0Y}N ~Q!The RWCH: band wavelength Hx, Hy, is relative to the RMS wavefront difference of the main light. Its unit is wavelength. Since the average OPD has been subtracted, this RMS actually refers to the standard wavefront deviation. See RWCE. For details, see RSCHBRWCE: the band gap Hx, Hy, and the RMS wavefront difference relative to the diffraction centroid. This operation is useful for minimizing the wavefront deviation, deviation from the wavefront Strehl ratio and the MTF area under the curve is proportional to the. Its unit is wavelength. See RWCH. For details, see RSCHANAR: the angle difference radius of the main light relative to the main wavelength measured on an image plane. This number is defined as 1-cos theta theta, here is between the traced rays and the angle of the light. See TRARZERN: Zernike edge factor. The coefficient terms Int1, Int2, Hx, and Hy data values are used to specify the number of Zernike coefficient terms (1-37), the wavelength number, the sampling density (1=32*32, 2=64*64, etc.) and the field of view position. Note that if you have multiple ZERN operands with only variable number of entries, they should be placed in adjacent rows at the edit interface. Otherwise, the computation speed will be reducedTRAC: vertical axis aberrations relative to the center of mass in the direction of the image plane radius. Unlike other operands, TRAC works correctly according to the distribution of other TRAC operands in the edit function of the evaluation function. TRAC operands must be grouped together by field of view and wavelength. ZEMAX will trace all the TRAC rays of a common field of view together, and then calculate the centroid of all the light based on these collective data. Only the default evaluation function tool is used to import this operand into the evaluation function editing interface, rather than advising the user to use it directly.OPDX: this sphere can minimize the RMS wavefront difference relative to the optical path difference of a moving and tilted sphere; here, ZEMAX uses the centroid reference. OPDX has the same constraints as TRAC. See TRAC for more details.RSRE: grid wavelength Hx, Hy, measured by the length of a lens unit, relative to the geometric centroid of the RMS dot size (ray aberration). This operand is similar to RSCE, except that it uses the light of the rectangular mesh rather than the Gauss integral method. This operand is generally accepted for vignetting. A grid value of 1 represents 4 rays, 2 represents a trace, each quadrant tracks a 2*2 mesh (16 rays), and 3 represents a 3*3 grid (36 rays) per quadrant, and so on. The symmetry of the system has been taken into accountRSRH: similar to RSRE, but the reference point is the main light.RWRH: similar to RSRH, it's only calculating wavefrontdifferences, not speckle sizesRWRE: similar to RSRE, it's only calculating wavefront differences, not speckle sizes.The x component of TRAD:TRAR. TRAD has the same constraints as TRAC. See TRAC for more details.The Y component of TRAE:TRAR. TRAD has the same constraints as TRAC. See TRAC for more detailsX: vertical aberrations relative to the center of mass measured in the direction of the image plane TRCX.See TRAC. Only the default evaluation function tool is used to import this operand into the evaluation function editing interface, rather than advising the user to use it directly.Y: vertical aberrations relative to the center of mass measured in the direction of the image plane TRCYDISG: generalized distortion, the reference field wavelength is. Expressed as a percentage. This operator calculates the distortion of any light at any wavelength and in any field of view, taking any field of view as a reference. The method of use and the assumptions you make are the same as those described in the chapter on the analysis menuFCGS: normalized sagittal field curvature. The field values are calculated for each wavelength, each field of view. Normalization of this value yields a reasonable result, evenfor non rotationally symmetric systems. See the analysis of the field features in the menu Chapter 32 "1 & S"FCGT: normalized meridional field curve.DISC: normalized distortion. This operand calculates the normalized distortion of the entire visible field, and obtains the absolute value of the maximum non-linear value for the f- theta condition. This operand is very useful for the design of those f- theta lenses. -Y 0uB;OPDM: the optical path difference relative to the average OPD; this operand is calculated from the average OPD of all rays on the pupil as a reference to the OPD value. OPDM has the same constraints as TRAC. See TRACn BZ=Ytl A for more detailsBSER: Aiming error. Semi coordinates are divided by the effective focal length is defined as the main light aiming error traced on the field axis. This definition will produce measurements of the angular deviation of the image. A`mP-MKTp'Id9C'+^]2, modulation letter xDTy 7$KZD@FpGL 8, XqMTFT: the square wave modulation transfer function value of meridian. Sampling density wavelength. It calculates the diffraction MTF value. The parameter Int1 must be an integer (1, 2, 3)... ) 1 produces a sampling density of 32*32, 2 producesa sampling density of 64*64, and so forth. The Int2 must be a valid wavelength number, or 0, representing all wavelengths. The value of Hx must be a valid field number (1, 2)... .). Hy is the spatial frequency expressed in cycles per millimeter. If the computational accuracy of the sampling density relative to MTF is too low, all operands MTF will get zero values. If both meridian and sagittal MTF are required, they are placed in adjacent rows, MTFT and MTFS, which are computed simultaneously. See the instructions for the use of operand MTF in this chapter in more detail. P.BD4 T $MTFS: the modulation transfer function value of the arc vector. For details, see "MTFT"". F, G, l, jMTFA: the mean value of the modulation transfer function of the arc vector and the meridian. For details, see "MTFT"". |'p dg!MSWT: the square wave modulation transfer function value of meridian. For details, see "MTFT"". &Lt{p l8u6MSWS: the square wave modulation transfer function value of the arc vector. V4W0 shall 6MSWA: the mean square of the square wave modulation transfer function of the sagittal and meridional. For details, see "MTFT"". \`cp=OY[ZGMTA: the geometric transfer function of arc vector and meridian, the mean of response curve. The parameter Int1 must be an integer (1, 2,...) 1 produce 32*32 sampling density, 2 produce 64*64 sampling density, and so on.Int2 can be any valid wavelength number, or 0, representing all wavelengths. The value of Hx must be a valid field number (1, 2,...). Hy is the spatial frequency expressed in cycles per millimeter. Px is a marker, and if it is 0, the diffraction limit is used to scale the transfer function values (recommended), otherwise they are not scaled. See the section on the use of operand MTF in this chapter for more details. O$zn+5f9/GMTS: the geometric transfer function response curve of the arrow vector. See the operands GMTA., I, SDlS, G in detailGMTT: Meridian geometric transfer function response curve, in detail, see operand GMTA.WbP, Bp{| Gy =< Y3. Basic optical properties /X2, A, uU (#]. /EFFL: the effective focal length is expressed in units of lens length. It is aimed at the paraxial system, and for non paraxial systems may be inaccurate U:o`/4 "XL"PIMH: the image height on the paraxial image plane of the specified wavelength. A@+3- 0/PMAG: paraxial magnification. This is the ratio of the height of the paraxial main light to the height of the paraxial image plane. Only useful for finite far conjugate systems. Note thatthe near axis image plane can be used even though the system cannot be ideal focused. T w% CwCR5AMAG: angle magnification. This is the ratio of the angle of the main axis of the light between the space and the object space. For non paraxial systems, invalid 3, l, {DE \ENPP: relative to the first surface, the pupil position is indicated by the unit of lens length. This is the paraxial pupil position, valid only for the central system, 9 nG MEXPP: expressed as a unit of lens length relative to the pupil position of the first face. This is the paraxial pupil position, valid only for the central system, A>, Ktl, uLINV: the Lagrange invariants of a system, expressed in units of lens length. Calculate the value MJ 8 & s=y with the paraxial edge light and the main ray dataWFNO: work, F/#. This is calculated from the angle of the actual edge light in the space relative to the main light. SN e E; KPOWR: the weight of the specified number of surfaces (represented by reciprocal units of the length of the lens). This operand is only valid for standard surfaces. Surface numbering &HI^ \ =EPDI: the pupil diameter is expressed in units of lens length. P 85ISFN: like space, F/#. This operand is infinite and conjugatenear axis F/#. See "WFNO" D, bP\}#aEFLX: the effective focal length of the primary wavelength of the surface in the specified range on the fixed X plane is expressed in terms of the length of the lens. The number of the first surface and the number of the last surface. */x9_Lo ^EFLY: the effective focal length of the main wavelength of the surface in the specified range on the fixed Y plane. The jCg <L is expressed in terms of the length of the lens;SFNO: the sagittal work F/# calculated at any defined field of view and wavelength. See TFNO. The field of view is S -F|5 "Sp"TFNO: the meridional work F/# calculated at any defined field of view and wavelength. See SFNO. WN\ sfnJURIMAG: image resolution. No matter what the default settings are currently used, this operand has the same resolution as the result of the geometric image analysis. To use this operand, you must first define the set values in the geometry analysis feature,Then press the Save button in the settings box. The operand IMAE will have the same resolution (normalization) as the image analysis feature. See the instructions in the optimization of the operand IMAE below.。
优化函数1、像差SPHA(球差)：surf表面编号/wave波长/target设定目标值/weight权重指定表面产生的球差贡献值，以波长表示。
如果表面编号值为零，则为整个系统的总和COMA(彗差) ：surf表面编号/wave波长/target设定目标值/weight权重指定表面产生的贡献值，以波长表示。
如果表面编号值为0，则是针对整个系统。
这是由塞得和数计算得到的第三级彗差，对非近轴系统无效.ASTI(像散)：指定表面产生像散的贡献值，以波长表示。
如果表面编号值为0，则是针对整个系统。
这是由塞得和数计算得到的第三级色散，对非近轴系统无效FCUR(场曲)：指定表面产生的场曲贡献值，以波长表示。
如果表面编号值为0，则是计算整个系统的场曲。
这是由塞得系数计算出的第三级场曲，对非近轴系统无效.DIST(畸变)：指定表面产生的畸变贡献值，以波长表示。
如果表面编号值为0，则使用整个系统。
同样，如果表面编号值为0，则畸变以百分数形式给出。
这是由塞得系数计算出的第三级畸变，对与非近轴系统无效.DIMX(最大畸变值)：它与DIST 相似，只不过它仅规定了畸变的绝对值的上限。
视场的整数编号可以是0，这说明使用最大的视场坐标，也可以是任何有效的视场编号。
注意，最大的畸变不一定总是在最大视场处产生。
得到的值总是以百分数为单位，以系统作为一个整体。
这个操作数对于非旋转对称系统可能无效。
AXCL(轴向色差)：以镜头长度单位为单位的轴向色差。
这是两种定义的最边缘的波长的理想焦面的间隔。
这个距离是沿着Z 轴测量的。
对非近轴系统无效.LACL(垂轴色差)：这是定义的两种极端波长的主光线截点的y方向的距离。
对于非近轴系统无效TRAR(垂轴像差)：在像面半径方向测定的相对于主光线的垂轴像差.TRAX(x方向垂轴像差)：在像面x方向测定的相对于主光线的垂轴像差TRAY(Y方向垂轴像差)：在像面Y方向测定的相对于主光线的垂轴像差TRAI(垂轴像差)：在指定表面半口径方向测定的相对于主光线的垂轴像差.类似于TRAR，只不过是针对一个表面，而不是指定的像面.OPDC(光程差)：指定波长的主光线的光程差.PETZ(匹兹伐曲率半径)：以镜头长度单位表示,对非近轴系统无效PETC(匹兹伐曲率)：以镜头长度单位的倒数表示,对非近轴系统无效RSCH：相对于主光线的RMS 斑点尺寸（光线像差）。
zemax优化函数使用方法Zemax是一款常用于光学系统设计和优化的软件工具。
其中的优化函数是Zemax的一个重要功能，可以帮助用户通过自动搜索和调整系统参数，找到最优的设计方案。
本文将介绍Zemax优化函数的使用方法。
一、什么是优化函数在光学系统设计中，我们通常需要通过调整系统的各种参数来实现特定的设计要求。
而优化函数就是帮助我们在众多参数中找到最优解的工具。
其原理是通过数值计算和模拟，自动化地搜索参数空间，以寻找最佳的设计方案。
二、Zemax中的优化函数Zemax中的优化函数可以分为两大类：单变量优化和多变量优化。
单变量优化是指只有一个参数需要进行调整，而多变量优化则是同时调整多个参数。
下面将分别介绍这两种优化函数的使用方法。
1. 单变量优化函数单变量优化函数可以通过调整一个参数，来寻找最优解。
在Zemax 中，我们可以选择需要调整的参数，并设置其变化的范围和步长。
然后，通过运行优化函数，Zemax会自动搜索参数空间，并给出最优的结果。
2. 多变量优化函数多变量优化函数可以同时调整多个参数，以找到最优解。
在Zemax 中，我们可以选择多个参数，并设置它们的变化范围。
然后，通过运行优化函数，Zemax会自动搜索多个参数的组合，并给出最佳的设计方案。
三、使用优化函数的步骤使用Zemax的优化函数，一般需要按照以下步骤进行操作：1. 定义优化目标：首先，我们需要明确设计的目标和要求，例如最小化像差、最大化光学传输等。
这样才能设置正确的优化函数和参数。
2. 设置参数范围：根据设计要求，我们需要选择需要调整的参数，并设置它们的变化范围。
例如，镜片的曲率半径、透镜的厚度等。
3. 运行优化函数：在Zemax中，我们可以选择不同的优化函数进行计算。
例如，全局优化、局部优化等。
根据设计要求和参数设置，选择适合的优化函数，并运行它。
4. 分析结果：运行完优化函数后，Zemax会给出最优的设计方案。
我们可以通过分析结果，评估设计的优劣，并进行进一步的优化和改进。
zemax主要优化函数zemax主要优化函数表2008年07月28日星期一 00:53优化函数1、像差SPHA(球差):surf表面编号/wave波长/target设定目标值/weight权重指定表面产生的球差贡献值，以波长表示。
如果表面编号值为零，则为整个系统的总和COMA(彗差) :surf表面编号/wave波长/target设定目标值/weight权重指定表面产生的贡献值，以波长表示。
如果表面编号值为 0，则是针对整个系统。
这是由塞得和数计算得到的第三级彗差，对非近轴系统无效.ASTI(像散):指定表面产生像散的贡献值，以波长表示。
如果表面编号值为0，则是针对整个系统。
这是由塞得和数计算得到的第三级色散，对非近轴系统无效 FCUR(场曲):指定表面产生的场曲贡献值，以波长表示。
如果表面编号值为0，则是计算整个系统的场曲。
这是由塞得系数计算出的第三级场曲，对非近轴系统无效.DIST(畸变):指定表面产生的畸变贡献值，以波长表示。
如果表面编号值为0，则使用整个系统。
同样，如果表面编号值为 0，则畸变以百分数形式给出。
这是由塞得系数计算出的第三级畸变，对与非近轴系统无效.DIMX(最大畸变值):它与 DIST 相似，只不过它仅规定了畸变的绝对值的上限。
视场的整数编号可以是 0，这说明使用最大的视场坐标，也可以是任何有效的视场编号。
注意，最大的畸变不一定总是在最大视场处产生。
得到的值总是以百分数为单位，以系统作为一个整体。
这个操作数对于非旋转对称系统可能无效。
AXCL(轴向色差):以镜头长度单位为单位的轴向色差。
这是两种定义的最边缘的波长的理想焦面的间隔。
这个距离是沿着Z 轴测量的。
对非近轴系统无效. LACL(垂轴色差):这是定义的两种极端波长的主光线截点的y方向的距离。
对于非近轴系统无效TRAR(垂轴像差):在像面半径方向测定的相对于主光线的垂轴像差. TRAX(x方向垂轴像差):在像面x方向测定的相对于主光线的垂轴像差 TRAY(Y方向垂轴像差):在像面Y方向测定的相对于主光线的垂轴像差 TRAI(垂轴像差):在指定表面半口径方向测定的相对于主光线的垂轴像差.类似于 TRAR，只不过是针对一个表面，而不是指定的像面.OPDC(光程差):指定波长的主光线的光程差.PETZ(匹兹伐曲率半径):以镜头长度单位表示,对非近轴系统无效 PETC(匹兹伐曲率):以镜头长度单位的倒数表示,对非近轴系统无效 RSCH:相对于主光线的RMS 斑点尺寸(光线像差)。
Zemax-光学系统设计经验（1）---优化函数的使用使用Zemax设计光学系统，基本上就是根据设计要求，给出初始设计，然后优化系统。
初始设计需要对光学有系统的学习，需要长期的经验，不同的领域有不同的要求，初始设计会大不相同，zemax不会给你太多的帮助。
Zemax的精髓是能计算出光路图，然后使用operand(优化函数)优化各项光学参数。
1. default merit function，当属最有用的优化函数，配合EFFL （有效焦距）使用，基本可以设计大部分光学系统。
可以使用RMS spot radius and rms wavefront error。
另外设置好变量。
Zemax会自动生成优化系数(weight). 函数行的量取决于波长数，场(field)数，也决定了计算的快慢。
2. 自己设计优化函数。
第一步，需要知道你的优化目标：焦距，abberation，耦合效率。
设置好constraints.设计的constraints:Lens size, cost; edge and center thickeness; minimum number of lens; simple design; cheap举例来讲，耦合效率。
有两个最有用的函数：FICL, POPD. 前者快，后者慢，但后者对大多数系统要准确一些。
读一下manual，你会知道你要设置什么参数。
优化目标是1，weight 是1. 要知道在哪个surface上，还有在什么wavelength, field, 最重要的是什么优化数据，可以是耦合效率，也可以是beam size。
如果都要考虑，可以设置新的POPD函数，设置好优化系数(weight). 有例子，改一下merit function就可以了。
有些有用的优化函数：加减乘除：SUMM, DIFF, PROD,DIVI镜头数据：MXCG, MNCG, CTGT, MNCT,变量的设置也很有讲究，越多越靠近理想目标，但是降低速度和提高坏设计的几率.几点经验：1. 尽可能 use solve， instead of 变量2. 尽可能 use default merit function3. 理解constraints4. 理解和使用symmetry5. 去掉无用的变量.6. 知道怎么去改变设计。
