曹宇澤,馬文禮
兩步式靈敏度矩陣法在卡塞格林望遠(yuǎn)鏡裝調(diào)中的應(yīng)用
曹宇澤1,2,馬文禮1*
1中國(guó)科學(xué)院光電技術(shù)研究所,四川 成都 610209;2中國(guó)科學(xué)院大學(xué),北京 100049
為了調(diào)節(jié)視場(chǎng)較大的卡塞格林望遠(yuǎn)鏡的次鏡位置,提出了兩步式靈敏度矩陣模型的計(jì)算機(jī)輔助裝調(diào)方法。在分析了傳統(tǒng)的二次模型靈敏度矩陣法的缺陷的基礎(chǔ)上,根據(jù)靈敏度矩陣的特點(diǎn)加入了精調(diào)步驟,對(duì)傳統(tǒng)的靈敏度矩陣法進(jìn)行了改進(jìn)。針對(duì)卡塞格林系統(tǒng),分析了各項(xiàng)澤尼克系數(shù)與失調(diào)量之間的關(guān)系,并對(duì)300 mm口徑,0.6°視場(chǎng)的卡塞格林系統(tǒng)進(jìn)行了校正仿真。仿真結(jié)果顯示,傳統(tǒng)的靈敏度矩陣法校正后沿、、軸偏移及繞、軸傾斜的失調(diào)量的均值分別為:-0.0684 mm、-0.0892 mm、0.0015 mm、0.0498°和-0.0444°,全視場(chǎng)波像差RMS均小于0.1(=632.8 nm);兩步式靈敏度矩陣法校正后的均值分別為-0.0018 mm、-0.0012 mm、0.0002 mm、0.0008°和-0.0012°,全視場(chǎng)RMS均小于0.03,明顯優(yōu)于傳統(tǒng)的靈敏度矩陣法。
像差校正;靈敏度矩陣法;計(jì)算機(jī)輔助裝調(diào);卡塞格林系統(tǒng)
隨著望遠(yuǎn)鏡技術(shù)的發(fā)展,望遠(yuǎn)鏡的口徑和視場(chǎng)越來越大,光學(xué)系統(tǒng)的結(jié)構(gòu)越來越復(fù)雜,加工制造和裝調(diào)的難度也隨之提升[1]。通過干涉儀等檢測(cè)設(shè)備可以測(cè)量光學(xué)系統(tǒng)焦平面的光斑圖,并通過光斑圖計(jì)算得到各項(xiàng)澤尼克系數(shù)。為了得到良好的成像質(zhì)量,采用計(jì)算機(jī)輔助裝調(diào)技術(shù),對(duì)光學(xué)系統(tǒng)進(jìn)行實(shí)時(shí)檢測(cè)并與理論結(jié)果進(jìn)行比較,通過建立澤尼克系數(shù)與失調(diào)量之間的數(shù)學(xué)模型,精確地對(duì)元件的失調(diào)量進(jìn)行校正[2-3]。應(yīng)用最廣泛的計(jì)算機(jī)輔助裝調(diào)方法是靈敏度矩陣法。該方法是一種在分析像差特性的基礎(chǔ)上,通過建立失調(diào)量與像差系數(shù)的數(shù)學(xué)模型來校正像差的方法[4-5]。最早的靈敏度矩陣是一階模型[6-9],校正后可達(dá)到全視場(chǎng)RMS為0.1的結(jié)果。中國(guó)科學(xué)院長(zhǎng)春光機(jī)所的顧志遠(yuǎn)改進(jìn)了靈敏度矩陣法,將數(shù)學(xué)模型擴(kuò)展到二階,并用改進(jìn)的靈敏度矩陣法對(duì)500 mm口徑的離軸望遠(yuǎn)鏡進(jìn)行校正,得到了全視場(chǎng)RMS均小于0.1的結(jié)果[10]。
本文分析了傳統(tǒng)靈敏度矩陣法的缺陷,通過擬合的方法構(gòu)建靈敏度矩陣并加入了精調(diào)的步驟,進(jìn)一步提高校正精度。使用兩步式靈敏度矩陣法對(duì)300 mm口徑、0.6°視場(chǎng)的卡塞格林試驗(yàn)系統(tǒng)進(jìn)行校正仿真,得到了良好的校正結(jié)果。
在光學(xué)系統(tǒng)中,元件的失調(diào)量會(huì)引入像差,單一元件有五個(gè)自由度的失調(diào)量,分別為沿軸偏移、沿軸偏移、沿軸(光軸)偏移、繞軸傾斜和繞軸傾斜,用D、D、D、T、T來表示,光學(xué)系統(tǒng)的像差可以用澤尼克多項(xiàng)式形式表示。根據(jù)泰勒定理,失調(diào)狀態(tài)下光學(xué)系統(tǒng)澤尼克多項(xiàng)式系數(shù)可以表示為
傳統(tǒng)的靈敏度矩陣法使用的數(shù)學(xué)模型是失調(diào)量和像差之間二次相關(guān),僅進(jìn)行單次校正。通過構(gòu)建二次修正模型,得到各自由度的失調(diào)量和每個(gè)澤尼克系數(shù)之間的靈敏度矩陣,進(jìn)而對(duì)失調(diào)量進(jìn)行校正。二次修正模型表達(dá)式為
然而,由于不同自由度的失調(diào)量對(duì)澤尼克系數(shù)的影響并非完全相互獨(dú)立,失調(diào)量不同的情況下靈敏度矩陣也會(huì)隨之變化。因此,傳統(tǒng)的靈敏度矩陣法具有較大的誤差。用擬合方法構(gòu)建靈敏度矩陣并將像差校正分為粗調(diào)和精調(diào)兩個(gè)步驟可以有效減小靈敏度矩陣變化帶來的誤差。
由于靈敏度矩陣會(huì)隨著失調(diào)量的變化而改變,而且失調(diào)量越小靈敏度矩陣越接近線性,因此兩步式靈敏度矩陣法分為粗調(diào)和精調(diào)兩個(gè)步驟,粗調(diào)的目的是減小失調(diào)量,精調(diào)使用線性模型進(jìn)行精確校正。
這種擬合得到的靈敏度函數(shù)精度較高,通過靈敏度函數(shù)可以構(gòu)建對(duì)應(yīng)位置的修正模型:
通過修正模型可以計(jì)算出粗調(diào)的校正量。
粗調(diào)后失調(diào)量顯著減小,此時(shí)粗調(diào)中使用的靈敏度矩陣與粗調(diào)后對(duì)應(yīng)位置的靈敏度矩陣有較大的偏差,如果重復(fù)使用粗調(diào)的靈敏度矩陣會(huì)導(dǎo)致校正結(jié)果偏差很大。由于各自由度失調(diào)量的絕對(duì)值小于0.1 mm或0.1°時(shí)失調(diào)量與澤尼克系數(shù)近似線性相關(guān),因此精調(diào)采用失調(diào)量較小時(shí)的一階靈敏度矩陣。修正模型的表達(dá)式為
在構(gòu)建靈敏度矩陣時(shí),需要考慮所采用的澤尼克多項(xiàng)式系數(shù)。由于光學(xué)系統(tǒng)的失調(diào)量主要造成初級(jí)像差,對(duì)成像質(zhì)量影響最大的是傾斜、離焦、彗差和像散,因此靈敏度矩陣的構(gòu)建優(yōu)先考慮0、1、2、…、8這幾個(gè)系數(shù)。
在選擇靈敏度矩陣的系數(shù)時(shí),需要注意以下三點(diǎn):
1) 選用的澤尼克系數(shù)必須能夠反映所有自由度的失調(diào)量變化;
2) 澤尼克系數(shù)與失調(diào)量之間呈明顯的線性或二次相關(guān);
3) 澤尼克系數(shù)變化量不宜太小,能夠確保靈敏度函數(shù)的精度。
因此,兩步式靈敏度矩陣法與傳統(tǒng)的靈敏度矩陣法相比,主要從澤尼克系數(shù)的選取、靈敏度函數(shù)的計(jì)算方法和校正次數(shù)這三方面進(jìn)行優(yōu)化,提高校正精度。
表1 卡塞格林望遠(yuǎn)鏡光學(xué)系統(tǒng)參數(shù)
仿真使用300 mm口徑、0.6°視場(chǎng)的卡塞格林系統(tǒng),波段范圍450 nm~800 nm,設(shè)計(jì)位置的全視場(chǎng)波像差RMS均小于0.03,并且要求50線對(duì)/mm處的MTF值大于0.6,具體參數(shù)如表1所示。
光學(xué)系統(tǒng)的結(jié)構(gòu)圖如圖1所示。
校正過程中,以主鏡為基準(zhǔn),逐一添加透鏡并與設(shè)計(jì)位置對(duì)比,通過兩步式靈敏度矩陣法可以精確地校正每一塊透鏡的位置。本文主要通過校正次鏡位置來說明兩步式靈敏度矩陣法的可行性,因此在后續(xù)仿真中將次鏡以外的透鏡置于設(shè)計(jì)位置,僅研究次鏡的失調(diào)量校正。
以主鏡為基準(zhǔn),保持其他光學(xué)元件位置不變,對(duì)次鏡的5個(gè)自由度進(jìn)行校正。首先需要選擇靈敏度矩陣使用的澤尼克系數(shù)。在(0°,0°)、(0°,0.3°)、(0°,-0.3°)、(0.3°,0°)、(-0.3°,0°)這幾個(gè)視場(chǎng)進(jìn)行失調(diào)狀態(tài)仿真并分析各自由度的失調(diào)量變化對(duì)澤尼克系數(shù)的影響。其中軸上視場(chǎng)各自由度失調(diào)量與澤尼克系數(shù)的關(guān)系曲線圖如圖2所示。
圖中左列是受失調(diào)量影響較小的澤尼克系數(shù)曲線圖,右列是受失調(diào)量影響較大的澤尼克系數(shù)曲線圖,圖中未出現(xiàn)的澤尼克系數(shù)代表不受對(duì)應(yīng)自由度的影響。由曲線圖可知,各自由度失調(diào)量對(duì)澤尼克系數(shù)的影響有三種:無影響、線性相關(guān)、二次相關(guān),并且失調(diào)量對(duì)澤尼克系數(shù)的影響程度不同,其中D和T主要影響1和6,次要影響0、3、4和8;D和T主要影響2和7,次要影響0、3、4和8;D主要影響0和3,次要影響8。
圖1 卡塞格林系統(tǒng)的光學(xué)結(jié)構(gòu)圖
因此,根據(jù)澤尼克系數(shù)的選用原則,靈敏度矩陣的構(gòu)建采用0、1、2、3、6和7這六項(xiàng)。
初始失調(diào)量隨機(jī)選取,粗調(diào)后的失調(diào)量如表2。
為了說明不同失調(diào)狀態(tài)的靈敏度矩陣的差異,第三組初始狀態(tài)的軸上視場(chǎng)的靈敏度矩陣如表3所示(分號(hào)前后分別為二次系數(shù)和一次系數(shù)):
粗調(diào)后的五個(gè)自由度失調(diào)量的均值分別為-0.0626 mm、-0.0594 mm、0.0005 mm、0.0513°和-0.0424°。由圖2可以看出,只有受失調(diào)量影響較小的澤尼克系數(shù)才會(huì)與對(duì)應(yīng)失調(diào)量二次相關(guān)。對(duì)于單一自由度,當(dāng)失調(diào)量小于0.1 mm或0.1°時(shí),二次函數(shù)和一次函數(shù)的偏差對(duì)失調(diào)量的影響小于0.0001 mm,因此失調(diào)量小于0.1 mm或0.1°時(shí)失調(diào)量與澤尼克系數(shù)近似線性相關(guān),精調(diào)的靈敏度矩陣采用失調(diào)量為-0.0626 mm、-0.0594 mm、0.0005 mm、0.0513°和-0.0424°時(shí)的一階靈敏度矩陣。精調(diào)后的失調(diào)量如表4所示。
粗調(diào)后第三組的軸上視場(chǎng)的靈敏度矩陣如表5所示。
由于粗調(diào)后光學(xué)系統(tǒng)已經(jīng)較為接近設(shè)計(jì)位置,此時(shí)靈敏度函數(shù)用Matlab擬合為一階函數(shù)時(shí)的誤差已經(jīng)小于二階函數(shù)的誤差。
表2 粗調(diào)后的各自由度失調(diào)量
表3 初始狀態(tài)的靈敏度矩陣
表4 精調(diào)后的各自由度失調(diào)量
第三組精調(diào)采用的一階靈敏度矩陣如表6所示。
通過幾個(gè)靈敏度矩陣可以看出,粗調(diào)后的靈敏度矩陣與初始位置相比確實(shí)有了很大的變化。傳統(tǒng)的靈敏度矩陣法為單次校正,等同于默認(rèn)靈敏度矩陣處處相等。因此,兩步式靈敏度矩陣法確實(shí)比傳統(tǒng)的靈敏度矩陣法精度更高。同時(shí),采用精調(diào)矩陣也比重復(fù)使用粗調(diào)的二階矩陣精度更高。
傳統(tǒng)的靈敏度矩陣法校正結(jié)果如表7所示。
同時(shí)附上四組初始失調(diào)狀態(tài)校正前后的澤尼克系數(shù)表,如表8所示。
本文在分析靈敏度矩陣的特點(diǎn)以及失調(diào)量對(duì)澤尼克系數(shù)影響的基礎(chǔ)上,從更換選用的澤尼克系數(shù)、用擬合的方法構(gòu)建靈敏度矩陣和增添精調(diào)步驟這三個(gè)方面對(duì)傳統(tǒng)的靈敏度矩陣法進(jìn)行了改進(jìn)。對(duì)300 mm口徑、0.6°視場(chǎng)的卡塞格林望遠(yuǎn)鏡進(jìn)行了多次校正仿真。仿真結(jié)果顯示,傳統(tǒng)的靈敏度矩陣法校正后沿、、軸偏移及繞、軸傾斜的失調(diào)量的均值分別為:-0.0684 mm、-0.0892 mm、0.0015 mm、0.0498°和-0.0444°,全視場(chǎng)波像差RMS均小于0.1;兩步式靈敏度矩陣法校正后的均值分別為-0.0018 mm、-0.0012 mm、0.0002 mm、0.0008°和-0.0012°,全視場(chǎng)RMS均小于0.03,校正后的光學(xué)系統(tǒng)接近設(shè)計(jì)位置,明顯優(yōu)于傳統(tǒng)的靈敏度矩陣法,說明了兩步式靈敏度矩陣法對(duì)卡塞格林望遠(yuǎn)鏡的裝調(diào)具有很好的實(shí)用價(jià)值。
表5 粗調(diào)后的靈敏度矩陣
表6 精調(diào)采用的一階靈敏度矩陣
表7 傳統(tǒng)的靈敏度矩陣法校正后的失調(diào)量
表8 校正前后的澤尼克系數(shù)
[1] Zhou L F. Study on the alignment technique of large aperture reflecting telescope on-line[D]. Chengdu: University of Chinese Academy of Sciences (Institute of Optics and Electronics), 2016.
周龍峰. 大口徑反射式望遠(yuǎn)鏡在線調(diào)整技術(shù)研究[D]. 成都: 中國(guó)科學(xué)院研究生院(光電技術(shù)研究所), 2016.
[2] Figoski J W, Shrode T E, Moore G F. Computer-aided alignment of a wide-field, three-mirror, unobscured, high-resolution sensor[J]., 1989, 1049: 166.
[3] Egdall I M. Manufacture of a three-mirror wide-field optical system[J]., 1985, 24(2): 242285.
[4] Gong D, Tian T Y, Wang H. Computer-aided alignment of off-axis three-mirror system by using Zernike coefficients[J]., 2010, 18(8): 1754–1759.
鞏盾, 田鐵印, 王紅. 利用Zernike系數(shù)對(duì)離軸三反射系統(tǒng)進(jìn)行計(jì)算機(jī)輔助裝調(diào)[J]. 光學(xué)精密工程, 2010, 18(8): 1754–1759.
[5] Yang X F. Study on the computer-aided alignment of three-mirror optical system[D]. Changchun: Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, 2005.
楊曉飛. 三反射鏡光學(xué)系統(tǒng)的計(jì)算機(jī)輔助裝調(diào)技術(shù)研究[D]. 長(zhǎng)春: 中國(guó)科學(xué)院長(zhǎng)春光學(xué)精密機(jī)械與物理研究所, 2005.
[6] Kim S, Yang H S, Lee Y W,. Merit function regression method for efficient alignment control of two-mirror optical systems. [J]., 2007, 15(8): 5059–5068.
[7] Wang B, Jiang S L. Study on computer-aided alignment method of Cassegrain system[J]., 2008, 30(1): 50–54.
王彬, 蔣世磊. 卡塞格林系統(tǒng)計(jì)算機(jī)輔助裝調(diào)技術(shù)研究[J]. 光學(xué)儀器, 2008, 30(1): 50–54.
[8] Sun J W, Lv T Y, Yao L S,. Design and assembly of transmitter-telescope[J]., 2014, 22(2): 369–375.
孫敬偉, 呂天宇, 姚麗雙, 等. 發(fā)射望遠(yuǎn)鏡的設(shè)計(jì)與裝調(diào)[J]. 光學(xué)精密工程, 2014, 22(2): 369–375.
[9] Zhang X M, Jiang F, Kong L Y,. Research on optical alignment technology for Cassegrain system[J]., 2015, 36(4): 526–530.
張向明, 姜峰, 孔龍陽(yáng), 等. 卡塞格林系統(tǒng)光學(xué)裝調(diào)技術(shù)研究[J]. 應(yīng)用光學(xué), 2015, 36(4): 526–530.
[10] Gu Z Y, Yan C X, Li X B,. Application of modified sensitivity matrix method in alignment of off-axis telescope[J]., 2015, 23(9): 2595–2604.
顧志遠(yuǎn), 顏昌翔, 李曉冰, 等. 改進(jìn)的靈敏度矩陣法在離軸望遠(yuǎn)鏡裝調(diào)中的應(yīng)用[J]. 光學(xué)精密工程, 2015, 23(9): 2595–2604.
Application of two step sensitivity matrix method in Cassegrain telescope alignment
Cao Yuze1,2, Ma Wenli1*
1Institute of Optics and Electronics, Chinese Academy of Sciences, Chengdu, Sichuan 610209, China;2University of Chinese Academy of Sciences, Beijing 100049, China
Layout of Cassegrain system
Overview:With the development of telescope technology, the aperture and field of view of telescope are becoming larger and larger, the structure of optical system is becoming more and more complex, and the difficulty of fabrication and assembly is also increasing. The speckle pattern of the focal plane of the optical system can be measured by interferometer and other testing equipment, and the Zernike coefficients can be calculated by the speckle pattern. For Cassegrain telescope, in order to obtain good imaging quality, it is necessary to correct the position of its secondary mirror. By using computer-aided alignment technology, the optical system can be real-time detected and compared with the theoretical results. By establishing a mathematical model between Zernike coefficient and misalignment, the misalignment of the components can be corrected accurately. The most widely used computer-aided alignment method is the sensitivity matrix method. Sensitivity matrix method is a method of correcting aberration by establishing mathematical model of misalignment and Zernike coefficient on the basis of analyzing aberration characteristics. The traditional sensitivity matrix method only carries out single correction. According to the meaning of Zernike coefficient,3、4、5、6、7and8are chosen to construct the sensitivity matrix. Based on the analysis of the shortcomings of the traditional sensitivity matrix method of the two order model, a fine tuning step was added based on the characteristics of the sensitivity matrix. The calculation method of sensitivity is improved. According to the relationship between misalignment and Zernike coefficient, the selection principle of Zernike coefficient for constructing sensitivity matrix is proposed. The traditional sensitivity matrix method is improved. For the Cassegrain system, the relationship between the Zernike coefficients and the misadjustment was analyzed, and the calibration simulation of Cassegrain system with 300 mm aperture and 0.6° field of view was carried out. The simulation results show that after correction by traditional sensitivity matrix method, the mean values of offset along,,axes and tilt around,axes are -0.0684 mm, -0.0892 mm, 0.0015 mm, 0.0498° and -0.0444°, respectively, and the full field wavefront aberration RMS is less than 0.1(=632.8 nm). After correction by two step sensitivity matrix correction method, the mean values are -0.0018 mm, -0.0012 mm, 0.0002 mm, 0.0008° and -0.0012°, respectively, and the full field wavefront aberration RMS is less than 0.03. The corrected optical system reaches the diffraction limit and approaches the design position, which is obviously superior to the traditional sensitivity matrix method.
Citation: Cao Y Z, Ma W LApplication of two step sensitivity matrix method in Cassegrain telescope alignment[J]., 2020, 47(2): 180536
Application of two step sensitivity matrix method in Cassegrain telescope alignment
Cao Yuze1,2, Ma Wenli1*
1Institute of Optics and Electronics, Chinese Academy of Sciences, Chengdu, Sichuan 610209, China;2University of Chinese Academy of Sciences, Beijing 100049, China
In order to adjust the position of the secondary mirror of Cassegrain telescope with large field of view, a computer aided adjustment method of two-step sensitivity matrix model was proposed. Based on the analysis of the shortcomings of the traditional sensitivity matrix method of the two order model, a fine tuning step was added based on the characteristics of the sensitivity matrix and the traditional sensitivity matrix method was improved. For the Cassegrain system, the relationship between the Zernike coefficients and the misalignment was analyzed, and the calibration simulation of Cassegrain system with 300 mm aperture and 0.6° field of view was carried out. The simulation results show that after correction by traditional sensitivity matrix method, the mean values of offset along,,axes and tilt around,axes are -0.0684 mm, -0.0892 mm, 0.0015 mm, 0.0498° and -0.0444°, respectively, and the full field wavefront aberration RMS is less than 0.1(=632.8 nm). After correction by two step sensitivity matrix correction method, the mean values are -0.0018 mm, -0.0012 mm, 0.0002 mm, 0.0008° and -0.0012°, respectively, and the full field wavefront aberration RMS is less than 0.03, which is obviously superior to the traditional sensitivity matrix method.
aberration correction; sensitivity matrix method; computer-aided alignment; Cassegrain system
TH743
A
10.12086/oee.2020.180536
: Cao Y Z, Ma W L. Application of two step sensitivity matrix method in Cassegrain telescope alignment[J]., 2020,47(2): 180536
2018-10-22;
2019-03-13
曹宇澤(1994-),男,碩士研究生,主要從事大口徑望遠(yuǎn)鏡的像差檢測(cè)與校正方面的研究。E-mail:qfwxdc@163.com
馬文禮(1962-),男,研究員,博士生導(dǎo)師,主要從事光電探測(cè)、精密機(jī)械、光電探測(cè)系統(tǒng)總體技術(shù)的研究及大型光電望遠(yuǎn)鏡的研制。E-mail:mawenli@ioe.ac.cn
曹宇澤,馬文禮. 兩步式靈敏度矩陣法在卡塞格林望遠(yuǎn)鏡裝調(diào)中的應(yīng)用[J]. 光電工程,2020,47(2): 180536
* E-mail: mawenli@ioe.ac.cn