基于粒子群优化的LAMOST初始波长定标方法

发布时间：2018-01-09 15:29

本文关键词：基于粒子群优化的LAMOST初始波长定标方法　出处：《天文学报》2011年05期 　论文类型：期刊论文

【摘要】：大天区面积多目标光纤光谱天文望远镜(Large Sky Area Multi-Object FiberSpectroscopic Telescope,LAMOST)初始波长定标操作,对于在先验定标系数附近搜索解空间内的每一点,通过插值法生成对应定标系数下的虚拟定标灯谱,再将生成的虚拟定标灯谱与实测定标灯谱作互相关运算,使得相关系数最大时对应的定标系数即为初始定标结果,其实质是一个多参数寻优问题.粒子群优化是一种基于群体智能的随机全局优化算法,具有实现简单、精度高、收敛快的特点.鉴于粒子群优化的优异性能,提出一种基于粒子群优化的LAMOST初始波长定标方法,并设计相应算法和测试实验.实验结果表明,基于粒子群优化的LAMOST初始波长定标在收敛性、解的质量、运行时间等方面都优于基于改进遗传算法参数寻优方法.
[Abstract]:Large Sky area Multitarget Optical Fiber Spectral Telescope (. Large Sky Area Multi-Object FiberSpectroscopic Telescope. LAMOST initial wavelength calibration operation, for each point in the solution space search near the prior calibration coefficient, the virtual calibration lamp spectrum under the corresponding calibration coefficient is generated by interpolation method. Then the generated virtual calibration lamp spectrum is cross-correlated with the actual measurement lamp spectrum, so that the calibration coefficient corresponding to the maximum correlation coefficient is the initial calibration result. Particle swarm optimization (PSO) is a stochastic global optimization algorithm based on swarm intelligence, which has the advantages of simple implementation, high precision and fast convergence. A LAMOST initial wavelength calibration method based on particle swarm optimization (PSO) is proposed, and the corresponding algorithm and experimental results are designed. The LAMOST initial wavelength calibration based on particle swarm optimization is superior to the improved genetic algorithm in terms of convergence, solution quality and running time.
【作者单位】：中国科学技术大学信号统计处理研究室;
【基金】：国家自然科学基金项目(11078016)资助
【分类号】：P111.2
【正文快照】： 1引言郭守敬望远镜，即LAMOST，，是一架已建成的具有4m口径、20m焦距及5o视场的反射施密特望远镜，它是目前世界上光谱获取率最高的望远镜[‘一2].该望远镜上布段4 000根光纤并平均分配到16台多目标光纤光谱仪上.在每台光谱仪内部，来自光纤的光谱经分光镜分为红(5700一9000A)

【参考文献】