星系的质量、年龄、金属丰度、恒星形成率、尘埃质量等基本物理参数的确定是认识星系形成和演化的基础，而星系的多波段能谱分布是确定这些物理参数的主要信息来源。因此，如何通过对星系多波段能谱的分析来可靠地确定其基本物理参数是人们认识星系的形成和演化时面临的一个重要而又基本的问题。然而，目前在观测和理论上存在的诸多不确定性和复杂性因素使得通过能谱分析方法来确定星系的基本物理参数仍然是一个富有挑战性的问题。有鉴于此，申请者准备在已有工作的基础上发展基于贝叶斯推理和人工神经网络的一套可靠、高效、通用的能谱分析方法，并且通过应用于红外明亮星系的基本物理参数确定而对整套方法进行验证。

The determination of basic physical parameters of galaxies， such as stellar mass, age, metallicity, star formation rate and dust mass, is the basis for our understanding of the formation and evolution of galaxies. Meanwhile, the spectral energy distributions (SEDs) of galaxies are the main source of information for the determination of these parameters of galaxies.So, how to reliably determine the basic physical parameters of galaxies from the analysis of their muti-wavelength SEDs is one of the most important and basic problem for our understanding of the formation and evolution of galaxies.However, due to some unresolved uncertainties and complexity in both the observational data sets and theoretical models, the determination of basic physical parameters of galaxies from SED fitting is still challenging.Given these, we are going to develop a suite of general methods for the reliable and efficient analysis of galaxy multi-wavelength SEDs based on Bayesian inference and artificial neural network. Besides, from the application of these methods to the physical parameter determination of IR-luminous galaxies, we are going to test these methods.