摘要
基于辐射传递求解的改进二流近似和全局智能优化的遗传算法,结合材料的法向-半球反射率和透过率,构建了纳米隔热材料吸收系数和散射系数的辨识模型。采用文献中报道的两种玻璃的辐射物性参数数据数值验证了辨识方法的可靠性。实验测量了纳米隔热材料法向-半球透过率和反射率数据,辨识获得材料在0.4~7.0μm波段的吸收系数和散射系数。研究表明:(1)构建的辨识方法能准确辨识材料的吸收和散射系数;(2)在0.4~7.0μm内,材料的吸收系数介于70~3900 m^-1,散射系数在180~3000 m^-1之间,不同波长下数值差异大,呈现强烈的光谱选择性;(3)在2.5μm以下材料内散射所占份额较大,在2.5μm以上吸收明显占优,整体上呈现出强吸收、弱散射特征。
An inverse method for retrieving absorption and scattering coefficients of nanoporous thermal insulation is proposed on the basis of a modified two flux method for solving radiative transfer,a genetic algorithm based global identification method,and normal-hemispherical reflectance and transmittance measurements.First,the inverse method is numerically validated by using radiative properties of two types of glass reported in the literature.Then,the normal-hemispherical reflectance and transmittance of nanoporous thermal insulation are measured,and the absorption and scattering coefficients of the material at wavelength between 0.4μm and 7.0μm are retrieved.The results show that(1)the proposed inverse method is able to retrieve accurately the absorption and scattering coefficients of nanoporous thermal insulation;(2)for wavelength of 0.4—7.0μm,the retrieved absorption coefficient is ranging from 70 m-1 to 3900 m^-1,while the scattering coefficient is between 180 m^-1 and 3000 m^-1,the values vary in a wide range for different wavelengths,and a strong spectral selectivity can be observed;and(3)for wavelength below 2.5μm,the fraction of scattering is larger than those for absorption,while absorption dominates for wavelength larger than 2.5μm,the material featuring strong absorption and weak scattering.
作者
刘华
李健
王江
陈忠灿
段彦军
张行周
蔡霖
Liu Hua;Li Jian;Wang Jiang;Chen Zhongcan;Duan Yanjun;Zhang Xingzhou;Cai Lin(Beiing Institute of Mechanical Equipment,Beijing,100854,China)
出处
《南京航空航天大学学报》
EI
CAS
CSCD
北大核心
2019年第S01期38-43,共6页
Journal of Nanjing University of Aeronautics & Astronautics
关键词
纳米隔热材料
吸收系数
散射系数
辨识
遗传算法
nanoporous thermal insulation
absorption coefficient
scattering coefficient
identification
genetic algorithm