确定恒星表面有效温度的非参数估计方法

恒星表面有效温度是恒星的一个重要物理参量，是恒星光谱差异的重要因素。文章采用非参数估计算法对恒星表面温度进行估计。首先对历史光谱数据进行主成分分析(PCA)处理，再根据PCA特征数据与其表面温度的对应关系建立温度的估计模型，该模型是基于高斯核函数的。方法不依赖对光谱进行精确测量，就可以得到较高估计精度的温度值，对大样本光谱分析具有重要意义。

Non-Parameter Estimation Algorithm to Determine Stellar Effective Temperature

The effective temperature of a star is one of the most important parameters, which determine the continuum and spectral lines in the stellar spectrum. A non-parameter estimation algorithm is proposed to estimate the stellar effective temperature in the present paper. Firstly, the spectrum data is processed by principal component analysis(PCA), then, an estimating model based on a Gaussian kernel function is set up using the PCA data and their temperatures. Experiments were carried out to verify the efficiency, and numerical robustness of the algorithm is also tested.