المفاهيم الأساسية
Developing a simple MLP-based FSCK model for accurate and efficient spectral predictions.
الملخص
Introduction to radiation modeling in high-temperature environments.
Challenges in modeling radiative properties for gases.
Evolution from line-by-line method to full-spectrum k-distribution (FSK) method.
Development of the full-spectrum correlated k-distribution (FSCK) method.
Comparison of different FSCK models, including table, TFM, and SFM models.
Structure and training process of the SFM model.
Implementation and efficiency comparison of different FSCK models.
Results and discussions on flame calculations using various models.
Conclusion on the accuracy, efficiency, and portability of the SFM model.
الإحصائيات
"Results show that the SFM model can achieve excellent accuracy that is even better than look-up tables at a tiny computational cost that is far less than that of TFM model."
"The storage of each neuron mainly concerns the values of weight and bias."
"The optimal hyperparameters of SFM model are 1.246×10^-3 for initial rate and 10^-7 for regularization factor."
اقتباسات
"The SFM model is not only more robust than the TFM model but also sufficient for radiative calculations."
"The SFM model gives better predictions of κP compared to the TFM model."