Improved global model for predicting gas radiative properties over a wide range of conditions

Calculating radiative transfer in gas mixtures, especially those containing CO2 and H2O, is important for high temperature applications, such as boilers furnaces. Of all the models that can perform calculations, weighted-sum-of-gray-gas (WSGG) model currently most widely used one solving engineering problems, due to its excellent computational efficiency. However, validity range of a WSGG usually limited because accuracy strongly depends on conditions at which coefficients are generated. In this work, general developed predicting total emissivities mixtures CO2, any non-absorbing gases. The be wide conditions: temperatures 300–2900 K, pressures 1–60 bar path lengths 0.01–100 m mixture compositions. based artificial neural networks (ANN) trained using around 4.5 million samples generated by statistical narrow band model. Two ANNs were selected—considering both speed—and compared against existing models. first ANN has more than an order magnitude smaller mean error models, speed comparable second even higher model, but with slightly lower speed.