Calculation Method of Infrared Radiation from Rocket Plume based on LS-SVM Model

With the rapid development of hypersonic and stealth technology, the technology of detecting and identifying targets through infrared radiation signals from the rocket plume during the working process of high-speed vehicles has received more and more attention. However, the generation mechanism of the infrared radiation signal from the rocket plume is very complicated, and the infrared radiation signal is affected by the flight environment and detection environment during the space transmission process, which makes the calculation of the infrared radiation from the rocket plume very complicated and time-consuming. Therefore, this paper proposes a method of calculating the infrared radiation from the rocket plume based on the least squares support vector machine (LS-SVM) and the kriging model, which is essentially a learning process for numerical calculation method of the infrared radiation from the rocket plume. This method establishes the least-squares support vector machine regression model and the kriging model of the infrared radiation from the rocket plume, respectively. Compared with the traditional numerical calculation methods, the least-squares support vector machine regression model and kriging model are simple and fast, and the calculation time for a single sample is only three thousandths of that of the traditional numerical calculation methods, and the average relative errors of the calculation results and the results of the traditional numerical calculations are within 5%, among which, the least-squares support vector machine regression model is more accurate than the kriging model. It is proved that the least-squares support vector machine regression model can effectively reduce the amount of engineering calculations without affecting the accuracy, and realise the high-volume calculation of infrared radiation of rocket plume.