NO_x emission prediction based on deep Boltzmann machine integrated with least square support vector regression

By analyzing the relationship between flame radical images and NO_x emission in combustion process, a prediction algorithm of NO_x emission was proposed based on deep Boltzmann machine integrated with least square support vector regression. The specific way is to use deep Boltzmann machine to automatically learn the image features of four flame radical images (OH^*, CN^*, CH^* and C₂^*), and then adopt least square support vector regression to establish the relationship between image features and NO_x emission, so as to further predict the NO_x emission. Results show that the predicted value of NO_x emission is in good agreement with the reference data. Compared with various image-based NO_x emission prediction algorithms, the proposed method has significant advantages in prediction accuracy.