To explore methods of reducing NOx emission from pulverized coal boilers, the effects of injecting ammonia solution and pyrolysis gas into the furnace on NOx emission were experimentally investigated on a 75 t/h pulverized coal boiler. Results show that the deep air staging with 30% separated over fire air (SOFA) creates a high temperature and strong reducing atmosphere in the reducing zone, providing the prerequisites for NOx reduction by ammonia solution and pyrolysis gas. Compared with deep air staging itself, NOx emission can be reduced by 16.7% when ammonia solution is injected from the reducing zone with a normalized stoichiometric ratio of 2.0. However, NOx reduction efficiency is largely affected by its injection position. Similarly, NOx emission is decreased by 28.2% through injecting pyrolysis gas with its calorific value of 10% into the furnace, while a further increase of pyrolysis gas input will not increase NOx reduction efficiency. When ammonia solution and pyrolysis gas are simultaneously injected into the furnace under deep air staging conditions, the overall NOx reduction efficiency reaches 92.0% and NOx emission is decreased to 39.1 mg/m³. Considering the increasingly strict NOx emission standard, these findings can provide theoretical and practical guides to the future NOx reduction in pulverized coal boilers.

A hyperspectral image classification method based on a mixed structure with a 3D multi-shortcut-link network (MSLN) was proposed for the features of few labeled samples, excess noise, and heterogeneous homogeneity of features in hyperspectral images. First, the spatial–spectral joint features of hyperspectral cube data were extracted through 3D convolution operation; then, the deep network was constructed and the 3D MSLN mixed structure was used to fuse shallow representational features and deep abstract features, while the hybrid activation function was utilized to ensure the integrity of nonlinear data. Finally, the global self-adaptive average pooling and L-softmax classifier were introduced to implement the terrain classification of hyperspectral images. The mixed structure proposed in this study could extract multi-channel features with a vast receptive field and reduce the continuous decay of shallow features while improving the utilization of representational features and enhancing the expressiveness of the deep network. The use of the dropout mechanism and L-softmax classifier endowed the learned features with a better generalization property and intraclass cohesion and interclass separation properties. Through experimental comparative analysis of six groups of datasets, the results showed that this method, compared with the existing deep-learning-based hyperspectral image classification methods, could satisfactorily address the issues of degeneration of the deep network and “the same object with distinct spectra, and distinct objects with the same spectrum.” It could also effectively improve the terrain classification accuracy of hyperspectral images, as evinced by the overall classification accuracies of all classes of terrain objects in the six groups of datasets: 97.698%, 98.851%, 99.54%, 97.961%, 97.698%, and 99.138%.