The flow behavior of molten slag directly impacts the stability of slag removal processes in high-temperature furnaces, with alkali metal oxides exerting a significant influence on slag flow characteristics. To elucidate the structural evolution mechanism by which Na₂O modulates the viscosity-temperature properties of silicate slag, molecular dynamics simulations, FactSage thermodynamic calculations, and Raman spectroscopy analysis are employed to investigate the effects of NaO content on slag structure and viscosity under fixed SiO₂/Al₂O₃ conditions. Results indicate that as Na₂O content increases within the 0 wt%-8.91 wt% range, the complex structural unit [Si₂O₅]²⁻(Q³) continuously depolymerizes into [SiO₄]⁴⁻(Q⁰) and [Si₂O₇]⁶⁻(Q¹), leading to increased non-bridging oxygen (NBO) proportion and decreased polymerization degree (DOP). Concurrently, Na⁺ compensates for Al³⁺ charges, promoting the transformation of unstable [AlO₆] octahedra into more stable [AlO₄] tetrahedra. The network depolymerization effect of Na₂O dominates its macroscopic viscosity-temperature behavior, resulting in a significant viscosity decrease with a good linear correlation between viscosity and degree of polymerization. Below the liquidus temperature range, when Na₂O<8.91 wt%, its addition suppresses slag crystallization tendencies. This promotes the transformation of high-melting-point minerals like wollastonite and calcium feldspar into low-melting-point feldspars, thereby enhancing slag fluidity and reducing viscosity. This study reveals the microscopic mechanism by which Na₂O regulates the viscosity-temperature characteristics of silicate slag, providing a theoretical basis for safe slag disposal.

Coal-fired power plants serve as primary sources of NO_x emissions, and the efficient operation of SCR denitrification systems is crucial for reducing pollutant emissions. However, the highly dynamic changes of data during NO_x prediction processes limit the accuracy of predictive models. Therefore, a hybrid prediction framework based on modal energy difference and sample entropy, which combining variational mode decomposition (MEVMD) with genetic algorithm (GA) to optimize convolutional neural network (CNN) and long short-term memory network (LSTM), is proposed. Firstly, abnormal data are corrected using the 3σ criterion; 20 key auxiliary variables are selected via Pearson correlation coefficients. The maximum information coefficient (MIC) is employed to determine the delay time for each variable, achieving temporal alignment between features and target variables. Secondly, adaptive variational mode decomposition (VMD) precisely extracts multi-frequency features from NO_x time-series signals. Hyperparameters are optimized via GA to achieve adaptive modeling of multiple sub-modes. Finally, prediction results are generated through data reconstruction. Experimental results demonstrate that the proposed model achieves RMSE of 0.9492, MAE of 0.4969, and R² of 0.9767, outperforming comparison models.