Transformers in the distribution network must have their carbon emissions analyzed in order to meet the “double carbon” goal. Using an oil-immersed distribution transformer as the research object, this paper develops a “cradle-to-grave” carbon accounting model. A life cycle assessment (LCA)-based methodology is proposed to account for carbon emissions and total energy demand over the full life cycle and further analyze the carbon emissions within each stage. A case study is presented using data from a 200 kVA transformer manufactured by a transformer plant in Xinjiang. According to the data, there are 112.18 t of carbon emissions and 798.21 GJ of total energy consumed. Distribution network transformers have carbon emissions of 282 kg, 782 kg, 122.96 kg, 11,079.64 kg, and −88.6 kg throughout the manufacture and manufacturing stage, transportation stage, construction and installation stage, operation stage, and waste recycling stage, respectively. There are 4.12 GJ, 9.06 GJ, 1.34 GJ, 785.97 GJ, and −0.28 GJ in total energy requirements. According to the study, which covered 99.03% of the entire life cycle, the operation stage had the largest percentage of carbon emissions. In addition to streamlining the production process and using more energy-efficient equipment, the waste recovery stage successfully decreased the environmental impact of carbon emissions. Sensitivity analysis shows that the silicon steel sheet and transformer oil has a significant impact on the carbon emissions of distribution network transformers during the life cycle, and the higher the grade of silicon steel sheet, the lower the carbon emissions, and the synthetic ester transformer oil has the most comprehensive performance and the lowest carbon emissions.

Against the backdrop of large-scale integration of renewables into the power grid, frequent load-following operation of thermal power units substantially increases the difficulty of controlling boiler NOx emissions. Accurate forecasting of boiler NOx emissions is crucial for guiding efficient and clean operation under such flexible operating conditions. However, under frequent load-following conditions, NOx dynamics are highly nonlinear and non-stationary, making it challenging to achieve accurate prediction using only time-domain information. To address these issues, we propose DTTE-Net, a time–frequency feature fusion framework for predicting SCR-inlet NOx concentration in coal-fired boilers. DTTE-Net consists of three components: a time-domain branch, a frequency-domain branch, and a gated feature fusion module. The time-domain branch captures short-term fluctuations and long-range temporal dependencies, while the frequency-domain branch extracts complementary spectral representations to enhance the characterization of non-stationary fluctuations. The gated feature fusion module then adaptively integrates the two-domain features by using a gated mechanism and produces the NOx concentration forecast. In addition, a Gaussian kernel-based loss is introduced to improve robustness to nonlinear error structures. Experiments on real distributed control system data from a 660 MW ultra-supercritical coal-fired unit show that DTTE-Net outperforms existing baseline models, achieving lower forecasting errors and higher R².