Guangxi Hydraulic Power Machinery Research Institute

To achieve accurate prediction and control of shield tunneling attitude under complex geological conditions, this study takes a water conveyance tunnel project in Western China, which crosses the Shaoling Plateau with Class V surrounding rock, as the research object, and conducts shield tunneling attitude prediction and parameter sensitivity analysis. A total of 604 sets of tunneling data were collected from loess strata and high-pressure water-rich sections, and median filtering was used for noise reduction. An evaluation system containing 9 types of algorithms including ensemble learning and linear regression was constructed. Parameters were optimized through grid search and 5-fold cross-validation, and predictions were made with the horizontal and vertical deviations of the shield head/tail as outputs. The sensitivity of input parameters to different attitudes was analyzed. The results show that the ensemble learning algorithm has the best prediction performance, with error indicators significantly lower than those of linear and most nonlinear models, and the goodness of fit reaches 0.70, which can effectively capture the dynamic changes of attitudes. Parameter sensitivity analysis indicates that the attitude deviations of different parts and directions are affected by parameters with specificity, while total thrust, cutter head speed/torque, and soil chamber pressure are common sensitive parameters. This research provides technical support and method references for engineering construction and similar scenarios.

In the transformer operation process, as the time continues to rise, the transformer winding and core will generate heat. Transformer oil, as an important heat dissipation and insulation part of the transformer, absorbs the heat generated by the transformer winding and the iron core, resulting in varying degrees of aging of the transformer oil, which increases the conductivity of the transformer oil and the relative dielectric constant, resulting in a decline in the insulation performance of the transformer. This phenomenon is the main cause of transformer aging breakdown. Therefore, the electric field calculation of transformer insulation aging degree considering the influence of temperature is studied. First, the transformer temperature experiment is carried out with the use of a temperature control box to make the transformer reach the predetermined temperature. Second, the accelerated aging experiment, power frequency voltage withstand experiment, and transformer oil dielectric strength experiment are carried out to obtain the power frequency voltage withstand experiment results and the relative dielectric constant data of transformer oil under different aging conditions, and the experimental data are brought into the electric field calculation by Fourier function fitting. Finally, under different aging conditions of transformer oil, electric field simulation and numerical calculation were carried out and calculation results of the electric field are obtained.

As China advances its National Woody Oilseed Strategy (2021–2035) to enhance edible oil self-sufficiency, reconciling camellia oil production with carbon neutrality goals demands precise emission diagnostics. In this study, a new forest-to-oil module was developed, utilizing operational data from 37 forest farms and processing plants in China. In accordance with the GHG Protocol Corporate Standard (Scope 1, 2 & 3), the LCA system for camellia oil spans from the nursery gate to the oil-mill gate, covering all direct (Scope 1), indirect energy (Scope 2) and upstream value-chain (Scope 3) greenhouse gas (GHG) emissions (assessed with global warming potential, GWP) associated with both forest farm operations and oil mill processes. Results show a carbon intensity of 3.91 t CO 2 -eq per ton oil, translating to annual sectoral emissions of 7.81 Mt CO 2 -eq based on China’s 2025 camellia oil output (2.00 million tons). Organic fertilizer (45.27% of total) and compound fertilizers (29.91% of total) dominate emission sources. Spatial analysis identifies three critical provinces—Hunan (29.46%), Jiangxi (22.43%), and Guangxi (12.83%)—collectively responsible for 64.72% of national production emissions, the disparity stems from regional grid emissions and interprovincial transport distance variations. Scenario modeling demonstrates 4.16–14.57% emission reduction potential through: (1) precision fertilization targeting 25%–30% nitrogen efficiency improvement, (2) renewable energy integration in processing (40% solar/wind penetration), and (3) intermodal logistics optimization. These findings provide spatially explicit mitigation pathways, emphasizing the necessity of province-specific policies balancing oil security and decarbonization targets. These insights not only guide forestry units in lowering their greenhouse gas emissions but also highlight the importance of sustainability in the camellia oil industry, offering essential support for its enhancement. • LCA of camellia oil production was based on the forest farms and enterprises in China. • Life cycle carbon footprint of camellia oil production is 3.91 tCO 2 eq per ton products. • Organic fertilizers (45.27%), compound fertilizers (29.91%) and electricity (9.05%) dominate emissions. • Mitigation potential (38–52%) lies in precision fertilization, renewables, and logistics optimization. • Hunan, Jiangxi, and Guangxi account for 64.72% of China’s camellia oil carbon footprint.