The global horizontal irradiance (GHI) is the most important metric for evaluating solar resources. The accurate prediction of GHI is of great significance for effectively assessing solar energy resources and selecting photovoltaic power stations. Considering the time series nature of the GHI and monitoring sites dispersed over different latitudes, longitudes, and altitudes, this study proposes a model combining deep neural networks and deep convolutional neural networks for the multi-step prediction of GHI. The model utilizes parallel temporal convolutional networks and gate recurrent unit attention for the prediction, and the final prediction result is obtained by multilayer perceptron. The results show that, compared to the second-ranked algorithm, the proposed model improves the evaluation metrics of mean absolute error, mean absolute percentage error, and root mean square error by 24.4%, 33.33%, and 24.3%, respectively.

The deployment of energy storage systems can play a role in peak and frequency regulation, solve the issue of limited flexibility in cleaner power systems in China, and ensure the stability and safety of the power grid. This paper presents a comprehensive environmental impact analysis of a lithium iron phosphate (LFP) battery system for the storage and delivery of 1 kW-hour of electricity. Quantities of copper, graphite, aluminum, lithium iron phosphate, and electricity consumption are set as uncertainty and sensitivity parameters with a variation of [90%, 110%]. The results show that global warming potential is 9.08E+01 kg CO₂ eq. and fossil resource use is 1.21E+03 MJ, with uncertainty ranges of [8.54E+01, 9.23E+01] and [1.15E+03, 1.23E+03], respectively. Electricity consumption during the manufacture and installation process is the greatest contributor to climate change (CO₂ eq. emissions), accounting for 39.71% and largely owing to non-renewable sources, followed by cathode materials at 27.85% and anode materials at 18.36%. The disposal and recycling process offers emission reductions but requires an additional 1.17% use of fossil resources. Sobol T indices for the quantity of electricity are the highest for acidification, climate change, fossil resource use, and ionizing radiation. By considering the pathway of China’s electricity mix from 2020 to 2050 under the Paris Agreement’s 2° target, the potential for environmental emission reduction in the system is evaluated. The results show that the greener electricity mix could lead to a 24.59% reduction in acidification impact, a 35.74% reduction in climate change impact, a 33.24% reduction in fossil resource use, and a 44.13% reduction in ionizing radiation impact. This study offers a comprehensive view of the environmental impact reductions associated with the lithium iron phosphate battery and its industry.

With the support of the dual-carbon strategy and related policies, China’s offshore wind power has experienced rapid development. However, constrained by the inherent intermittency and volatility of wind power, large-scale expansion poses significant challenges to grid integration and exacerbates government fiscal burdens. To address these critical issues, this paper proposes a multi-market coupling trading model integrating energy storage-equipped offshore wind power into electricity–carbon–green certificate markets for large-scale grid networks. Firstly, a day-ahead electricity market optimization model that incorporates energy storage is established to maximize power revenue by coordinating offshore wind power generation, thermal power dispatch, and energy storage charging/discharging strategies. Subsequently, carbon market and green certificate market optimization models are developed to quantify Chinese Certified Emission Reduction (CCER) volume, carbon quotas, carbon emissions, market revenues, green certificate quantities, pricing mechanisms, and associated economic benefits. To validate the model’s effectiveness, a gradient ascent-optimized game-theoretic model and a double auction mechanism are introduced as benchmark comparisons. The simulation results demonstrate that the proposed model increases market revenues by 17.13% and 36.18%, respectively, compared to the two benchmark models. It not only improves wind power penetration and comprehensive profitability but also effectively alleviates government subsidy pressures through coordinated carbon–green certificate trading mechanisms.

The configuration of energy storage systems in offshore wind farms can effectively suppress fluctuations in wind power and enhance the stability of the power grid. However, the economic balance between the cost of energy storage systems and the fluctuations in wind power remains an urgent challenge to be addressed, especially against the backdrop of widespread spot trading in the electricity market. How to achieve effective wind power stabilization at the lowest cost has become a key issue. This paper proposes three different energy storage configuration strategies and adopts the non-dominated sorting genetic algorithm (NSGA-II) to conduct multi-objective optimization of the system. NSGA-II performed stably in dual-objective scenarios and effectively balanced the relationship between the investment cost of the energy storage system and power fluctuations through the explicit elite strategy. Furthermore, this study analyzed the correlation between the rated power and rated capacity of the energy storage system and the battery life, and corrected the battery life of the Pareto frontier solution obtained by NSGA-II. The research results show that when only considering the investment cost of the energy storage, the optimal configuration was a rated power of 4 MW and a rated capacity of 28 MWh, which could better balance the investment economy and power fluctuation. When further considering the participation of energy storage systems in the electricity spot market, the economic efficiency of the energy storage systems could be significantly improved through the fixed-period electricity price arbitrage method. At this point, the optimal configuration was a rated power of 8 MW and a rated capacity of 37 MWh. The corresponding project investment cost was CNY 242.77 million, and the annual fluctuation rate of the wind power output decreased to 17.84%.

To enhance the prediction accuracy of offshore wind speed, this study employs an interpolation algorithm to improve spatial resolution based on the ERA5 reanalysis dataset. The objective is to identify the optimal interpolation method and apply it to wind energy assessments in the South China Sea. This paper compares the interpolation effects and accuracy of Linear, Cubic, and Bicubic interpolation methods on wind speed data, with the optimal method subsequently applied to evaluate wind resources in the South China Sea for 2023. The findings indicate that, while different interpolation methods minimally affect the correlation of wind speed data, there are notable differences in their impact on overall accuracy. The Cubic interpolation method proved to be the most effective, tripling spatial resolution and reducing wind speed errors in ERA5 data by 26%. Using this method, wind resource assessments were conducted in selected areas of the South China Sea. Results reveal that the annual available operational hours for wind turbines in most parts of the region range from 2000 to 4000 h, with fluctuations in turbine output power increasing alongside available operational hours.

Wind power prediction plays a significant role in enhancing the effectiveness of power system operation and decision‐making. Given the inherent stochastic nature of meteorological events, achieving highly accurate forecasts for wind power poses considerable challenges. To address this challenge, this paper initially leverages the time series learning capability of recurrent neural networks (RNN) to extract sequential information from historical wind power data. Subsequently, the information extracted from the convolutional layer is transferred to the light gradient boosting machine (LGBM), utilizing the feature extraction capability of convolutional neural networks (CNN). Furthermore, an optimal weighted combination is employed for the short‐term prediction of wind power. Finally, a multi‐step wind power prediction method of integrated CNN–RNN–LGBM is proposed in this paper. Simulation results demonstrate that the proposed CNN–RNN–LGBM framework outperforms other models during global training. Meanwhile, transferring the information from CNN to LGBM can improve its performance, proving the feature extraction ability of CNN.