To enhance the cost-effectiveness of smoothing wind power fluctuations, this study develops a novel hybrid energy storage system power allocation strategy that integrates an improved Kalman filter with variational mode decomposition (VMD). First, the Kalman filter is used for initial suppression of wind power fluctuations. The optimal weighting coefficients are determined based on the Pareto frontier to construct an objective function incorporating both fluctuation magnitude and energy storage consumption. Simulated annealing is then employed to solve for the optimal Q and R values, thereby achieving balanced suppression of wind power fluctuations. Subsequently, the power allocation strategy, which fully considers the characteristics of batteries and supercapacitors, is implemented through VMD. The Grey Wolf Optimizer is introduced to determine the optimal parameter combination [k, α] for VMD. Finally, simulation results based on actual wind power generation data demonstrate that the proposed strategy can effectively mitigate power fluctuations, reduce energy storage consumption, and consequently improve overall system economic performance.

Aiming at the uncertainty of wind and storage power plant output and price volatility in the electricity market, this study proposes a collaborative bidding framework that integrates hybrid tariff prediction and dynamic risk management. First, a hybrid tariff prediction model combining independent component analysis and time series network is constructed to provide data support; second, a dynamic Conditional Value-at-Risk threshold model is designed to adaptively switch between aggressive and conservative bidding strategies based on the predicted value of tariff fluctuations; finally, a spread-driven dual-market synergistic mechanism is set up to dynamically adjust bidding ratios and storage scheduling schemes by comparing the day-ahead and real-time tariffs. The empirical analysis shows that this strategy can effectively improve the economic return and operational adaptability of the wind storage system, and verifies the core value of energy storage in multi-market synergy.

Configuring energy storage can effectively reduce the abandonment of wind and solar energy, thereby enhancing the consumption capacity of new energy. In this paper, a power grid electricity balance model was established, and the "renewable energy consumption characteristic curve" was extracted. Secondly, an "equal integration principle" optimization model for energy storage and renewable energy consumption was established, incorporating the consideration of battery capacity degradation. Targeting the annual total consumption of renewable energy as the goal, the model seeks a balance between surplus and deficiency in energy storage capacity. Energy storage is configured based on the characteristics of renewable energy, and renewable energy installation is planned based on the principle of system balance, achieving a coordinated planning strategy. This approach realizes the four-dimensional organic unity of "renewable energy characteristics," "renewable energy consumption goals," "energy storage configuration," and "system balance."

The increasing integration of distributed generation (DG) into the distribution network has led to greater complexity in power flow distribution and transient current characteristics during single-phase grounding faults. As a result, traditional fault location methods are no longer sufficient. Adapting existing fault location methods to accommodate changing permeability has become an urgent issue. In response, a fault location method utilizing a convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) network is proposed. The CNN is used to extract detailed longitudinal features from fault zero sequence current data at a specific time, compressing the data length to reduce subsequent network training parameters. Additionally, a cascade network with BiLSTM as its core is constructed to capture historical horizontal features of fault data during the fault evolution process. An Attention mechanism is integrated to ensure that the model focuses on changes in fault time and location data, thereby enhancing fault location accuracy. The simulation results demonstrate that the proposed method is capable of accurately identifying single-phase grounding faults, offering high precision and robustness in positioning, and exhibiting strong adaptability across various permeability fault scenarios.

In order to improve the safety of electric vehicle charging pile operation, a charging pile performance evaluation method is proposed by analyzing the current situation of charging pile performance evaluation for the problem that charging pile performance faults are difficult to identify. Firstly, the EV charger performance evaluation index system is established by analyzing the technical standards of EV non-vehicle chargers and combining them with the actual situation. Secondly, objective weights based on random forest and subjective weights based on gray correlation analysis are introduced to improve the rationality of the combination weights. Then, the affiliation function of the normal cloud model, which conforms to the state distribution of charger performance, is established and combined with the combined weights to realize the performance evaluation of EV chargers. Finally, the operational data of one charger is taken as an example for performance evaluation and analysis, and the results show that the proposed evaluation method can discover the weak links of the charger in time, which verifies the feasibility and practicability of the method.