With the deepening of Germany's energy transition, accelerating the phase-out of coal power and promoting the integration of a high proportion of renewable energy have become key strategies. The smooth advancement of Germany's coal power phase-out can be attributed to a broad societal consensus on coal phase-out and the government's flexible use of a combination of market mechanisms and administrative directives. In response to the dual challenges of phasing out coal power and the continuous increase in renewable energy penetration, Germany has successfully maintained a high level of security and stability in its power system by continuously optimizing feed-in tariff policies for renewable energy, leveraging balancing groups, enhancing cross-border grid interconnections, and strengthening the supporting role of conventional power sources. This paper provides an in-depth analysis of Germany's coal power phase-out mechanism and its measures to ensure power supply security, summarizes relevant experiences and lessons, and explores the implications of Germany's practices for China. The aim is to offer valuable insights for the construction of a new power system in China under the carbon neutrality goal.

To address the overestimation bias and poor decision accuracy of conventional microgrid cluster energy management methods, an energy management strategy based on improved double deep Q-network is proposed. Firstly, this study constructed a dual-objective value network framework based on clipped double Q-learning, which enhances decision-making precision by suppressing value overestimation bias through parallel computation of temporal difference (TD) targets for dual value networks and clipping high TD target values. And then, a dynamic greedy strategy was adopted to calculate the value function of all possible actions based on the current state, avoiding persistent exploitation of the greedy actions to ensure sufficient exploration and prevent premature convergence of the agent. Finally, a case study of a microgrid cluster with three sub-microgrids was conducted for verification. The simulation results show that compared to the energy management strategies based on model predictive control and conventional double deep Q-network, the proposed method achieves superior optimization performance and convergence characteristics, while reducing system operating costs by 44.62% and 26.39% respectively.

The configuration of electricity-hydrogen hybrid energy storage in active distribution networks is of vital importance for enhancing the economic efficiency of electricity usage, promoting the consumption of renewable energy, reducing voltage fluctuations, and ensuring the balance between supply and demand as well as system stability. Firstly, an active distribution network operation framework containing electricity-hydrogen hybrid energy storage system is designed, and an electric-hydrogen hybrid energy storage model in the distribution network domain is constructed. Then, an upper-level optimization model is constructed to minimize the comprehensive electricity cost, and a lower-level optimization model is established with the goals of minimizing the wind and solar curtailment rate, voltage fluctuation rate, grid purchase fluctuation, and the total load fluctuation. Finally, a modified IEEE 33-node distribution network is used as a case study to validate the effectiveness of the proposed model. The results demonstrate that the model can achieve optimal capacity configuration for electricity-hydrogen hybrid energy storage system, and optimal dispatch of the active distribution networks. Compared to scenarios without energy storage or with single-medium energy storage, the hybrid electricity-hydrogen system proves to be more economically viable and enhances the self-regulating capability of the active distribution networks. Time-of-use electricity prices and seasonal characteristics collectively drive the coordinated operation of the electricity-hydrogen hybrid energy storage system, forming a synergistic dispatching mode characterized by "electrochemical storage providing rapid response + hydrogen storage enabling long-duration regulation."

To address the challenges posed by intermittent fluctuation of high-penetration distributed photovoltaic (DPV) integration to the secure and stable operation of the power system, a market mechanism and trading strategy is proposed for collaborative participation of source-grid-load-storage in distribution-side transactions. Firstly, the distribution-side distributed energy market was constructed, and a market mechanism was designed considering the characteristics of distributed energy. Secondly, an operation model of each entity was established considering electricity transaction among entities, and an optimization model for multi-agent cooperative operation of source-grid-load-storage was established based on Nash bargaining theory. Finally, the distributed alternating direction method of multiplier (ADMM) was employed to obtain bids and profits of all parties. Simulation results show that, after considering the market mechanism and the revenue distribution of the distribution system operator, the overall alliance market revenue increased by 29.8%, and the local PV consumption rate rose by 59.8%, verifying the effectiveness of the proposed market mechanism and trading strategy.

With the growth of global energy demand and the advancement of sustainable development goals, energy management of microgrids, as an important means to address energy supply, improve energy efficiency and promote green energy utilization, is faced with high dimensionality, complexity and dynamic challenges. In this paper, we propose a distributed reinforcement learning-driven energy optimization and management strategy for microgrid clusters, aiming to enhance the efficiency and sustainability of microgrid clusters in energy scheduling and management through intelligent means. Aiming at the challenges of the microgrid cluster, such as large dynamic changes in load and complex topology, adaptive decision-making and collaborative optimization of the microgrid cluster in a distributed environment is achieved by constructing an objective optimization function and introducing a distributed reinforcement learning algorithm; and power generation of the microgrid cluster is achieved by treating each power point in the microgrid as an agent and utilizing information sharing to achieve the maximization of the global benefit and minimization of the power generation cost, storage and load demand management; finally, the results of the real case show that the proposed strategy is able to maintain the dynamic balance between power supply and demand, resulting in a saving of about 18% of the total power generation cost compared to the traditional methodology techniques.

The accounting of cross-regional electricity indirect carbon emissions is crucial to the carbon budget management of major power-sending and power-receiving regions during China's "15th Five-Year Plan" period. To address such issues as significant inaccuracies in traditional territorial-based accounting methods and insufficient integration of green electricity certificates, four alternative accounting methods, including the power source-based emission factor method, the power import-based emission factor method, the net coal-power transfer adjustment method, and the green certificate consumption adjustment method, are proposed to quantitatively assess the scale of carbon emission transfers via cross-regional and inter-provincial electricity transmission in 2023, and the impacts of different accounting methods on regional carbon budgets are compared. The study reveals that the accounting results of indirect carbon emissions from cross-regional electricity are varied significantly with different methods, with fluctuations exceeding 10% in key power-importing regions such as East China and North China; the accounting mechanism based on the "national average emission factor adjusted with green certificates" can dynamically refine the carbon emission responsibilities of power-importing regions. It is recommended that the accounting of indirect carbon emissions from cross-regional electricity transmission should be grounded in regional coordinated development and the efficiency of power resource allocation, with comprehensive consideration of the responsibility-sharing mechanisms between sending and receiving regions as well as the green certificate system.

Accurately assessing the renewable energy accommodation capacity of rural power systems is of significant guidance for the planning and development of new rural power systems. However, due to the low voltage levels, large impedance ratios, and higher network loss ratios in rural power systems, the traditional DC power flow-based time-series production simulations are prone to inaccuracies when evaluating their accommodation capacity. Therefor this paper proposes a time-series production simulation model and solution method based on refined AC power flow. Firstly, considering the voltage and network loss characteristics, a refined operational model for rural power systems based on AC power flow model is established. Then, to address the computational challenges of this model in annual time-series production simulation, the AC power flow is convexified and relaxed using second-order cone relaxation techniques to reduce model complexity. Furthermore, a time segmentation strategy that balances both grid scale and computational efficiency is introduced, It decomposes the time-series production model into multiple subproblems solved via rolling optimization, drastically improving computational performance. Finally, a county-level rural power system was used as an example for simulation testing. and the results verified the effectiveness of the method in accurately assessing the capacity for renewable energy integration.

To meet the new demands of high-quality development of distribution networks and enhance their capacity to accommodate large-scale distributed generation and electric vehicle (EV) loads, this paper proposes a robust joint planning method for soft open points (SOP) and distributed energy storage systems (DESS) in AC/DC hybrid distribution networks, with consideration of EV demand response. Firstly, to address source-load uncertainty, typical and extreme daily operation scenarios are extracted using K-means clustering, and a scenario probability uncertainty set is constructed with l₁-norm and infinity-norm constraints to adjust the model’s conservativeness. And then, the response behaviors of EV users to real-time price are characterized by a demand price elasticity coefficient. A two-stage robust optimization model is formulated to minimize the annual total cost, and the second-order cone relaxation and McCormick envelopes are used to convexify the model. Scenario probability variables are expanded in binary form to enable worst-case scenario search within the uncertainty set. Candidate SOP locations are extended based on network partitioning. The model is solved efficiently by applying duality theory and the inexact column-and-constraint generation (i-C&CG) algorithm. Finally, the effectiveness of the proposed model in supporting voltage, ensuring renewable energy accommodation, and reducing losses is verified in a 69-bus system.

To address the issue that current regulation potential evaluation methods for flexibility resources fail to consider the optimal operation status of distribution transformer zones, resulting in adverse impacts on the zones' autonomous operation when supporting the upper-level grid's power regulation, this paper proposes a modeling method for regulation potential range that incorporates autonomous operation optimization of distribution transformer zones and source-load uncertainties. Firstly, the scattered flexibility resources within the distribution transformer zone are aggregated, and a regulation model is established. Then, the interaction power baseline between distribution transformer zones and the upper-level grid is optimized to achieve the optimal operation of the distribution zone. Finally, accounting for source-load uncertainties and based on baseline interactive power, a transformer zone regulation potential range model is developed to enhance the upward/downward regulation margins, with solution via the jellyfish search algorithm. Simulation results demonstrate that the proposed method can accurately evaluate the regulation potential range of the transformer zones, providing reliable references for the upper-level grid to allocate regulation tasks effectively, and achieve a balance between optimal operation and regulation capability of the transformer zones.

With the continuous increase in the penetration rate of residential photovoltaic (PV), voltage regulation in low-voltage (LV) distribution networks has attracted growing attention. To address the voltage regulation requirements across different time scales, a coordinated control method for LV distribution networks with high-penetration residential PV is proposed. First, aiming at minimizing network losses and voltage deviation, a day-ahead and intra-day centralized optimization control model is established to determine the operational setpoints for voltage regulation resources. Then, the optimized results for residential PV systems are applied in the real-time control stage, and an adaptive control strategy for residential PV is improved to dynamically adjust the output power based on node voltage measurements. Finally, simulation analysis on a 21-node LV distribution network demonstrates that the proposed method significantly improves network losses and voltage profiles, effectively suppresses voltage violations, and ensures the secure and stable operation of the distribution network.

Under the vision of high-proportion renewable energy development, in order to effectively shorten the energy storage payback period, enhance renewable energy accommodation, and reduce distribution network carbon emissions, this paper proposes a multi-objective bilevel planning model for energy storage systems (ESSs) in distribution networks that considers refined charging/discharging strategies and carbon benefits. Firstly, typical photovoltaic scenarios are generated using an improved Wasserstein generative adversarial network with gradient penalty (WGAN-GP) and the K-medoids clustering algorithm. Secondly, a refined charging/discharging model for ESS is established, and a carbon benefit model is constructed based on both the carbon emission reduction enabled by ESS and its lifecycle carbon emissions. And then, a bilevel ESS planning and operation model for distribution network is constructed considering the refined charging/discharging strategies and carbon benefits. The upper-level model aims to minimize the total daily cost for optimal energy storage configuration, while the lower-level model pursues the minimization of operational costs and voltage deviation, as well as the maximization of carbon benefits from energy storage, to achieve optimized distribution network operation. Subsequently, the bilevel model is transformed into a single-layer multi-objective model by modeling the inter-layer coupling variables. The multi-objective model is then solved using the normalized normal constraint (NNC) method, and the optimal compromise solution is selected via the entropy-weighted TOPSIS method. Finally, the effectiveness of the proposed model is verified through numerical case studies based on the IEEE 33-node system.

In recent years, with the continuous implementation of the "county-wide rooftop distributed photovoltaic (DPV) development program" promoted by the National Energy Administration, the integration scale of PV in China has steadily expanded, enhancing the utilization efficiency of renewable energy. However, the large-scale integration of PV users, particularly at the ends of low-voltage distribution networks. has led to issues such as localized voltage violations and abnormal power losses, thereby compromising power system's operational security and economic efficiency. To address the aforementioned issues, this paper proposes a bi-level optimal allocation method for distributed generation siting and sizing based on second-order cone programming relaxation. Firstly, to balance system's operational security and economic efficiency, a bi-level optimization model is established with the dual objectives of minimizing the system's daily comprehensive operating cost and maximizing PV utilization, which takes account such factors as the aging, maintenance costs and operational safety of distributed energy storage systems. Then, the second-order cone programming relaxation technique is employed to transform the power flow model, converting the original non-convex model into a convex one to reduce computational complexity, and the multi-objective particle swarm optimization algorithm is used to solve the model. Finally, using the improved IEEE 33-bus system and an actual distribution network in Hebei for case study, a comparison with the traditional planning approaches is conducted to verify the superiority of the proposed method. The results indicate that the proposed method can enhance power quality, reduce network losses, and increase the DPV hosting capacity while ensuring economic efficiency of distribution networks.

Building a new-type power system dominated by new energy sources is a critical pathway towards achieving China's "Dual Carbon" strategic goals. Since the inception of the "14th Five-Year Plan" period, China's power system has undergone profound transformations in its generation mix, grid configuration, and technological foundation, making the enhancement of power supply-demand synergy capability under these new circumstances an urgent imperative. Focusing on the new-type power system, this study firstly defines the fundamental concept and connotation of the supply-demand synergy and systematically identifies the critical scenarios that require enhanced synergy. Then, the study conducts an in-depth analysis of all the factors influencing supply-demand synergy and their underlying mechanisms from perspective such as policy, market, technology, etc. Finally, on the basis of an in-depth deconstruction from multiple perspectives such as space, time, resource, and form, this study innovatively constructs a theoretical framework model for the synergy of supply-demand across all factor to provide a solid theoretical foundation for the innovative upgrading of the new-type power system, and to contribute to the effective planning of its high-quality development blueprintt.

Demand response is a key means of regulating the flexibility of the distribution network and optimizing the dispatch of demand-side resources. To fully utilize the adjustable of demand-side resources, a demand-side resource cluster optimization dispatch method based on load baseline reconstruction in the distribution network is proposed. A fuzzy set is used to construct an model for the output of new energy, and a load baseline model considering the prediction error of new energy is reconstructed. The adjustable power domain of the adjustable load resource cluster is quantified based the idea of time-period decoupling dimensionality reduction, and an intraday optimization model for the demand-side resource cluster participating in the baseline-type response is proposed. Through the example, it can be seen that the proposed method can not only improve the new energy consumption capacity, but also improve the economic benefits of the power grid.

Aiming at the power quality problems brought by the access of large-scale distributed power (DG) to the distribution network, this paper proposes distribution network power flow optimization control method based on hybrid power electronic transformer (HPET), which aims to improve the accommodation capacity of the distribution network for DG. The steady-state of HPET are described by establishing a voltage source model and a power injection model, and the HPET sequential linear programming (SLP) method is proposed by considering the nonlinear voltage capability and the constraints of power electronic devices. Through example simulation verification, the proposed method provides a low-cost and efficient voltage/reactive power control solution for the distribution network high-penetration DG access by coordinating the operation of multiple HPET, which improves the utilization of DG in the distribution network.

Pumped storage demonstrates significant peak-shaving and valley-filling effects. While combined operations with nuclear power can stabilize the utilization hours of nuclear plants, they may also compress the generation space for renewable energy. To address this conflict, this paper explores the planning of "nuclear-pumped storage" combined operation from system benefits. First, the reasonable installation ratio is studied from two dimensions: output equivalence and economic benefits. On this basis, using an 8 760-hour production simulation, it obtains key data profiles such as system generation costs, renewable energy accommodation rates, and coal power generation volumes for different combined operation scales. Finally, based on the objective weights assigned to key indicators by the entropy weight method, it integrates multiple data into a comprehensive evaluation to accurately recommend the optimal combined operation scale. The results indicate that the established analysis method can meet the requirements for combined operation planning across multiple indicator dimensions; conducting combined operation at a reasonable scale can achieve optimal overall benefits under the constraints of key indicators.

The uncertainty of the source and load increases the difficulty of real-time situation awareness of the distribution network. To this end, an affine power flow calculation for the low-voltage distribution network considering the uncertainty of distributed photovoltaic output is proposed. The power flow model of the low-voltage distribution network is established considering uncertainty of photovoltaic output and load. The influence of the neutral point voltage on the power flow calculation results is considered, and the affine expression of the distribution network state is established, thereby reducing the expansion effect of the interval operation and obtaining a more compact interval of the state variables. Based on the standard distribution network test system, the proposed method is. The simulation results show that the proposed method can not only improve the calculation accuracy but also give more valuable evaluation results in the analysis of power quality.

Accurately identifying the key feature factors that influence short-term power load forecasting is an effective means for enhancing forecast accuracy. To address the issue in multidimensional datasets where non-critical features can easily lead to poor fitting capability of prediction models, consequently reducing model accuracy, this paper proposes a short-term power load forecasting method that combines the eXtreme Gradient Boosting (XGBoost) decision tree with an improved Informer model. Firstly, to evaluate the importance of feature factors from multi-dimensional historical load data, the coverage metric of XGBoost decision tree is adopted as an indicator to assess feature importance, thereby enabling accurate screening of the feature factors participating in model training. Subsequently, an improved Informer short-term load forecasting model is constructed. By optimizing the positional encoding design, the selected key features are combined with positional markers of different time scales to form input vectors for the encoder. Finally, ablation experiments are designed to conduct a comparative analysis of model convergence speed and prediction accuracy across different time scales. Experimental results indicate that, compared to other models, the XGB-Informer model demonstrates significant advantages in both prediction accuracy and convergence speed, verifying the effectiveness and superiority of the proposed method.

To address issues such as slow filling speed and high cost in existing SF₆/N₂ gas-insulated switchgear (GIS) filling devices, a direct pipeline filling method based on heated-pipeline gas flow control technology is proposed. This approach eliminates the need for built-in mass flow controllers (MFCs). Through theoretical analysis and ANSYS simulations, experimental parameters for the gas flow–temperature rise test platform were determined. A 3-meter heated pipeline was constructed, and gas flow prediction models were established for initial temperature differences of 20 °C and 30 °C. Precise gas flow regulation was achieved using a cascade fuzzy PID controller. Experimental results show that the cascade fuzzy PID controller outperforms the cascade PID controller across all performance metrics. When the initial temperature difference for SF₆ is 20 °C or 30 °C and 30 °C for N₂, the deviation in the SF₆ proportion in the mixed gas remains below 1.0%. A prototype filling device was designed and tested, demonstrating that when the filling pressure exceeds 0.2 MPa, the SF₆ proportion deviation complies with the standard DL/T 2243—2021, meeting on-site filling requirements.

To solve the problem of fair benefit distribution among provinces in regional electricity market joint clearing, this paper proposes a regional electricity market gain allocation mechanism that incorporates energy-reserve joint clearing. First, a regional electricity market joint clearing model is constructed with the objective of minimizing the total operational cost of the entire network, which coordinates the optimization of energy production, reserve capacity, and inter-provincial transmission costs. Second, based on Shapley value theory, a gain allocation method is designed to quantify the marginal contribution of each province in joint operations, with further refinement to the levels of power generators and consumers to ensure fairness in benefit distribution. Finally, simulation verification is conducted on a three-region interconnected system. The results show that compared to traditional independent clearing models, the proposed joint clearing mechanism can reduce the total system cost by approximately 3% and decrease load shedding by about 80%. The research results provide theoretical support for the collaborative operation and benefit allocation of regional electricity markets.

With the continuous increase of the renewable energy penetration and the increment of the adjustable units capacity, it has been an important trend for them to participate the electricity market auction transaction. In order to deeply analyze the competitive game behavior between renewable energy generators (REG) and conventional power generators (CPG) in the regional electricity market, multiple leaders-one follower (MLOF) Nash-Stackelberg (NS) game model is proposed. The upper layer constructs a bidding strategy model for the joint participation of renewable and conventional power generators in the electricity energy and ancillary service market based on the green certificate trading mechanism. The lower layer is a joint market clearing model of electricity-reserve-, which determines the market clearing results and price signals. On this basis, an efficient distributed algorithm is carried out to solve the complex game model. So as to improve the computational and protect the data privacy of all parties. Simulation analysis shows that the proposed model can effectively improve the market returns of renewable energy generators, and enhance their game ability in multi-subject.