In order to give full play to the role of energy storage "peak valley arbitrage" in the grid-connected photovoltaic storage microgrid to reduce the operation cost of microgrid, and to reduce the effect of photovoltaic and load power fluctuations on the arbitrage effect, a dual-time scales coordinated operation strategy of photovoltaic storage microgrid source-grid-load-storage is proposed based on time-of-use electric price and with energy storage as the main scheduling control object. In the day-ahead optimal scheduling model, the energy storage system loss costs caused by the frequency and depth of energy storage charging and discharging are considered. The objective function is established with the tareget to minimiz the operation cost of microgrid, and optimize charging and discharging period and amount of energy storage under different photovoltaic outputs and load conditions. In the intraday real-time control model, various photovoltaic outputs and load conditions are divided into different operation scenarios according to the time-of-use electric price mechanism, the energy storage day-ahead scheduling plan and charging and discharging constraints. The calculation formulas of energy storage charging and discharging power in various operation scenarios are given to control the energy storage real-time output, adjust day-ahead output plan and reduce effects of photovoltaic and load power fluctuations on the arbitrage. The proposed control strategy is verified by simulation examples. The results show that the proposed strategy plays a good role of "peak-valley arbitrage" of energy storage, and the operation economy of microgrid is improved. At the same time, the influences of photovoltaic and load power fluctuation are reduced.

In view of the problem that power supply enterprises have a large number of power supply tasks and occupy a large amount of manpower and material costs in China, an operation model of power-guaranteed microgrid based on multi-party game among photovoltaic investors, energy storage investors and microgrid operators is proposed, considering the technical economy of distributed photovoltaic, user-side energy storage and microgrid in the field of power supply protection. Firstly, for the current emergency power supply vehicles or power generation vehicles, there are problems such as low asset utilization, environmental protection, and noise pollution, the idea of power supply protection for photovoltaic-storage microgrid instead of emergency power supply vehicles or power generation vehicles is discussed in detail. Then, based on the benefits and subsidies of microgrid participating in power supply protection, considering the distribution of photovoltaic power generation and its selling price among the photovoltaic -storage-microgrid, the investment cost and benefit composition of the three investment entities are analyzed, and the cooperative game method is used to optimize and discuss the interest equilibrium points of the three investment entities, so that the benefits of the three entities can be optimized at the same time. Finally, the economic benefits of photovoltaic investors, energy storage investors, and microgrid operators of the operation model proposed in this paper are analyzed through examples.

With the rapid development of plug-in electric vehicles （PEVs） and hydrogen fuel vehicles （HFVs）, introducing hybrid energy supply stations （HESSs） into microgrids is an important way to realize a green and low-carbon energy system. A HESS is a new type of energy infrastructure that can not only charge PEVs, but also refill HFVs. Currently, there is a lack of in-depth research on energy scheduling for microgrids with HESSs. Thus, this paper focuses on the distributed energy scheduling problem of multi-microgrids and virtual power plants with HESSs under the scenario of massive heterogeneous device integration. Firstly, a refined model of the HESS is formulated, and a cloud-edge collaboration based two-layer distributed energy scheduling framework is proposed to achieve optimal scheduling of heterogeneous devices such as electric vehicles, energy storage, and renewable energy sources. The upper and lower layers of the framework respectively consider the optimal scheduling of energy for a virtual power plant and multi-microgrids. Then, an analytical target cascading （ATC）-based distributed optimization algorithm is employed to decouple the upper- and lower-layer problems, thereby achieving the decentralized autonomy of the virtual power plant and microgrids. Finally, the effectiveness of the proposed cloud-edge collaborative method in achieving distributed optimization of the virtual power plant and microgrids is verified through case studies.

In the current digital age, Internet data centers （Internet data center, IDCs） have become important components of power grid as large power consumers. Firstly, IDC waste heat recovery technology is adopted to establish a IDC equivalent thermal parameter model, which is integrated into the micro-energy network to achieve multi-energy complementarity of cooling, heating, and power. Secondly, the complex relationship between the required cooling capacity and waste heat during the operation of IDCs is analyzed in depth, and lithium bromide absorption refrigeration machines and electric refrigeration machines are used to meet the cooling load requirements of IDCs. In order to more effectively cope with the changing characteristics of batch processing load in the time dimension, a highly flexible data load allocation strategy has been developed, and a two-layer programming model has been established. The typical day selection method for scenario reduction is used to address source load uncertainty, and the Tent mapping grey wolf optimization algorithm （TMGWO） based on Tent mapping and CPLEX are introduced for joint solution. Finally, through simulation analysis of a micro-energy network with a IDC, numerical results show that the proposed method can effectively improve the system's economy and environmental friendliness.

The interests of microgrid users driven by incentive-based demand response are incorporated into the classic environmental economic dispatch model. An improved electricity satisfaction model for users with diverse mixed loads is proposed, serving as a metric to evaluate user-side benefits. A multi-objective dispatch model is established with the goals of minimizing system operating costs, environmental costs, and maximizing user electricity satisfaction. In order to improve the performance of the grey wolf optimization algorithm in solving multi-objective problems, the algorithm is improved by using strategies such as searching population variation factor and nonlinear cosine transform of convergence factor. Then the improved multi-objective grey wolf algorithm is applied to solve the multi-objective dispatch model. At the same time, the optimal compromise solution is selected from the Pareto solution set based on fuzzy set theory to analyze and compare the dispatch results of various load responses and different types of users. Finally, the simulation results demonstrate that a well-designed demand response mechanism can enhance both economic and environmental benefits while ensuring user satisfaction with electricity consumption. This approach achieves a win-win situation for the power system, the environment, and users benefits, offering valuable theoretical support for the development of a green and sustainable power management system.

Under the "dual carbon goal", with the deepening reform of China's energy market and the gradual formation of the distributed multi-energy market, the integrated power microgrid and its community appear on the user side. The microgrid cluster formed by the interconnection of adjacent microgrids is a multi-energy comprehensive energy system, which is jointly affected by many factors such as multi-energy interaction and coupling, complex game decision relations and uncertainty of energy sources and loads. Therefore, the energy mechanism of this kind of comprehensive energy system is complex, and the management scheme formulation is difficult. On the basis, a multi-agent comprehensive energy microgrid group alliance energy management strategy is proposed based on cooperative game and Gaussian mixture clustering. Through simulation example analysis, it can be seen that the proposed method significantly reduces the operating costs of various microgrids, up to 24 % for each microgrid, thus improving the operating efficiency of the entire microgrid system and maximizing the benefits of the alliance. The influence of higher source load prediction errors on the operating cost of each agent and alliance are less than 3 %.

To adapt to the new trend of a large number of market entities participating in the investment and construction of distribution networks, a distribution network source-network-load-storage collaborative planning method is proposed considering the mixed game relationship of multiple agents. Firstly, the interactive behaviors and respective interests of four main agents are analyzed in the planning process, which are distribution network operators, distributed power generation operators, energy storage operators, and power users. And the mixed game relationship of master-slave cooperation is established. Then, a hybrid game planning model is constructed with the distribution network operator as the leader with an embedded source-load-storage side follower alliance cooperative game. A solution strategy combining target cascade analysis method and adaptive coefficient alternating direction multiplier method is proposed to achieve decoupling and distributed iterative solution of the two-layer model. Finally, the case analysis shows that the planning method can effectively balance the interests and demands of all parties. The investment and utilization of distributed new energy are promoted. And the economic benefits of the planning scheme is enhanced.

As renewable energy penetration continues to increase, the strong randomness of wind and solar renewable energy output leads to instability of grid frequency and deterioration of control performance. To address this, a multi-agent reinforcement learning approach is explored from the perspective of automatic generation control, that is the high dimensional cooperative soft actor-critic algorithm. The proposed algorithm encourages agents to engage in random exploration within a maximum entropy framework to address the inability of Q-learning and its derivative algorithm to rapidly update the Q-table in response to environmental changes. It also utilizes a Gaussian distribution strategy to generate continuous action values, enabling the algorithm to find collaborative optimal solutions in high dimensional continuous state spaces, thus solving the traditional reinforcement learning curse of dimensionality of high dimensionality "state-action". Frequency instability and declining control performance caused by the strong randomness of renewable energy outputs are effectively addressed. Through simulation experiments on an improved IEEE standard two-area load frequency control model and the Central China three-area load frequency control model, the effectiveness of the algorithm is validated. And the proposed algorithm has superior control performance and frequency stability compared to other algorithms.

With the continuous evolution of new generation information and communication technologies such as 5G, cloud computing, and artificial intelligence, the world is rapidly entering the fast lane of the digital economy. A dual-layer optimization scheduling method for data centers based on multi-agent proximal strategy network is proposed to address the uncertainty of renewable energy and workload prediction in data centers. Firstly, a dual-layer spatiotemporal optimization scheduling framework for data centers is established, which provides detailed modeling of data center workloads, IT equipment, and air conditioning equipment; On this basis, a dual-layer optimization scheduling model for data centers is proposed. The upper layer schedules the time dimension with the goal of minimizing the total operating cost of IDC operators, while the lower layer schedules the space dimension with the goal of minimizing the operating cost of each IDC. Then, the principle of multi-agent proximal strategy network algorithm is introduced, and the state space, action space, and reward function of the dual-layer optimization scheduling model for data centers are designed. Finally, offline training and online scheduling decisions are conducted for the examples. Simulation results show that the proposed model and method can effectively reduce system costs and energy consumption, achieve optimal workload allocation, and have good economy and robustness.

To address the challenges posed by the increasing penetration of wind power on grid frequency regulation, this paper establishes a life-cycle cost-benefit calculation model for wind farms equipped with energy storage systems participating in primary frequency regulation services. The model considers energy storage costs, real-time electricity revenue, and primary frequency regulation revenue, with energy storage capacity and power as decision variables. To tackle the difficulty of accurately evaluating primary frequency regulation actions, a frequency regulation model for the combined wind-storage system is developed using an open-loop processing method based on the closed-loop dynamic process of the power system frequency. This model incorporates wind turbine control and energy storage system state-of-charge (SOC) constraints, and proposes an open-loop calculation method for primary frequency regulation actions using annual frequency curves as inputs. By combining the wind-storage system frequency regulation simulation model with annual operational frequency and power data of the wind farm, the change rule of the annual energy storage costs and primary frequency regulation revenue are obtained. The results show that for a wind farm with an installed capacity of 30 600 kW, the optimal primary frequency regulation revenue is achieved when the energy storage power is 4 553.6 kW and the energy storage capacity is 9 822.6 kWh.

Peer-to-peer （P2P） trading provides an important way for promoting the consumption of renewable energy by prosumers and the reform of power markets. The information transmission in P2P transactions relies on the cyber-physical system （CPS）, however, issues such as data bias, transmission delays, and prosumers' own information protection make it difficult for trading participants to accurately obtain the state information of other prosumers, thereby creating an incomplete information environment. In addition, the bounded rationality of prosumers intensifies the uncertainty of trading behavior. To address these issues, a stochastic game-based optimization method for prosumer P2P trading strategy considering incomplete information is proposed. Firstly, Harsanyi transformation is used to convert incomplete information into complete but imperfect information. The resulting transformation is then refined using prospect theory to better reflect prosumers' actual psychological preferences. Secondly, a stochastic game decision-making model for prosumer P2P trading is constructed. By incorporating the Markov decision process （MDP） framework into the stochastic game, behavioral uncertainty in trading is effectively reduced, enhancing the stability and effectiveness of trading strategies. Then，to mitigate the “curse of dimensionality” caused by the exponential growth of state spaces in the stochastic game, an adaptive state-tree pruning technique is proposed to significantly reduce computational complexity. Simulation results show that the proposed method effectively alleviates the impacts of incomplete information, reduces behavioral uncertainty, optimizes trading strategies, and improves overall economic performance.

With the restart of China′s certified voluntary emission reduction （ CCER ） market, the benefits of CCER are mainly realized through carbon market transactions, and the CCER carbon emission reduction strategy based on carbon market is closely related to the energy scheduling strategy based on electric energy market. In order to give full play to the energy-carbon flexibility aggregation effect of integrated energy system （ IES ）, implement low-carbon and power demand scheduling, and promote low-carbon and efficient operation of IES and low-carbon transformation of power system, a market-oriented operation strategy of IES electricity carbon considering low-carbon benefits is proposed. A bi-layer optimization game model of IES participating in the joint clearing of electricity carbon market is constructed. The upper layer is the bidding decision model of IES electricity carbon market. The carbon quota allocation model is constructed and the CCER voucher-carbon quota offset mechanism is quantified. With the minimum comprehensive cost of IES electricity carbon operation as the goal, the heterogeneous energy capacity plan and the electricity carbon market bidding strategy are formulated. The lower layer is the electricity carbon market clearing model, which aims to maximize the social welfare of the market and meet the market electricity carbon demand clearing. The bi-layer model is solved iteratively, and finally the optimal capacity decision of IES participating in the operation of the electricity carbon market is obtained. The results of the case study verify that the proposed strategy can reduce the carbon emissions of the system and improve the operating efficiency of the IES.

In order to facilitate the operation regulation of distribution network and reduce the line loss rate of distribution network system to improve its transmission efficiency, a cluster division method of high-proportion distributed photovoltaic（PV）access to distribution network considering the matching degree of multi-section source-load features is proposed. Firstly, the distributed PV output curve connected to the distribution network and the load characteristic curve of each region are comprehensively discussed, and the matching degree index of multi-section source-load features is designed to maximize the utilization of PV in the same cluster. Moreover, the modular degree index considering the cluster structure and electrical distance, as well as the capacity matching index and reactive power compensation index considering the cluster function are comprehensively designed. The comprehensive index system is constructed to realize the division of distribution network cluster. Then, when the optimization algorithm is used to divide the distribution network cluster, the constraints based on the network adjacency matrix are added to avoid the existence of isolated nodes. Finally, taking the IEEE 33-node system as an example, the cluster division and the PV output optimization after the cluster division are carried out. The results show that the proposed cluster division considering the multi-section source-load features matching has better source-load features matching, stronger electrical coupling, and better active and reactive power balance. Compared with the case without considering the cluster division, the photovoltaic power output optimization after the distribution network cluster division can reduce the line loss rate of the distribution network system, thus improving the transmission efficiency of the distribution network.

Despite extensive research on non-intrusive load monitoring(NILM), existing models face challenges in accurately predicting multiple operational states of appliances, leading to significant decreases in prediction accuracy. To address this issue, this paper proposes non-intrusive load monitoring model based on multi-algorithm fusion. Firstly, the REDD low-frequency dataset is preprocessed using time-based interpolation and oversampling. Secondly, the model employs graph convolutional networks(GCN) and convolutional neural networks(CNN) to extract power features, which are then fed into a self-attention mechanism and long short-term memory(LSTM) networks. This effectively captures the key features of the input signals, thereby improving the prediction accuracy for appliances with multiple operational states. Finally, simulation verification is conducted using the preprocessed REDD low-frequency dataset. The experimental results indicate that the proposed model outperforms comparative models in terms of MAE, SAE, and R^{\mathrm{2}}, demonstrating its effectiveness in load disaggregation.

CO₂ emissions from industrial parks account for about 31% of the total national CO₂ emissions, and the low-carbon development of parks plays an important role in mitigating climate change. In this paper, the key information of single and multiple CO₂ emission sources in industrial parks is inverted and studied by taking an industrial zone in Hubei Province as an example. A forward model of CO₂ diffusion in the industrial park is established based on the AERMOD system to obtain the data set required for inversion. The BP neural network optimized by particle swarm optimization （PSO）, whale optimization algorithm （WOA） and pelican optimization algorithm （POA） is used to calculate the inversion of CO₂ emission source location and emission intensity in the industrial park. The results show that the POA-BP inversion model has an R2 of 0.965 for single source coordinates and an R² of 0.938 for emission intensity, and an R² of 0.97 for multiple source coordinates and an R² of 0.988 for emission intensity, which has a higher inversion accuracy and stability than other models, and it can achieve a more accurate location of CO₂ sources in the industrial park and provide a better solution for the industrial park. It can realize more accurate positioning and provide decision support for industrial parks to cope with climate change and promote green transformation.