With the large-scale integration of distributed energy resources, the operational characteristics of the traditional distribution networks have undergone significant changes, leading to such problems as dispersed loads, poor real-time observability, and incomplete data, which severely impact the state monitoring and operational optimization of the distribution networks. To address above problems, we propose a distribution network state estimation method based on Bayesian-optimized convolutional neural networks (CNN) and long short-term memory (LSTM) networks with incomplete real-time measurement data. The method is divided into two phases: offline learning and online state estimation. In the offline learning phase, generative adversarial networks are used to generate the required samples for training the CNN-LSTM model, and the Bayesian optimization algorithm is employed to adjust the hyperparameters, thereby enhancing the accuracy of the algorithm. In the online state estimation phase, the state estimation is performed online with incomplete real-time data of the distribution network and the trained CNN-LSTM model. Finally, simulation analysis is conducted on the IEEE 33 and IEEE 123 networks, which confirms the effectiveness and accuracy of the proposed state estimation method.

To address the voltage limit violations and bidirectional power flow problems arising from high-penetration integration of distributed renewable energy, this paper proposes a two-layer active-reactive power cooperative optimization method to achieve cooperative optimal dispatch of active and reactive power in off-grid microgrids, ensuring the secure and stable operation of the system while enhancing operational economy. The lower-level model optimizes slow-regulating discrete devices based on mixed-integer second-order cone programming, while the upper-level model optimizes fast-regulating continuous devices using a multi-agent deep policy gradient algorithm. The two-layer model coordinates both active and reactive power flows of the microgrid, enabling real-time monitoring of the microgrid's status and online decision-making for the optimization of device regulation, without reliance on precise power flow models or complex communication systems. Finally, the feasibility and effectiveness of the two-layer optimization model are validated in the improved IEEE 33-bus microgrid system.

Traditional PV power interval forecasting relies on specific probabilistic distribution assumptions, which often result in inconsistencies between the assumed probability distributions and the actual heteroscedastic nature of PV power distributions, thus affecting the accuracy and confidence level of interval predictions. To address this issue, an ultra-short-term PV power interval forecasting method based on time-series decomposition and conformal quantile regression (CQR) is proposed. Firstly, the PV power series is modeled as the sum of three additive subseries: trend components, periodic components, and autoregressive components, based on the NeuralProphet time-series decomposition framework. Then, piecewise linear models, Fourier series decomposition models, and AR-Net models are respectively employed to fit the three subseries, with the Fourier series decomposition model enhancing the fitting capability for daily and seasonal periodicities of PV power. Finally, by calculating the prediction uncertainty of the CQR model, the quantile interval of the prediction results are determined based on conformal scores, enabling dynamic adjustment of the prediction interval width without the need for preset probability distributions. Case studies demonstrate that the proposed method outperforms the advanced Transformer-based algorithms like TimesNet and Informer in deterministic PV power forecasting tasks, and with the introduction of the daily and seasonal periodic components, the prediction error is further reduced by 11.65%. In interval forecasting tasks, the proposed method surpasses the traditional quantile regression algorithms in terms of prediction interval coverage rate, normalized interval width, and coverage width-based criterion.

With the popularization of smart meters and the gradual improvement of grid informatization and digitization, non-intrusive load monitoring (NILM) on the demand side of residential customers' energy usage is becoming one of the key technologies for power supply companies to boost energy efficiency. Regarding the problems of the current non-intrusive load recognition algorithms, such as feature redundancy, high computational overhead, and low recognition performance, the paper proposes a non-intrusive load recognition method integrating SENet attention mechanism and GA-CNN. Firstly, the SENet attention mechanism is embedded in a convolutional neural network (CNN) to improve the characterizaion of key features and reduce feature redundancy. Secondly, the U-I trajectory map of the residential load is extracted and weighted pixelated to obtain the WVI (Weighted pixelated VI) feature matrix through computation, which is applied as the feature coefficient to train the SENet-CNN model. Finally, by virtue of the genetic algorithm, the SENet-CNN model is trained and the hyperparameters of the CNN-SENet model are optimized to improve the model load recognition performance and computational efficiency. The experimental results show that the proposed method can reduce the computational overhead of non-intrusive load identification, accurately identify the residential load categories, and significantly improve the efficiency of non-intrusive load identification.

The report of the 20^th National Congress of the Communist Party of China points out it is necessary to accelerate the green transformation of the development mode that carrying out green power trading is an important way to promote the green and low-carbon transformation of energy. In recent years, China's green power market construction has achieved remarkable results. With the green power market transaction scale continues expansion, the implementation of green power consumption certification has become an urgent need for policy landing and market players. The construction of green power consumption certification system has become a key issue facing the development of the green power market. In this paper, on the basis of in-depth analysis of situation of Chinese green power consumption certification, a systematic summary of the relevant mechanisms of mature green power market in Europe and the United States is conducted. And a systematic summary has been conducted on the green power consumption certification mechanism of typical international organizations, from the consumption accounting account system, consumption accounting methods, consumption accounting results, consumption accounting platforms and other aspects. Based on the current situation of green power consumption certification in China, inspiration for China and suggestions for the next step of China's initiatives to carry out green power consumption accounting certification is proposed.

Energy is the foundation of national economy and the main field for carbon reduction, and its future development of energy supply and demand has a significant impact on constructing the new energy systems and achieving the carbon peak and carbon neutralization targets. Considering multiple factors such as economic and social development status, energy consumption characteristics of each province, an indicator system for carbon quota allocation is constructed, encompassing aggregate and relative metrics, and balancing equity with efficiency, and a provincial-level energy supply-demand forcasting method under carbon quota constraints is proposed to achieve comprehensive balance analysis and prediction of primary and secondary energy of different varieties and industries covering the entire process. Based on China's carbon peak target in 2030, an empirical analysis is carried out with Shandong Province as an example to verify the effectiveness of the proposed method, which provides a reference for different provinces in carbon emissions control and energy transformation.

To investigate the investment decision optimization issue under the condition of uncertain cost of multiple types of low-carbon technologies faced by power generation enterprises to actively achieve the goal of "dual carbon", this paper takes the medium- and long-term time-scale decision-making optimization of independent power generation enterprises as the research perspective. Considering the cost uncertainty of various technologies within the investment period, a two-stage robust model is constructed based on the expected output and installed capacity. In the first stage, the objective is to maximize the net profit of the enterprise's operations. In the second stage, the cost uncertainty of various technology investments and the expected output provided in the first stage are considered as constraints. A comprehensive consideration of the annual investment intensity of the enterprise and the total investment limit for the investment plan period is made to minimize the total investment cost. The second-stage model is transformed into an equivalent linearized model for subsequent solving of the problem in a robust equivalent form. The column-and-constraint generation algorithm is employed to solve the two-stage problem. The effectiveness of the proposed model is verified by comparing the evolution of the power generation company's power structure under different scenarios, which can provide valuable insights for power generation companies adopting low-carbon technologies.

Regarding the volatility and uncertainties in the operation of independent microgrid with photovoltaic energy storage, a control strategy based on sliding mode is proposed to enhance the robustness against various disturbances and improve the system dynamic performance. In order to effectively reduce the adverse effect of external solar irradiation, a non-singular fast terminal sliding mode control (NFTSMC) based on front DC/DC converter is proposed, in which incremental conductance method (InC) is implemented , to boost the maximum power point tracking (MPPT) performance by means of adjusting the photovoltaic output voltage tracking reference voltage. In order to alleviate the adverse effect of load change on the system performance, a control scheme based on sliding mode control of full bridge inverter is proposed to achieve low steady state error and rapid dynamic response. Furthermore, in order to maintain the voltage stability of the DC bus, the traditional dual-circuit PI control scheme is used to realize the bidirectional DC/DC converter to ensure the power balance of the system. Finally, simulation tests are carried out in Matlab/simulink environment to verify the effectiveness of the control strategy.

To fully capitalize on the medium- and long-term storage benefits of hydrogen energy, a two-stage weekly optimization scheduling strategy for electric-hydrogen coupled microgrids is proposed, utilizing a scenario-based stochastic optimization model. Firstly, the mathematical models for the electric-hydrogen coupled equipment within the microgrid are developed. Secondly, aiming to minimize the weekly operation cost, the cycles of electric and hydrogen energy storage are set as weekly and daily, respectively, and the first-stage weekly scheduling model of the microgrid based on weekly prediction data is established. Then, to account for the uncertainty in wind power generation, typical scenarios of prediction errors are introduced. A day-ahead scheduling model considering uncertainty is constructed, aiming to minimize the sum of the daily expected operational cost and the deviation penalty for the two-stage hydrogen storage tank's state of charge. The final weekly operation scheme is determined through rolling optimization. Finally, case studies demonstrate that the proposed strategy effectively reduces the microgrid's operating costs and enhances energy utilization.

With the large-scale grid connection of electric hydrogen production load and the frequent changes of power grid operation mode, the access mode and distribution location of electric hydrogen production equipment cluster have an increasingly significant impact on the stability of power grid. Therefore, this paper proposes a planning method of electric hydrogen production capacity and distribution location considering static voltage stability margin constraints. Firstly, based on the steady-state model of alkaline electrolytic cell, its operating characteristics are analyzed, and the grid-connected mode of transformerless direct-hanging electrolytic cell is constructed. The internal current of hydrogen production system is involved in the Newton-Raphson method iteration of continuous power flow calculation, and the static voltage stability margin of the current system is obtained from the PV curve. Based on the net present value evaluation index, the economic analysis model of the whole life cycle of the hydrogen production system is established. Based on the constraint of static voltage stability margin and the goal of maximizing the net present value of the system, a capacity planning model for the hydrogen production system of the electrolytic cell is proposed. The decision variable is the total capacity of the electrolytic cell. Based on the evaluation index of power transmission distribution factor, the layout planning model of the electrolytic cell at the key nodes of the system is constructed, and the weight of each node is calculated to obtain the capacity configuration result of the electrolytic cell hydrogen production system. Finally, the IEEE 39 node system is used for example analysis. The results show that the proposed method ensures the grid-connected stability of large-scale electric hydrogen production load while taking into account the economy, effectively consumes the redundant configuration of the system and balances the power flow distribution.

Aiming at the problem that the current residual effective life prediction (RUL) technique for proton exchange membrane fuel cells (PEMFCs) has poor prediction effect in the medium and long term, a residual life prediction method based on the Improved Gray Wolf Optimization algorithm (IGWO) and Echo State Network (ESN) is proposed, in which the voltage of the electric stack is firstly selected as a health indicator, and the PEMFC dataset is processed by using convolutional smoothing filtering method to carry out data Smoothing and normalization are used to effectively reduce the interference of outliers on the subsequent model training. Then the reserve pool parameters of the ESN are optimized using the local and global optimization search capability of IGWO, and the IGWO-ESN network model is constructed, and the processed dataset is used for the training of the remaining life prediction model of the PEMFC, and finally it is compared with the traditional ESN for verification. The results show that the improved ESN model predicts the root mean square error and average absolute percentage error of 0.0342 and 0.9315%, respectively, and the prediction accuracy is significantly improved compared with the ordinary ESN model, and the prediction accuracy of the medium- and long-term RUL is also higher.

The electricity price directly reflects the attributes of electricity commodities. However, influenced by various factors such as market supply and demand, power generation company bids, and line transmission capacity, the electricity price signals often exhibit diverse forms of anomalies. Recognizing and tracing these abnormal electricity price signals are crucial daily tasks for power trading centers at all levels. However, the existing tracing methods generally rely on manual experiences to analyze the causes of abnormal electricity prices, which is inefficient and difficult to ensure objective and scientific traceability of the causes of abnormal electricity prices. This paper proposes an abnormal electricity price cause tracing method based on comparative analysis of key features. Firstly, based on the historical electricity price data features, the abnormal spike amplitude and abnormal average value of electricity prices are categorized. And then, using the principal component analysis method, the collections of key features for each type of abnormal electricity price signals are established. Finally, using the alternative algorithms, the influencing degree of each element within the collection of key features on the electricity price are calculated respectively, and ranked based on their importance, thereby achieving filtering and tracing of the causes for abnormal electricity prices. The effectiveness of the proposed methodin is verified through case studies based on numerous actual power market data. The average accuracy rate of the proposed method for tracing the causes of average electricity price anomalies and spike anomalies reaches more than 85%, which can effectively reduce the labor costs in the process of tracing the causes of abnormal electricity prices.

To improve renewable energy consumption under the background of dual carbon, and solve problem of multi-source coordinated scheduling with DC transmission, a multi-source coordinated scheduling of receiving power system considering voltage stability based on information gap decision theory (IGDT) is proposed. Firstly, the voltage stability indicator is constructed, which indicates distance between the system voltage state and the voltage stability limit state. Then, aim at minimizing the system operation cost and wind curtailment penalty, a multi-source power receiving ability promotion model considering thermal/ cascade hydro/wind power generation, energy storage and DC transmission coordination is proposed. The models of multi-sources in the system are established, and the AC power flow and hydropower conversion constraints are linearized by Taylor series expansion and triangle approximation. On this basis, voltage stability constraints are generated through voltage stability margin threshold, and voltage stability constraints are added to multi-source coordinated optimization problem for the receiving power system through each iteration to achieve the improvement of voltage stability indicator. To consider uncertainty of wind power output and load demand, a multi-source coordinated scheduling of receiving power system considering voltage stability based on IGDT is proposed. Finally, the effectiveness of the proposed model is verified by case studies.

In order to improve the observability of distribution network topology, this paper constructs a universal topology identification framework. This framework includes three topology identification methods to handle different situations. The first method utilizes the sparsity feature of the node admittance matrix to propose an improved three-stage heuristic program for single topology recognition; The second method, based on the improvement of a single topology identification method, classifies the records in the measurement dataset with a given number of topologies into a fixed number of topology categories, and then determines the node admittance matrix for each topology category. A multi topology identification model is constructed to handle the measurement dataset with a given number of topologies; The third method designs a two-stage program for measurement datasets containing unknown topological quantities, which determines the number of topologies, record classification, and node admittance matrix for a single topology. Finally, two numerical examples were used to verify the effectiveness of the proposed model in identifying distribution network topology under three different topologies.

With the large-scale access of new energy to the power grid, the frequency response performance of the power station is becoming increasingly critical, and its rapid response capability has become an important support for ensuring the stability of the power grid. In view of this challenge, an efficient test design scheme for the rapid frequency response technology of new power stations is proposed. Firstly, the inherent defects of the phase locked loop (PLL) and the characteristics of the frequency response index under the condition of grid connection are analyzed., a fixed trajectory control strategy is proposed to suppress the power fluctuation under the frequency disturbance condition. Then, the rapid frequency response test system architecture is constructed, which supports voltage feed and generation control (AGC) command feedback to realize dynamic response test. Finally, the developed engineering application system can adapt to a variety of test environments, and the verification scheme not only the requirements of relevant standards, but also improves accuracy and efficiency.

A fast testing scheme for secondary system equipment based on cloud-side optimization is proposed. The cloud-side optimization capability is utilized to propose a fast testing architecture secondary system equipment and to achieve the reduction of the dimension from the sub-module to the test package. The operations of switching the applied quantity, throwing and closing the soft and hard plates and starting and stopping the terminal are further considered, and a fast testing mathematical model aiming at the shortest time is proposed. The test sequence is optimized based on the improved whale optimization algorithm (SWOA). The effectiveness and accuracy of the proposed scheme are verified through simulation result.

In view of the poor identification of the effective data in online monitoring of high-voltage capacitor voltage transformer (CVT), this paper proposes a method for effective data identification based on the analysis of data correlation coefficients, with utilization of the data characteristics that multiple online monitoring data sources exhibit linear correlation. To address the prevalent problem in current high-voltage CVT fault diagnosis, such as limited information sources, poor accuracy, and significant interference in partial discharge devices leading to compromised fault signal detection and accuracy, a fault diagnosis method based on multi-dimensional information fusion is proposed. Firstly, factor analysis is employed to perform data-level information fusion on the diagnostic indicators of CVT by extracting the variance contribution values of common factors corresponding to each fault type as eigenvalues that reflect the differences among fault types. Subsequently, fuzzy theory is utilized for feature-level information fusion, and the variance contribution values of common factors are taken as input parameters for the membership function to identify the fault types of CVT, thus achieving the accurate diagnosis of high-voltage CVT faults. The validity of the proposed method is demonstrated through case studies, which can provide a theoretical reference and practical experience for CVT fault diagnosis.

For offshore wind power transmission systems based on diode rectifier-based DC transmission, the voltage and frequency support of the offshore AC grid based on the grid-forming control of offshore wind turbines has become a key issue. For the currently most studied Q/f droop-based grid-forming control (GFM) scheme and the newly emerged P/f droop-based GFM control scheme, a detailed analysis of the multi-machine power coupling characteristics and the stability aspects of the two schemes is carried out. The analysis of the control mechanism reveals that the Q/f droop-based GFM control scheme has coupling characteristics between active and reactive power control path, while the P/f droop-based GFM control scheme has a natural power decoupling characteristic. Detailed analysis based on the small signal model shows that the power control coupling characteristics of the Q/f droop-based GFM control scheme may lead to power oscillation problems between multiple machines. And the P/f droop-based GFM control scheme has a natural decoupling characteristic and has better stability robustness under different load conditions, making it more conducive to achieving the synchronous and stable operation of hundreds of wind turbines in offshore high-capacity wind farms.

With the large-scale integration of distributed generation into microgrids, the volatility of renewable energy generation and system random disturbances pose significant threats to the frequency stability and operational control of islanded microgrids. To address this, a secondary frequency control method based on deep reinforcement learning is proposed. The droop control characteristics of islanded microgrids are analyzed, and a secondary frequency controller structure based on deep Q-Networks is presented. The frequency deviation is used as the state input variable, and the design of the state space, action space, reward function, neural network, and hyperparameters in the deep Q-Networks algorithm is carried out. The reward function balances the goals of frequency recovery and power allocation among distributed energy resources , ensuring consistency in action selection among the intelligent agents. An offline learning process is used to train the deep reinforcement learning-based secondary frequency controller. A simulation model of the islanded microgrid is developed in Matlab/Simulink, and multiple disturbance scenarios are tested to validate the controller's performance. The results show that, compared to traditional PID control and Q-Learning-based controllers, the proposed method achieves more stable secondary frequency control and adapts to coordinate the power allocation of distributed generation units according to their capacities, ensuring the stable operation of the system.

The short circuit between turns of the rotor winding is a weak characteristic electrical fault of the turbine generator, which makes the fault detection difficult. This paper investigates the weakening effect of the circulating magnetic field inside the parallel branch of the stator winding of a turbine generator on the asymmetric magnetic field generated by the rotor winding turn-to-turn short-circuit, derives the expression of the excitation magnetic field under the fault condition, obtains the expression of the circulating current of the parallel branch of the stator winding and the expression of the magnetic field generated by the circulating current, and finds out the offsetting effect of the circulating field on the original asymmetric magnetic field. A 300MW steam turbine generator is used as an example to carry out finite element simulation, which proves the weakening effect of the parallel branch structure of the stator winding on the short-circuit fault characteristics of the rotor winding, and the research conclusion is helpful to analyze the external appearance and causes of the short-circuit fault characteristics between turns of the rotor winding, and also provides theoretical support for the feature extraction in the fault diagnosis stage.