Power load forecasting is influenced by many uncertain events, so accurately predicting load has always been a key research direction in the industry. In response to the problem of low accuracy of traditional methods in short-term power load forecasting, a short-term power load visualization forecasting method is proposed based on two-dimensional variational mode decomposition （2D-VMD） and convolutional long short term memory neural networks（ConvLSTM） . Firstly, the Gramian angular fields （GAF） method is used to convert the preprocessed load data into a set of Gram angular field images, and then the images are decomposed into a series of sub-modes with different center frequencies through 2D-VMD and classified according to the center frequencies. ConvLSTM neural network is used to predict the image groups with different modes. Finally, the prediction results are reconstructed and inversely operated to obtain the load prediction values. The prediction results indicate that this method improves the accuracy of short-term load forecasting and provides a new method for power load forecasting.

Evaluation and prediction of power network system robustness are conducive to system managers' ability to perceive the current status of network system operations and take timely measures to cope with potential risks. Therefore, a robustness prediction model for power dispatching data network is proposed based on an improved whale optimization algorithm. Firstly, an evaluation index system for the robustness of the power dispatching data network is constructed and data dimensionality reduction processing is designed based on methods such as field extraction and formula mapping. Moreover, an improved chaotic mapping and adaptive weight-whale optimization algorithm （CA-WOA-BP） is proposed based on chaotic mapping and adaptive weights （WOA-BP） and a power network robustness prediction method is established. Experimental results show that compared with the WOA-BP algorithm, the proposed improved algorithm speeds up the convergence of the prediction model while overcoming the situation of falling into local optima and reduces the absolute error percentage of the predicted values by 5.3 %, which helps users to discover the robustness of power dispatching data network systems more accurately and timely.

Aiming at the large number of regional distribution transformer, a large number of new loads, distributed photovoltaics, etc., and the enhancement of the random voltage fluctuation of distribution transformer. The voltage quality of the substation users is facing challenge. In order to better analyze and predict the over-limit characteristics of regional distribution transformer voltage, a bilayer clustering regional distribution transformer voltage prediction method based on correlation feature screening is proposed. Firstly, the number of overrun days of regional distribution transformers are taken as the first layer clustering feature, and the distribution transformers with normal and over-limit voltage properties are obtained. Secondly, for the over-limit voltage distribution transformers, an optimal metric matrix combining Pearson′s correlation coefficient and Euclidean distance is proposed to extract the contained information of the original data as the input of K-means to realize the bilayer clustering of regional distribution transformer. On this basis, the representative distribution transformers in the cluster are selected to characterize the distribution transformers of this category, and the convolutional neural network-bidirectional long and short-term memory- attention（CNN-BiLSTM-Attention）model is used to predict the distribution transformer voltage, which can extract the bidirectional information features of the input data, weight the important features, and obtain the bidirectional feature information from multiple time scales for prediction. Finally, the effectiveness of the proposed method is verified in a certain area of Shanghai.

With the development of interconnected power systems, low-frequency oscillations are becoming increasingly frequent. Researching measures to suppress these oscillations in order to enhance the stability and reliability of power systems has become an urgent issue. A method to suppress low-frequency oscillations in power systems is proposed by utilizing magnetic controlled transformers with additional damping control. Firstly, the working principle of the magnetic controlled transformer is introduced and its mathematical model is derived. Secondly, a linearized model of a single machine infinite bus system incorporating magnetic control transformers is established. The damping characteristics of magnetic control transformers are analyzed, and the principle that the magnetic control transformer provides damping for the power system is explained. Then the method of designing additional damping controller parameters is studied primarily based on the frequency domain method, and calculation examples of single machine infinite bus systems are provided. Finally, the effectiveness and feasibility of the additional damping controller for the magnetic controlled transformer are verified by Simulink time-domain simulation.

Soft open point （SOP） can provide real-time power flow control, fast fault self-healing and feeder load balancing, and has been widely used in distribution networks. However, grid voltage dips （GVD） are typical large disturbance conditions, and the stable operation mechanism of distribution network SOP under these conditions cannot be analyzed through small signal methods. Therefore, the mixed potential theory （MPT） is used to study the stability of SOP under GVD conditions, and the parameter optimization strategies and virtual capacitor control strategies are proposed. Firstly, a nonlinear average model of SOP is established. On this basis, the large signal stability criterion （LSSC） for SOP under GVD conditions is derived using MPT. In addition, in order to enhance the stability of the SOP system, parameter optimization strategies and control strategies for constructing virtual capacitors are proposed based on LSSC. Finally, the effectiveness of the above method is verified through MATLAB/Simulink time-domain simulation, namely: 1） based on LSSC, the stability of SOP under GVD can be effectively analyzed; 2） the system prameter optimization strategy based on LSSC and the construction of virtual capacitor control strategy can effectively improve the stable operation ability of the system under large disturbance conditions.

Applications of deep learning methods in the classification task of power quality disturbances (PQDs) are becoming increasingly popular. Aiming at the limited availability of measured label data compared to the ability to generate simulated data in large batches, a disturbance classification method is proposed based on recursive graph theory and pre-trained transfer learning. Firstly, the PQDs signals are transformed into two-dimensional recursive images using the recursive graph algorithm. Then, a VGG-16 deep learning network is pre-trained using a large amount of simulated data, and the model's weight parameters are saved. Finally, through transfer learning method, the fully connected layer of the model is fine-tuned using a limited amount of measured data, enabling deep feature extraction and classification of PQDs signals under the constraint of limited training samples. Simulation and measured data validation demonstrate that the proposed method achieves high classification accuracy even with limited labeled data.

With the continuous expansion of electric taxi fleets in various cities, their large-scale and unregulated charging behaviors will not only aggravate the fluctuation of the power grid load curve, threaten the stability and security of the power grid, but also reduce the equipment utilization rates of remote charging stations. Therefore, an electric taxi orderly charging guidance system is established based on the information cooperation of “vehicle-station-network”, which proposes a spatiotemporal bi-level charging optimization of electric taxi considering the feasibility of user participation in scheduling. In the temporal level model, a multi-objective optimization model is established based on minimizing the peak-valley difference, standard deviation and the charging cost of users. This model optimizes the charging time load of electric taxis in advance and reserves charging time windows for each user. In the spatial level model, real-time scheduling of the spatial load of electric taxi charging is carried out with the objectives of balancing the average utilization rate of each charging station and minimizing users′ charging time cost, in order to allocate the optimal charging station for each user. Finally, the effectiveness of the proposed optimization strategy is verified through a case simulation, demonstrating its ability to balance the interests of power distribution network, charging station operators and electric taxi users.

The increasing penetration rate of wind power and photovoltaics （PV） in new power systems exacerbates voltage fluctuations in distribution networks, while energy storage （ES） and electric vehicles （EV） play important roles in reducing voltage fluctuations in distribution networks. At the same time, smart meters, smart sensors and improved communication networks are widely deployed, the amount of data available is increasing, and data-driven technology is emerging.This paper proposes a multi-agent deep reinforcement learning （MADRL） based dual time scale active and reactive power coordinated voltage control strategy for distribution networks.Using the double deep Q-network algorithm （DDQN） to solve the optimization problems of capacitor banks （CBs）, on line tap transformers （OLTC）, and ES active and reactive power at a slow time scale. At a fast time scale, the EA-MASAC algorithm with attention mechanism is used to enhance the reactive power of PV, wind turbines （WT）, and static var compensators （SVCs）, as well as the active power of EVs. Finally, the effectiveness of the proposed method is verified on an IEEE-33 node system.

In view of the fluctuation of wind farm output, a method of capacity allocation of hybrid energy storage system （HESS） based on a two-stage wind power fluctuation smoothing model is proposed, and evaluation indexes for evaluating the energy storage allocation effect are constructed. Firstly, the optimization model of wind power fluctuation smoothing is established, and the wind power sequence and HESS power sequence after smoothing are obtained. Secondly, the discrete wavelet transform is performed on the HESS power sequence, which is reasonably allocated to lithium batteries and super capacitors according to the principle of minimum cost of energy storage allocation combined with the characteristics of energy storage. Then, the capacity allocation of HESS under multiple scenarios is carried out. Finally, the evaluation indexes of energy storage allocation effect are constructed. Through the analysis of examples, it is shown that the proposed method can avoid the change of grid-connected wind power energy and improve the ability to follow the original wind power under the premise of achieving fluctuation smoothing. The evaluation index of energy storage allocation effect verifies the superiority of the proposed method in the capacity allocation of HESS.

Aiming at the problem of rough evaluation of voltage quality in distribution networks and the lack of consideration for improving system voltage quality and coordinating with reactive power compensation equipment through grid side energy storage, an optimized configuration method for grid-side energy storage is proposed that takes into account voltage timing characteristics, line loss, and operational economy of the distribution network. Firstly, the multi-attribute comprehensive evaluation index system of voltage quality is established based on the three dimensions of bidirectional voltage exceeding the limits, fluctuation characteristic and time series trend. Secondly, the calculation method of energy storage cycle life is refined, and then the energy storage benefit model of grid side is put forward considering line loss and cycle life. Then, a two-layer model of energy storage optimization configuration is established. The upper layer maximizes the life cycle benefit of energy storage and optimizes its power and capacity. Aiming at the maximum comprehensive evaluation value of network voltage quality and the minimum operation cost of reactive power regulation equipment after storage allocation, the optimization model of the operation strategy of the storage power layer is established. Finally, a numerical example is designed to verify the effectiveness of the proposed model and method.

With the widespread popularity of distributed photovoltaics, charging piles, 5G base stations and other new infrastructure, residential stations are prone to problems such as large peak-valley difference and reverse power transmission. In order to solve the above problems, by means of fully mining the energy storage potential of a variety of elastic resources, an optimal dispatching strategy is proposed, which takes into account the coordination of multiple elastic resources. Firstly, the mathematical model of each unit in the residential station area is established, and the energy consumption law and energy storage potential of the three elastic resources of electric vehicle, 5G base station and temperature control load are analyzed. Secondly, taking the power of the three elastic resources as the decision-making variable, considering the constraints such as power balance and equipment characteristics, an optimal scheduling model with peak shaving and valley filling and new energy consumption as the goals is established, and the MOMVO algorithm is used to solve the optimal scheduling scheme. Finally, considering the changes of electric vehicle penetration rate in the short-term, medium-term and long-term time scales, the proposed optimal scheduling strategy is simulated and verified by example. The results show that compared with the original station, after optimizing the scheduling of various elastic resources, the peak-valley difference of power load in the platform area decreases by 58.8% on average, eliminating the phenomenon of voltage exceeding the limit and photovoltaic power generation reverse transmission. Therefore, the proposed method can fully mine the energy storage potential of different elastic resources, effectively reduce the overload operation of distribution and enhance the absorption capacity of new energy, which is of great significance for improving the safety reliability and energy utilization efficiency of residential areas.

To reduce the impact of intermittency and volatility of offshore wind power on the power grid, energy storage devices are often introduced to smooth out its output power fluctuations. In response to the low-carbon development process and considering the uncertainty of wind power, the capacity allocation method for energy storage based on a typical wind power output scenario has limitations. Taking into account the electricity sales revenue, carbon trading revenue, and various costs of the energy storage system of offshore wind farms, a capacity allocation method for hybrid energy storage system of offshore wind farm considering wind power uncertainty is proposed. The random opportunity constrained programming method is adopted to deal with the uncertainty of offshore wind power output, and wavelet packet decomposition method is used to allocate power between lithium batteries and supercapacitors, the allocation scheme is combined with economic evaluation to establish an optimized operation model for wind and energy storage system, with the goal of maximizing the daily net profit of the system, optimizing its capacity configuration. Taking the measured data of a domestic offshore wind farm as an example for simulation, the effectiveness and economy of the proposed scheme are verified. At the same time, the system adopting hybrid energy storage is more superior than single energy storage.

With the proposal of the "dual carbon" target and the gradual deepening of the construction requirements of new power systems, industrial users are no longer traditional passive demanders, but need to make bidding decisions in the market according to their own production processes and planning loads. They need to face the uncertainty of electricity prices while making optimal bidding decisions that do not affect their own production continuity. A scenario construction based on industrial process modeling is proposed for industrial users to participate in the day-ahead and real-time electricity market, fully considering the uncertainty of electricity prices and production continuity. A two-stage bidding risk avoidance model is proposed based on risk optimization technology. Furthermore, in order to study the optimal bidding strategy for industrial users, the rationality of historical data is considered to improve the accuracy of target day simulation and obtain more accurate load data as the boundary of market bidding optimization algorithm. Taking industrial cement production as an example, modeling and analysis are conducted on various stages of cement productions. A simulation is designed using day-ahead and real-time electricity price data from Guangdong Province. The simulation results prove from multiple aspects that the proposed method can optimize users' electricity market bidding strategies and reduce user costs.

According to calculations, achieving clean energy electrification will reduce global carbon emissions by 73.2%. The power industry plays a crucial role in CO₂ emissions reduction, making it essential to guide zero-carbon electricity consumption on the consumer side to support the achievement of the carbon peaking and carbon neutrality goals. Green electricity consumption calculation is a key technical requirement for zero-carbon electricity user certification, yet the current electricity market lacks relevant calculation methods. To address this issue, related research is conducted on the calculation method and market mechanism of green electricity consumption by zero carbon electricity users in the market environment. Firstly, based on an improved power transmission distribution factor power flow tracing method, a multi-level power flow tracing model is constructed, enabling real-time, multi-level, and high-precision calculation of green electricity consumption for zero-carbon electricity users. On this basis, a calculating method and market mechanism for grid-sourced green electricity consumption by zero-carbon electricity users are proposed, which covers access indicators, green electricity consumption calculation methods, and settlement mechanisms in line with existing electricity market rules. Finally, the proposed model is validated through a case study of a practical power grid in a southern city, demonstrating its effectiveness in providing efficient calculation methods and transaction mechanisms for cross-province, cross-regional, and neighboring grid-sourced green electricity users. Furthermore, the model guides zero-carbon electricity users to concentrate in distributed energy areas and shift loads to periods with higher system green power proportion, promoting the consumption of new energy.

In recent years, frequency domain reflection method (FDR) has been widely used in defect detection of power cables. However, multi-reflection of traveling waves and the reflection coefficient spectrum of FDR are non-periodic signals, which leads to false peaks in the positioning spectrum of FDR, and this problem can easily lead to serious misjudgment of the location of defects. In this paper, a new method for power cable defect location based on Pisarenko harmonic decomposition is proposed. The linear difference model is used to fit the probability density distribution of reflection coefficient spectrum and construct an auto-regressive moving average (ARMA) model to replace the fast Fourier transform (FFT) in the traditional FDR. The proposed method can effectively overcome the influence of attenuation, multi-reflection and dispersion effect on cable defect location. Then, Pisarenko harmonic decomposition method is used to estimate all frequencies in the reflection coefficient spectrum and further complete the accuracy location of defects. Finally, the proposed method is validated through simulation data and on-site data from a 58 km submarine cable. The results show that the proposed method can not only weaken the influence of false peaks on defect location, but also improve the location resolution.

Raindrops attached to the surface of grading ring on the power equipments such as insulators and bushings under rainfall and high humidity environment may trigger or intensify corona discharge phenomelons, which have become one of the main sources of surface corona effects in DC power equipment. Mastering the deformation characteristics of raindrops on the conductor′s surface is crucial to reveal the mechanism of corona discharge triggered by raindrops and to guide the anti-corona design of equipment. Starting from the difference in light transmission properties among solid, liquid, and gas, a shadow method optical platform is built to experimentally study the deformation phenomenon of the raindrops under a DC electric field. At the same time, based on the equivalence principle of fluid body force, a fast numerical simulation method of the three-dimensional electrohydrodynamic model is proposed. The deformation mechanism of raindrops is analyzed from experimental and simulation perspectives. The research results show that raindrop images with uniform background grayscale and clear gas-liquid boundaries can be obtained in the constructed optical platform. The proposed numerical method can significantly reduce computational time and ensure accuracy. Research has found that there is a reciprocating deformation phenomenon of raindrops in a DC electric field, similar to the AC electric field. And its deformation rate has a certain periodicity in the time domain. Analysis shows that the reciprocating deformation phenomenon of raindrops comes from the Maxwell stress on the surface of the raindrops perpendicular to the electric field direction. The research results can provide a reference for the study of the electroinduced deformation phenomenon of the raindrops and lay a foundation for further research on the corona discharge mechanism triggered by raindrop deformation.

The influence of controllable hybrid reactive power compensation on the transient characteristics of transmission lines needs to be studied urgently. Compared with conventional transmission lines, the actual breaking conditions of ultra high voltage(UHV)transmission line circuit breakers with controllable hybrid reactive power compensation are more stringent, which pose a great threat to system insulation. The most serious over-standard condition of transient recovery voltage of UHV controllable hybrid reactive power compensation line is taken as the research object. An equivalent model for single-phase grounding faults in UHV hybrid reactive power compensation lines is established, and the influencing factors of transient recovery voltage are analyzed. The transient recovery voltage(TRV) suppression measure of controllable hybrid reactive power compensation line is proposed. By combining the theoretical analysis of equivalent parameter lumped circuit with PSCAD/EMTDC simulation, the suppression effects and feasibility of fast linkage bypass series compensation, circuit breaker with opening resistance, and fast grounding switch on transient recovery voltage are discussed, and the best suppression measure of transient recovery voltage is obtained, which provides a practical reference for UHV engineering construction.