Abbreviation (ISO4): South Power Sys Technol
Editor in chief:
Under the background of the new power system, parameter identification and correction based on measurement data are key requirements for the digitalization and intelligence of distribution networks. A parameter identification method is proposed that integrates the mathematical model of the distribution network with measurement data collected from the low-voltage side of distribution transformers, aiming to address the online identification and correction of line and transformer parameters. Firstly, the topology is obtained by parsing CIM files, and the initial parameters of components are set according to asset data. Then, a quasi-Newton-Raphson method is employed to construct the Jacobian matrix based on multiple sets of measurement data, enabling iterative parameters updated by solving underdetermined or overdetermined equations. Finally, the identified parameters are used to recalculate the low-voltage side voltages, which are then compared with the measured values for validation. Experimental results based on data from a practical distribution network in South China demonstrate that, with increasing data sets and iteration times, the average relative errors converge to a stable level. This verifies that the proposed method achieves high accuracy and practicality, and can provide effective support for power flow calculation, state estimation, and digital twin construction in distribution networks.
Driven by the "dual carbon" goals and energy transition, distributed photovoltaic systems with their characteristics of "numerous points and extensive coverage" are connected to distribution networks, making power interactions between distribution network levels more complex. Traditional supply-side regulation methods struggle to meet the operational demands of multi-source collaboration and multi-objective coordination. To address this, a multi-level multi-objective collaborative optimization method for distribution networks-substation areas-users is proposed, considering dynamic comfort zones. Firstly, a multi-level structure and hierarchical scheduling framework for distribution networks, transformer areas, and users is established. The distribution network is divided into the distribution network layer, transformer area layer, and user layer. Within each layer, source-grid-load-storage resources achieve collaborative interaction, while bidirectional power transfer and information exchange occur between layers, promoting rational resource allocation and utilization. Secondly, a flexible resource model incorporating dynamic comfort zones is developed to quantify residential users′ resource regulation potential under varying electricity prices, enabling real-time adjustment of regulation boundaries. Then, a multi-level multi-objective collaborative optimization model for distribution networks-substation areas-users is formulated, comprehensively considering optimization objectives across different levels to establish a bottom-up coordination mechanism. Finally, the proposed strategy is validated through case simulations. The results demonstrate that the method effectively maximizes and utilizes source-grid-load-storage resources across all levels, significantly improving photovoltaic integration, optimizing load curves, reducing user electricity costs, and ensuring safe and economical distribution network operation
The widespread integration of distributed power sources affects the power flow distribution of the system. Reactive power optimization can maintain voltage stability and improve operational economy. To address the insufficient regulation capability of traditional reactive power devices, the introduction of a soft open point with self-energy storage （E-SOP） enhances the reactive power-voltage optimization performance of distribution networks. In the day-ahead stage, a long-term reactive power optimization model is developed to determine the scheduling strategies for on-load tap-changing transformers, switchable capacitors, and energy storage devices. In the day-to-day stage, a rolling optimization model based on short-term source-load forecasts is established to refine the day-ahead optimization results. The proposed multi-time scale reactive power optimization method for flexible distribution networks considers the modeling accuracy of E-SOP loss characteristics, achieving coordinated optimization of continuous and discrete reactive power resources while reducing operational costs and voltage deviations. Based on convex relaxation techniques, the aforementioned reactive power optimization model is transformed into a mixed-integer second-order cone programming problem, and an accelerated technique based on an induced objective function （IOF） is further introduced to enhance the computational efficiency of the reactive power optimization model. The IEEE 33-node system after retrofitting is used as an example for analysis, and the results verify the feasibility and effectiveness of the proposed method.
A grid-forming photovoltaic energy storage access optimization control strategy based on improved chaotic particle swarm optimization is proposed to address the issues of voltage fluctuations, low utilization of renewable energy, and line losses caused by the integration of traditional photovoltaic energy storage grid-forming inverters into active distribution stations. Firstly, a multi-objective optimization model is constructed with the objectives of minimizing load voltage deviation, maximizing photovoltaic power generation, and minimizing line losses. According to system requirements, the optimization priority for load voltage deviation is set to be the highest and the line loss is the lowest. Multi-objective problems are transformed into three single objective optimization problems through priority sorting. To improve the efficiency of solving, the initial parameter selection strategy of the chaotic particle swarm algorithm is improved. The simulation results show that the proposed method has faster convergence speed and higher solution accuracy, which verify the effectiveness of the control strategy.
Ultra-high-penetration distributed photovoltaics （PV） in distribution networks pose significant challenges, including severe reverse power flow overloads and voltage violations. Addressing the strong coupling characteristics between active-reactive power （P/Q） and voltage under complex network impedances of distribution networks, a active-reactive power coordinated voltage support optimization method is proposed. This method achieves multi-objective collaborative optimization operation of distributed renewable energy and energy storage, balancing low operational costs with high power quality. Firstly, a rolling optimization method for three-phase power flow voltage model predictive control （MPC） based on active-reactive power coupling is proposed. Via a multi-period voltage MPC framework, the number of decision variables and constraints is effectively reduced, which improves solution efficiency while ensuring solution accuracy and achieves fast solution of the online voltage control problem. Furthermore, a pre-iterative method based on fixed-point iteration is proposed for the strongly coupled three-phase power flow model. By distributing multi-step iterations across the rolling optimization horizon, the solution accuracy is significantly improved without increasing the number of per-step iteration. Finally, the effectiveness of the proposed control system in suppressing voltage exceeding limits is verified by integrating an IEEE 13-node system with ultra-high-penetration PV.
To address the issues of insufficient dimensions, lack of objectivity, and weak coupling with incident data in current power supply enterprise risk assessments, a comprehensive systematic risk assessment method based on generalized information is proposed. Firstly, complex risk factors faced by power supply enterprise are systematically analyzed. A comprehensive systematic risk assessment framework based on generalized information is established, enabling a multi-dimensional characterization of safety risks. On the basis, considering the differences between assessment dimensions and risk inter-coupling, an indicator weighting method integrating fault tree logic structures and expert knowledge is proposed to accurately determine the importance weights of each dimension. Subsequently, recognizing the impact of assessment indicator characteristic cycles on systemic risk, data features embedded within incident samples are deeply mined. A risk dynamic identification model based on accident energy mapping is developed, and an improved grey wolf optimizer algorithm is proposed to solve the model. This enhanced algorithm incorporates a large-scale incident sample-driven convergence mechanism and introduces chaotic initialization along with a cosine-based dynamic optimization mechanism to improve population diversity and balance global exploration and local exploitation capabilities across different solution stages, which significantly enhances solution efficiency and precision. Finally, the effectiveness of the proposed method is validated through a case study on a regional power supply enterprise.
In view of the large communication volume required by traditional centralized methods in active distribution network fault section location, as well as the challenges faced by existing artificial intelligence methods such as complex models and resource constraints when deployed in a distributed manner, a two-stage collaborative intelligent fault section location method of distributed deployment and centralized diagnosis is proposed. This method first uses variational mode decomposition to extract high-dimensional local fault features from the current signals collected by each distributed measurement unit, and then completes the preliminary diagnosis at the terminal through a lightweight multi-layer perceptron model, only uploading the low-dimensional fault section probability vector. Subsequently, at the centralized fusion end, taking these probability vectors as meta-features, meta-learning data is generated with the aid of K-Fold cross-validation, and a collaborative neural network based on Stacking is constructed to optimize and fuse multi-source information end-to-end, in order to output the final fault section discrimination result. The simulation experiments show that the proposed method can effectively integrate distributed observation information and achieve high-precision fault section location under various fault conditions. Its performance is significantly better than that of a single local model and a simple fusion strategy, providing a new and effective way for the rapid and accurate location of distribution network faults.
Aiming at the problem of multimodality risk coupling and resilience improvement in”source-network-load-storage” collaborative optimization after large-scale distributed generation access to distribution network, a dynamic reconfiguration method of elastic distribution network integrating multimodality vulnerability perception and demand response collaborative optimization is proposed. Firstly, based on the complex network theory, a topology-operation multi-dimensional vulnerability assessment model is constructed. By improving the betweenness center to quantify the topological hub of the branch, the voltage sensitivity coefficient is used to characterize the dynamic disturbance sensitivity of the node, and the principal component analysis is used to realize the unsupervised fusion of structural vulnerability and operational vulnerability, so as to accurately identify high-risk branches. Secondly, a second-order cone programming model with resilience constraints is constructed, and the comprehensive vulnerability index is embedded into the objective function as a penalty term to realize the collaborative optimization of energy storage dynamic scheduling, demand response elastic load and network topology. Finally, based on the simulation of IEEE 33-node system, the results verify that the proposed distribution network reconfiguration strategy reduces the network loss by 55.1 %, while the new energy consumption rate increases by 15.17 %. The comprehensive vulnerability-driven reconfiguration makes the high-risk nodes clear, the N-1 fault recovery time is shortened by 70 %, and the peak-time voltage fluctuation is suppressed by 40 %. Pareto frontier analysis reveals that 6.5 % economic cost increment can exchange for 22.7 % vulnerability index improvement, which verifies the multi-objective coordination of safety-economy-elasticity. It provides an “evaluation-optimization-feedback” closed-loop theoretical framework and basis for the new distribution network.
Neutral and ground lines wiring faults on the user side of low-voltage distribution systems（LVDS） lead to common yet highly concealed leakage current faults in practical engineering, posing significant challenges for operation and maintenance in LVDS. Existing methods for locating leakage faults struggle to address insufficient sample sizes and multicollinearity problems among users currents, making it difficult to rapidly and accurately identify leakage fault users in large-scale low-voltage power substation area. Firstly, the impacts and hazards of user-side wiring faults and leakage faults in the context of current graded residual current protection（RCD） and grounding system protection configurations in LVDS are analyzed. Further, the differences in the impact of load current on residual current in the substation area when there are faults in the neutral and ground wire connections for individual and multiple users are investigated. To address the limitations of insufficient samples due to restricted measurement capabilities in substation area and the multicollinearity among user load current curves, a ridge regression-based method is proposed for identifying leakage user caused by wiring fault in LVDS. This method can accurately identify leakage fault user in small sample sizes. Further considering the number of faults and the differences in their phases, simulation experiments of different wiring fault leakage cenarios are carried out to verify the effectiveness of the proposed method, and comparative tests against multiple linear regression method demonstrate its superiority.
With the proposal of dual-carbon target, energy storage is an important way to promote distributed PV local balanced consumption and reduce carbon emissions from the distribution network, but the traditional energy storage configuration scheme usually only considers the electric power balance, and how to comprehensively consider the electric-carbon characteristics puts forward higher requirements for the optimal configuration of energy storage. Firstly, a cluster division method based on electrical distance and net load carbon flow rate is proposed by introducing the node net load carbon flow rate index for cluster division of distribution networks, which improves the efficiency of distribution network management and the accuracy of carbon emission management. On the basis of the cluster classification, an energy storage allocation method that takes into account the carbon emission cost of the distribution network is further proposed, which takes into account the carbon emission cost and network loss cost, as well as the economic cost and benefit of energy storage allocation, establishes a carbon balance mechanism for energy storage, quantifies the impact of the change of the carbon content within the energy storage on the distribution network, and avoids carbon dioxide emissions into or out of the energy storage equipment. Finally, the results of an arithmetic simulation in a distribution network in Anhui province show that the proposed method reduces the total carbon emissions of the distribution network by 11.12 % compared to the unconfigured energy storage system.
Photovoltaic energy storage configuration is the current development trend, but the value of energy storage investment directly affects the interests of energy storage owners and the actual market behavior, it is necessary to consider the interaction between energy storage benefits and costs, and carry out further research on the basis of traditional multi-objective optimization of the grid side or the maximum economic benefit of energy storage. Therefore, a two-layer optimal configuration model of distribution network energy storage with distributed photovoltaic, which adopts multi-objective particle swarm optimization and considers the optimal marginal benefit. Firstly, the concept of marginal benefit of energy storage is put forward, the marginal benefit model of energy storage is established, and the benefit of energy storage is measured from the perspective of energy storage owners. Then, a two-layer optimization model is established. The outer layer optimization takes the maximum marginal benefit and the minimum voltage deviation after energy storage access as the target, and takes the network loss rate as the constraint condition. The inner layer optimization takes the maximum daily energy storage income as the target to optimize the energy storage output. Finally, the IEEE 33-node system is used for energy storage optimization calculation. The results show that on the basis of traditional energy storage capacity optimization with net income over the whole life cycle, based on the optimal marginal benefit of the energy storage owner, the secondary allocation of energy storage capacity with the goal of considering marginal benefit and voltage deviation can improve the net income, line loss rate and voltage deviation after energy storage access.
The installation of energy storage systems can mitigate the intensified vulnerability caused by the integration of large-scale distributed photovoltaic (PV) systems into the distribution network. Establishing a comprehensive distribution network vulnerability assessment system can effectively enhance the benefits of energy storage integration. The operational vulnerability of the distribution network is evaluated by considering PV integration and fluctuations in voltage and load, a comprehensive vulnerability index for the distribution network is established. A distribution network energy storage optimal allocation model is developed with vulnerability indicators，economic indicators and active power losses indicators as objective functions. The PV and load uncertainties are addressed using the K-means clustering algorithm, and the model is solved with an improved multiple objective differential evolution algorithm (IMODE). By initializing the population with chaotic sequences, adaptive mutation and crossover factors are introduced, along with optimizations to the mutation operator strategy. Simulation verification is conducted on a distribution network case in a county of Gansu Province. The proposed strategy effectively ensures the safe operation of the distribution network while considering economic factors, reducing network vulnerability by 18.9 % and active power loss by 23.9 %. The proposed vulnerability indexes and model can quantitatively assess distribution network vulnerability and reflect the improvement effects of energy storage integration, providing technical references for the installation and operation of energy storage systems.
To address the problem of unreasonable energy storage configurations caused by ignoring the differences in source-load power matching and control demands, which impacts efficiency and harms stakeholders' interests, an optimized energy storage configuration method is proposed considering the variations in photovoltaic-load scenarios. Firstly, from the perspectives of photocharge power differences, matching and distribution, the differences in demand for energy storage configuration operation in different typical scenarios are analyzed. Secondly, according to the differences in energy storage control demand of each scenario, the original probability coefficients of each scenario are corrected to add demand focus. Then, based on the new probability coefficients of each scenario, an energy storage optimization configuration model is established, with the goal of improving operating efficiency and optimal loss cost of energy storage, and a variety of methods are used to convexize the nonlinear terms in the model. Finally, based on the improved IEEE 33-node distribution network, a Gurobi solver is used to obtain the energy storage configuration scheme and verify the effectiveness of the proposed configuration method.
With the rapid growth of terminal energy micro-unit consumption, represented by residential buildings, the optimization of power dispatch in user-side new energy microgrid systems face challenges. A hierarchical optimization strategy for new energy microgrids, considering source-load uncertainty and electric vehicles, is proposed to ensure residential electricity demand and improve system operational efficiency. The upper-level model optimizes residential electricity loads and electric vehicle charging loads, with objectives focused on maximizing electric vehicle charging benefits and resident electricity satisfaction while accounting for demand response uncertainty. The lower-level model aims to enhance system operational efficiency and external impact, employing an interval information fuzzy planning model to construct and solve nonlinear multi-objective optimization problems based on upper-level data, thereby obtaining the optimal power output scheme for system equipment. Results demonstrate that the proposed dispatch strategy effectively improves resident electricity satisfaction while ensuring overall system performance, mitigating the impacts of uncertainty factors.
Isolated microgrids are susceptible to multiple disturbances while they are operating independently. Fluctuations in renewable energy outputs and load variations can distrupt power balance, causing voltage and frequency deviations. And long-distance communication and network congestion may induce control command transmission delays that compromise the real-time performance of secondary control. To address these challenges, a fixed-time distributed secondary control algorithm based on a dynamic event-triggered mechanism（DETC-P） is proposed. To counter periodic disturbances, an H∞ robust control strategy is adopted to strengthen disturbance rejection capability.For communication delay issues, model order reduction techniques are employed to simplify the dynamic models of distributed generation units, enhancing delay compensation efficiency. The proposed dynamic event-triggering mechanism adaptively adjusts triggering thresholds based on voltage and frequency deviations, significantly reducing communication burden between nodes while ensuring a lower bound on event-triggering intervals and avoiding Zeno behavior. Finally, a system model of the islanded microgrid is constructed, and simulation experiments are conducted to validate the feasibility of the proposed algorithm.
Accurate ultra-short-term prediction of photovoltaic （PV） power is pivotal in mitigating the adverse effects of PV output uncertainty on power systems, thereby furnishing a dependable foundation for grid decision-making and scheduling. Therefore, a method for ultra-short-term prediction of photovoltaic output based on deep learning technology is proposed. Firstly, leveraging historical PV data, feature correlation analysis and K-means++ weather clustering are performed to discern pertinent patterns. And employing strategically positioned quantiles, a novel bidirectional long short term memory （BiLSTM） neural network is then constructed to capture bidirectional temporal dependencies. Then, a self-attention mechanism （SA） is introduced to dynamically emphasize key sequential information and a particle swarm optimization algorithm is integrated to optimize the parameters of the neural network. Optimal parameters are then assimilated into the BiLSTM-SA optimization framework for point prediction. Finally, through meticulous error analysis, a quantile regression （QR） is developed to delineate QR-BiLSTM-SA probability prediction model. The results show that the proposed method achieves accuracy exceeding 95% in ultra-short-term PV output probability prediction and has a good generalization ability, offering a robust foundation for the operation and scheduling of modern power systems.
ISSN 1674-0629 (Print)
Started from China Southern Power Grid Corporation
Published by: China Southern Power Grid Corporation
