Wind-intensive power system is under significant pressure during operation to perform peak regulation, reserve provision, and wind power accommodation. To alleviate the energy management burden of power system, large-scale adiabatic compressed air energy storage (A-CAES) with parallel compression and constant/sliding-pressure expansion designs can be incorporated into the dispatchable resources during day-ahead dispatch, thereby fully utilizing its energy throughput capability. Therefore, considering the participation of A-CAES, this paper develops a day-ahead dispatch strategy for wind-intensive power system. Firstly, a nonlinear thermodynamic mathematical model of A-CAES is established considering its variable-parameter characteristics. Different from previous research, this model integrates the effects of parallel compression and constant/sliding-pressure expansion. Secondly, a day-ahead dispatch model for the power system integrated with A-CAES, thermal power units, and wind farms is constructed, which includes the constraints imposed by the two aforementioned configurations. Furthermore, an iterative solution method based on the framework of master problem and sub-problem is designed to efficiently and accurately solve the nonlinear dispatch problem. Finally, case studies are conducted. The results indicate that large-scale A-CAES with parallel compression and constant/sliding-pressure expansion designs plays a constructive role in promoting the safe, economic, and clean operation of power system.

Abstract To address the “trend-fluctuation-noise” multi-frequency coupling characteristics exhibited by net load in distribution grids under high renewable energy penetration, and the resulting prediction accuracy issues, this paper proposes a combined forecasting method based on wavelet decomposition, three-model collaborative prediction and Ruppel Fox Optimization weighted fusion. Firstly, wavelet decomposition decouples the net load sequence into component signals with distinct frequency characteristics. Secondly, based on the characteristics of each component, LSTM, LGBM, and RF models are differentially introduced for component prediction. The RFO algorithm, which possesses strong global optimization capabilities, is incorporated to adaptively optimize the weighting coefficients of the three models within each component. Finally, the weighted fusion results of all components are reconstructed via inverse wavelet transform to obtain the final net load forecast for the distribution network. Case studies based on the ISO-NE dataset demonstrate that the proposed method achieves an average absolute percentage error of only 2.97% on the validation set. This represents a reduction of 38.4% to 65.7% compared to traditional single models and a decrease of 4.6% to 5.2% relative to the particle swarm optimization-weighted fusion model. The research findings confirm that this method can accurately capture the multidimensional characteristics of net load, effectively enhancing prediction accuracy.

Both frequency instability risks and wind power accommodation burdens of wind-intensive power system (WIPS) merit sufficient attention. In this regard, large-scale adiabatic compressed air energy storage (A-CAES) with parallel compression and constant/sliding-pressure expansion designs can provide high-quality rotational inertia support, primary frequency regulation, and peak regulation services. Its rational dispatch highly benefits the secure and economic operation of WIPS. Consequently, a frequency security-constrained dispatch (FSCD) strategy for A-CAES-integrated WIPS is proposed. First, the variable-parameter thermodynamic and variable-gain frequency response models of A-CAES are constructed, based on which the analytical expressions for frequency security indices are derived. Second, an FSCD model is developed by incorporating the boundary constraints of the aforementioned indices under positive and negative power disturbances. Third, to address both temporal-coupled and temporal-independent nonlinear constraints, a temporal-classified iterative method is designed. Finally, case studies are conducted. The results indicate that the FSCD strategy can effectively leverage the roles of A-CAES in stabilizing frequency and promoting wind utilization.

Reasonable planning and reliable black-start sources are particularly important for the rapid restoration of power system after blackout. During restoration, the frequency regulation capability of system is weak. It is necessary to consider frequency security constraints in decision-making to ensure the feasibility of restoration plans. Additionally, large-scale A-CAES with parallel compression and constant/sliding-pressure expansion designs possesses advantages of high energy storage density, flexible operation, and simultaneous charge-discharge. It is suitable as a black-start source. In summary, an FSCR strategy for A-CAES-integrated power system is proposed. First, the variable-parameter thermodynamic model of A-CAES with the above two designs is established, considering simultaneous compression and expansion. Second, from a practical perspective, the frequency response model of system is developed by taking SSPR as disturbance, and an equivalent calculation method for frequency security indices is further proposed. Third, incorporating frequency security constraints as well as restoration constraints of units, loads, and topology, the FSCR model is constructed. An iterative solution method is adopted to address its nonlinearity. Finally, case studies are conducted. The results show that, compared with the existing LSPS-based FSCR strategy, the proposed strategy does not compromise restoration performance and even improves load cumulative restoration by 1.1%, while the number of time points with frequency violations decreases from 5 to 0. This indicates that the proposed strategy can utilize A-CAES to restore power system efficiently while mitigating frequency deviation during this period.

The integration of multi-type resources provides new ideas for the black-start of active distribution networks (ADNs). However, the inability to deal with uncertainty will lead to problems such as frequency/voltage crossing limits, scheduling difficulties, and even restoration failures. To this end, an ADN black-start strategy considering source-load bilateral uncertainty and multi-type resources is proposed. The forecast error uncertainties of renewable energy sources (RESs) and loads are characterized in intervals based on Copula theory, which are then introduced into the black-start model of ADNs and solved by the column-and-constraint generation algorithm. Case studies based on the improved IEEE 33-node system indicate that the proposed strategy can effectively cope with source-load bilateral uncertainty and realize robust restoration of ADNs. The system also achieves a 99.97 % RESs consumption ratio and a 67.36 % power utilization ratio of energy storage devices. Compared with existing methods, our strategy can give a more economical and faster restoration scheme while considering safety, which can be deployed in dispatch centers to help operators make full use of existing resources to achieve black-start safely and stably after outages. However, the computational time will increase if it is migrated to grids with large topologies, which needs to be further investigated.

With the increasing penetration of distributed photovoltaics (DPV) in distribution networks, the coupling and interaction between distribution networks and the upper-level grid have been continuously enhanced. To mitigate the risk of grid instability caused by the large-scale integration of DPV, it is imperative to establish transient equivalent models for distribution networks with high-penetration DPV. However, existing studies fail to meet the requirements of timeliness and interpretability for equivalent modeling of distribution networks in practical applications. In view of the above, this study proposes a transient equivalent modeling method for distribution networks with high-penetration DPV. Firstly, based on the intermediate equivalent model of distribution networks and industry standards, a clustering method is proposed using unit types and nodal voltages. Secondly, the initial clustering results are optimized based on nodal voltage magnitudes and total power outputs of sub-clusters. Thirdly, equivalent models for DPV are developed, and the equivalent impedance between each equivalent DPV units and the point of common coupling (PCC) of the distribution network is calculated. Subsequently, the equivalent load power is calculated based on the exchange power at PCC, as well as the output of the equivalent DPV model. Finally, simulations on the PSASP7 platform validate the proposed method.

This paper proposes a stepwise expert-teaching reinforcement learning framework for intelligent frequency control in hydro–thermal–wind–solar–compressed air energy storage (CAES) integrated systems under high renewable energy penetration. The proposed method addresses the frequency stability challenge in low-inertia, high-volatility power systems, particularly in Southwest China, where large-scale renewable-energy-based energy bases are rapidly emerging. A load frequency control (LFC) model is constructed to serve as the training and validation environment, reflecting the dynamic characteristics of the hybrid system. The stepwise expert-teaching PPO (SETP) framework introduces a stepwise training mechanism in which expert knowledge is embedded to guide the policy learning process and training parameters are dynamically adjusted based on observed performance. Comparative simulations under multiple disturbance scenarios are conducted on benchmark systems. Results show that the proposed method outperforms standard proximal policy optimization (PPO) and traditional PI control in both transient response and coordination performance.

Large-scale adiabatic compressed air energy storage (A-CAES) is a crucial technology for achieving high penetration of renewable energy. The rational formulation of its power schedules is a prerequisite for maximizing its value. However, existing A-CAES dispatch strategies suffer from oversimplified or overidealized models and low solving efficiency. In response, this paper develops a nonlinear dispatch model for A-CAES under variable working conditions, aiming to accurately capture its off-design operating characteristics, and proposes a rapid solution method. Firstly, a nonlinear thermodynamic mathematical model for A-CAES is established under compression, generation, and shutdown conditions, comprehensively considering the variations in ambient temperature, isentropic efficiency, compression ratio, expansion ratio, and temperatures of air reservoir and hot water tank. Then, the aforementioned A-CAES model is integrated into the day-ahead economic dispatch model of power system, which also includes thermal power units and wind farms. Furthermore, to address the challenge of straightforwardly solving the strongly nonlinear dispatch model, an iterative solution method based on the framework of master problem and sub-problem is proposed. Finally, case studies are conducted. The results indicate that the solution time of the proposed method is 412 s, which is only 1 % of the conventional method, while achieving almost identical economic objectives.

The increasingly frequent severe convective weather poses a serious threat to the reliability of distribution networks. In this regard, an optimal maintenance model that considers pre-disaster defense and post-disaster mitigation stages is proposed to meet the demand for resilience enhancement. During the pre-disaster defense stage, the equipment maintenance plan and the corresponding operation strategy are decided in conjunction with the failure probability of each device. During the post-disaster mitigation stage, the emergency dispatch strategy of the grid is developed considering the diversity of damage results. On these bases, the optimal pre-disaster maintenance strategy considering post-disaster situations is identified by integrating the two stages' outage losses and operating costs. Finally, the numerical results show that the proposed strategy can give an optimal maintenance plan by combining the fragility of each device, which effectively improves the outage prevention and management capability of the distribution network under severe convective weather.

The parallel operation of bidirectional converters is a promising way to increase the power rating, efficiency and reliability, and maintain the power balance in an AC/DC hybrid microgrid. However, zero-sequence circulating current (ZSCC) and active-reactive disturbance current will be generated under the unbalanced conditions of the DC side and AC side. This paper aims to study the performance characteristics of the parallel converters and propose an effective control scheme to control the ZSCC and active-reactive disturbance current. Firstly, the generation mechanism of ZSCC and active-reactive disturbance current are analyzed based on an averaged phase-leg model. The coupling relationship is derived and two categories of disturbance can be found: the primary disturbance caused by the unbalanced condition of one side, and the secondary disturbance generated by the primary disturbance and the unbalanced condition of the other side. On this basis, a decentralized control scheme is proposed to suppress the primary disturbance, thereby eliminating the ZSCC and active power oscillation without the distortion of output current. Finally, simulation cases in PSCAD/EMTDC verify the correctness of the theoretical analysis and the validity of the proposed control scheme.

The autonomous economic control for the islanded AC/DC interconnected microgrid has various modes due to the restriction of the feasible regions of inner and inter-microgrid resources. To properly handle the cooperative relationship between resources with different regulation characteristics and realize the frequency, voltage stability, and economic operation of the AC/DC interconnected microgrid, a distributed control strategy considering the regulation boundary of the consensus variable is proposed. Firstly, the influence of the variable feasible regions on the consensus control objective is analyzed, and the fundamental principle of the consensus control algorithm considering the variable regulation boundary is expounded. Secondly, the trilevel control strategy consisting of the local equipment control and inner and inter-microgrid cooperative control is constructed. The local equipment control adopts droop control, and the cooperative control adopts consensus control to realize the elimination of the frequency and voltage deviation and the economic dispatch of active power within and between microgrids. Finally, the AC/DC interconnected microgrid model is built based on PSCAD/EMTDC, and the simulation results verify the effectiveness of the proposed control strategy. To properly handle the cooperative relationship between resources with different regulation characteristics and realize the frequency, voltage stability, and economic operation of the AC/DC interconnected microgrid, a distributed control strategy considering the regulation boundary of the consensus variable was proposed in this paper. The proposed control strategy consisted of the local equipment and inner and inter-microgrid cooperative control. The simulation results based on PSCAD/EMTDC verify the effectiveness of the proposed control strategy.image

The linkage, coordination, and complementary cooperation of energy supply can improve the efficiency of transportation and utilization. At present, the level of new energy consumption needs to be improved, the coordination of the source network load storage link is insufficient, and the insufficient complementarity of various types of power sources in the power system. This article fully explores the differences and complementarities of various types of wind-solar-hydro-thermal-storage power sources, a hierarchical environmental and economic dispatch model for the power system has been established. Among them, the upper level model takes the flexible consumption of new energy as the optimization goal, the middle level model uses a combination of hydropower station and energy storage to minimizes the fluctuation variance and peak-to-valley difference of the net load curve, the lower level model aims to achieve optimal environmental and economic benefits of the power system, comprehensively considers the coal consumption cost, startup and shutdown cost, energy storage operation cost, and pollutant emissions of thermal power units, determines the startup and shutdown mode and output power of thermal power units. Finally, an improved IEEE 6-machine 30-node system is used as an example for simulation analysis, the results show that after applying the proposed hierarchical environmental and economic dispatch strategy of the power system, the fluctuation variance and the peak-to-valley difference of the net load curve have been reduced by 46.3% and 31.5%, respectively, and the environmental and economic benefits of the system is improved by 5.1% compared with the traditional economic dispatch strategy. It can meet the requirements of energy system cleaning and decarbonization while improving the operation economy, which verifies the effectiveness of the proposed environmental economic dispatch model.

针对城镇配电网供电能力不足的问题,设计了一种集中-分布式混合压缩空气储能(CAES)电站的基本架构.该架构依托集中式先进绝热压缩空气储能(AA-CAES)电站建立分布式液态空气储能电站,可更有效地利用AA-CAES电站大规模储气室的优势,延展AA-CAES电站的功能.然后,建立了集中-分布式混合CAES电站的运行模型.该模型能准确描述混合CAES电站空气压缩、液化、贮存、运输、汽化、膨胀发电等过程中的能量流通、转化、存储和释放机理.基于混合CAES电站的运行模型,结合全寿命周期成本理论,建立了混合CAES电站的优化规划与运行模型.上述模型还包含扩建输电线路、增建燃气轮机、安装电池储能等备选方案的相关约束,以便对比安装混合CAES电站和其他方案的经济效益.最后,通过算例仿真验证了优化规划与运行模型的有效性.

With the increasing severity of global warming and the rapid evolution of cyber-attack techniques, the enhancement of distribution network (DN) resilience that can effectively cope with natural disasters and man-made attacks has received extensive attention. Hence, in the context of a high impact low probability (HILP) event where attackers launch malicious attacks after a natural disaster occurs, a multi-stage DN operation strategy for resilience enhancement is proposed considering multi-type resources, such as gas turbines, stationary energy storage systems, mobile energy storage systems (MESSs), repair crews (RCs) and network reconfiguration. Firstly, the operation process of DN is divided into five stages based on the time lag between the natural disaster and the man-made attacks: the normal operation stage, the disaster disruption stage, the post-disaster degradation stage, the man-made attack stage and the defense and recovery stage. Secondly, under the background of a coupled power-transportation system, the optimal scheduling model for each stage of DN is established considering the effects of dynamic changes in traffic flow on the movement process of MESSs and RCs. Finally, the model is converted into a 5-layer mixed integer linear programming (MILP) problem that can be solved by sophisticated optimization software according to the time series relationship, and a case study is carried out with a modified IEEE 33-node system and the corresponding transportation network. The results demonstrate that under extreme scenarios, the proposed model can guarantee the load-side power supply as much as possible through flexible cooperation between multi-type resources, effectively enhancing the resilience of the DN.

电网结构日益复杂,故障监测与防范的成本随之增大,而强对流天气的频繁出现使得设备故障与外界条件关联更加紧密,因此挖掘故障设备与故障因素间的关联关系,并对具有较高关联度的设备进行重点监测与防范,对电力系统的安全稳定运行具有重要意义.基于模糊频繁项挖掘算法,提出了 1种考虑气象因素的配电网设备故障关联挖掘模型.该模型首先从多元信息库中提取故障特征数据,采用Relief-F算法排除相关程度较小的冗余特征,通过数据预处理与数据整合,构建包含气象因素的故障关联特征库.其次,以故障关联特征库为基础,引入模糊集理论,提出基于模糊频繁项集挖掘算法的故障因素与故障设备关联模型构建方法.最后,基于故障关联模型进行了算例分析,结果验证了所提方法的正确与有效性.

In order to give full play to the frequency regulation ability of multiple types of resources such as wind power, energy storage, and controllable load in a microgrid, this paper proposes a hierarchical cooperative frequency regulation control strategy of wind-storage-load in a microgrid based on model prediction. Firstly, according to the operation characteristics of each resource in the microgrid, a hierarchical cooperative frequency regulation architecture of wind-storage-load is constructed. On this basis, the frequency regulation control models of wind power, energy storage, and controllable load are established, respectively, and the calculation method of the characteristic index of the system frequency response is proposed. Then, taking the maximum frequency deviation as the stratification index, a hierarchical cooperative frequency regulation control strategy of wind-storage-load based on model prediction is proposed, and a power compensation strategy for connecting the wind turbine frequency support is proposed for the wind turbine speed recovery stage. Finally, a microgrid model including wind power, energy storage, and controllable load is built on Matlab/Simulink for simulation analysis. The simulation results show that the proposed control strategy can control wind power, energy storage, and controllable load to participate in frequency modulation in advance, and improve the frequency stability of the system.

近年来,作为消纳可再生能源的有效方式之一,微电网在新型电力系统中扮演了愈来愈重要的角色,取得了显著的发展.但微电网的大量接入,改变了传统配电网的单级调度模式,使得调度过程中所面临的计算和通信任务日益繁重,而现行调度策略难以兼顾配电网运营商及微电网等多主体的利益诉求,也难以满足调度过程的计算高效性与通信私密性要求.对此,提出了一种基于多智能体算法的多微电网-配电网分层协同调度策略.首先,考虑配电网运营商与微电网在电力市场运行中的主从关系,构建了基于双层Stackelberg博弈的多微电网-配电网电力交易模型;然后,将多微电网-配电网协同调度表述为马尔可夫决策过程,采用基于数据驱动的多智能体深度策略性梯度算法求解Stackelberg均衡;最后,基于改进IEEE 33节点系统开展算例分析,验证了所提模型及算法的有效性.

电力线路周围树木生长与线路安全运行之间的矛盾(即树线矛盾)是影响乡镇配电网供电可靠性的重要因素.针对强对流天气来临之前乡镇配电网树线矛盾风险预警及清障处理需求,提出了一套"分区预警-总体权衡-优化处理"机制.首先,考虑强对流天气的地域关联特性,基于支持向量机建立了乡镇配电网各区域强对流天气-树线矛盾映射模型,实现树障接地风险的分区预警;然后,基于风险预警结果,以总体经济投入和停电损失最小为目标,建立了考虑强对流天气影响的乡镇配电网树线矛盾双层优化处理模型,决策各矛盾易发点的清障计划;最后,采用某乡镇强对流气象监测数据、树障接地故障记录以及改进的 IEEE 33 节点系统、123 节点系统进行了算例分析.结果表明,所提模型能够有效挖掘强对流天气与树线矛盾间的映射关系,并给出经济性最优的事前清障计划,实现乡镇配电网树障接地风险的分区预警及优化处理.

架空配电线路电杆使用数量大、分布范围广,在极端天气条件下易发生倾覆,对局部区域甚至整条配电线路的供电产生威胁.如何借助科学有效的方法对电杆的抗倾覆能力进行仿真建模与分析,现已成为配电网建设与改造过程中的关键技术问题.针对该问题,以10 kV钢筋混凝土电杆为例,首先,结合气象条件对其荷载及内力情况进行计算;其次,基于FLAC3D软件,提出一种架空配电线路电杆抗倾覆能力仿真建模与分析方法;最后,结合河南某地区土体参数开展算例分析,详细分析极端天气条件下地基基础的变形破坏过程并绘制荷载—位移曲线,可为架空配电线路电杆的安装和加固提供技术支撑.

Robust optimization (RO) is an important method to deal with the uncertainty of wind power. The main challenge is to reduce the redundancy of the uncertainty model, thereby improving the economy of the scheduling plan while ensuring its robustness. In this paper, aiming at the day-ahead robust scheduling problem, the uncertainty model of wind power is improved by fitting 3 typical characteristics. First, the kernel density estimation (KDE) model and the non-parametric Copula model are combined to fit the nonlinear correlation between forecast power and forecast error, and the forecast error boundary constraints are established. Second, by combining the Mallet algorithm, the autoregressive integrated moving average (ARIMA) model, and the t distribution model, the temporal dimension constraints are established to describe the time-series characteristics of wind power. Third, the spatial dimension constraints are established based on the high-dimensional non-parametric regular vine (R-vine) Copula model, and the spatial correlation of multiple wind farms is reflected. Based on the above wind power uncertainty model, a 2-stage day-ahead robust scheduling model is established. The case study shows that the proposed wind power uncertainty model helps to achieve the balance between economics and robustness of the scheduling plan.