Increasing penetration of distributed generators (DGs) brings new challenges to the control and protection of distribution networks. This article proposes a control and protection framework based on Multi-agent System (MAS), in which situation awareness of zone agents plays an important role. The main protection is designed to locate and isolate faults based on the peer-to-peer communication among the zone agents. Fault-tolerant mechanisms, including self-inspection and backup protection, are developed to cope with the malfunction of individual MAS and local communication failure. An automatic parallel restoration scheme that doesn't require complete grid information is proposed to restore the power supply. Comprehensive simulation studies are carried out on a 71-bus distribution network with DGs using DIgSILENT, and the results have shown the effectiveness of the proposed self-healing framework.

In order to adapt the rapid development of electric vehicles (EVs) in the future and reduce negative impacts of charging load on distribution networks, the reasonable charging facilities configuration of EVs is particularly important. This paper proposes an optimal planning method of charging piles. Firstly, a forecasting model of charging load is established based on the concept of trip chain and Monte Carlo Simulation Method (MCSM). Charging load profiles in different locations is then calculated. An optimal planning model is established to optimize the configuration of charging piles. Simulation results show that the proposed method can decrease both peak-valley difference and voltage deviation after the access of EV. This study can accurately forecast charging load demand in residential area, workplace and shopping center, and provide guidance for future planning of charging piles.

Medium- to long-term wind power output scenarios are crucial for power system planning and operational simulations. This paper proposes a two-stage hidden Markov model-based approach for modeling the time series output of multiple wind farms. First, based on the key features of the wind power output sequence, the daily typical patterns of wind power output are extracted. Then, the process of simulating the wind power output time-series is modeled as a two-layer temporal model. The upper layer uses a discrete hidden Markov model to describe the day-to-day transition process of wind power output patterns and the lower layer uses a Gaussian mixture hidden Markov model to describe the fluctuation process of wind power output values within each output pattern. Finally, the upper models corresponding to each quarter and the lower models corresponding to each pattern are trained respectively and the time-series scenarios of wind power output for multiple wind farms are generated quarter-by-quarter and day-by-day through Monte Carlo sampling. Validation using real-world wind power data demonstrates that the proposed method can effectively generate medium- to long-term output scenarios for multiple wind farms. Compared to traditional methods, the proposed method shows improvements in terms of accuracy, statistical characteristics, temporal correlation, and mutual correlation.

The widespread integration of photovoltaics (PVs) presents significant challenges to the operation and control of distribution systems, particularly in maintaining voltage stability at nodes with PV connections. To address these challenges, this paper proposes a voltage control algorithm based on distributed model‐free adaptive control (MFAC). The control objective is to achieve real‐time reactive‐power‐voltage coordination under constraints including PV power output limitations, voltage safety ranges, and the communication network topology. The proposed method estimates dynamic linearization parameters that represent the voltage control characteristics of the distribution systems by utilizing real‐time data from distributed PVs and enabling communication between adjacent nodes. Rather than relying on a precise network model, the algorithm achieves robust voltage control by estimating these parameters from historical and real‐time sampling data, employing a data‐driven approach to iteratively update control strategies. Multi‐scenario simulations of a 32‐bus power system demonstrated the effectiveness and robustness of the algorithm across diverse operating conditions.

Fast diagnosis of commutation failure in HVDC systems can improve the level of operation in the power systems. It is also the base to realize the prevention of commutation failure. By application of wavelet transform, the rapidity and correctness of faults diagnosing are enhanced. AC voltage and DC voltage in inverter side are taken as the signals for wavelet analysis. Wavelet coefficient d6 changed suddenly when fault occurred. This paper studied a diagnosing method based on the variance of wavelet coefficient d6. With moving sample windows, the sampling signals in the windows are analyzed with wavelet and the variance of wavelet coefficient d6 in real time. Therefore, the fault diagnoses are realized in real-time. The functions are realized in MATLAB/Simulink. The results of simulations show that the method can output the alarm signals after about half cycle when fault occurred. It can also tell out the AC fault without CF and DC line fault with CF fault. Above all, the diagnosing method using wavelet analysis ameliorates the rapidity and correctness of faults diagnosing.