The traditional Weighted Average Current (WAC) control method loses its loop-order reduction characteristic due to control delay, resulting in significant instability risk for the system in weak grid conditions. This paper, based on the equivalent impedance model, analyzed that the main cause of system instability is the introduction of a positive feedback loop associated with grid impedance through the voltage feedforward channel, resulting in phase lag. To address this issue, the paper proposes the introduction of a Second-Order Generalized Integrator (SOGI) in the voltage feedforward channel to attenuate the impact of the positive feedback loop on the system, improve the phase margin of the equivalent output impedance in the medium to low frequency range, and enhance the system's adaptability to a wide range of grid impedance variations. Furthermore, in order to eliminate control errors in the indirect control of WAC and enhance the system's background harmonics suppression capability, the quasi-Proportional Resonant (QPR) controller is combined to improve the quality of grid-connected current. Finally, the effectiveness and feasibility of this improved WAC control strategy are verified through simulation in MATLAB/Simulink. © Beijing Paike Culture Commu. Co., Ltd. 2024.

In order to alleviate the increasingly prominent energy problems and environmental problems, clean energy based microgrid technology has been widely studied and applied. Aiming at the coordination and optimization scheduling problem of multi-microgrid, the power transaction mechanism of multi-microgrid system is designed from the two aspects of multi-microgrid energy interaction and economic operation. Firstly, the two-layer scheduling mathematical model of multi-microgrid is established. Then, the improved Gray Wolf algorithm PSO-GWO is proposed and applied to the optimal scheduling problem of multiple micro networks. Finally, through simulation experiments, PSO-GWO, PSO and GWO are applied to solve the mathematical model of multi-microgrid optimal scheduling respectively, which proves that the established multi-micro-grid double-layer optimal scheduling model is feasible and effectively promotes the economic and stable operation of multi-micro-grid system. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Bidirectional DC-DC is widely used in energy storage systems and other fields, and its reliability is closely related to the stable operation of equipment. Electrolytic capacitor is considered as the component with the highest failure rate in the bidirectional DC-DC converter circuit, and its performance greatly affects the efficiency of the whole circuit. In this paper, a capacitance fault prediction model of bidirectional DC-DC converter is proposed based on support vector machine regression algorithm and LSTM-SVR algorithm. Firstly, the failure mechanism of electrolytic capacitor of the bidirectional Buck/Boost converter is analyzed, and the parameters reflecting its degradation law are determined. The capacitor degradation is simulated, and ripple voltage is selected as the fault characteristic parameter. Then, Support Vector Regression and Grid Search are used to optimize the long and short term memory network to determine its optimal network structure. Finally, through simulation experiments, compared with LSTM and SVR models, the prediction accuracy of LSTM-SVR reaches more than 97%, the accuracy of the model was verified. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Computational efficiency is an important aspect of power flow algorithms. A new combination algorithm based on higher-order Newton and simplified Newton method is proposed. Firstly, the solution principles of the high-order Newton method and the classical Newton method are compared and analyzed; Furthermore, the high-order Newton method and the simplified Newton method are combined, and the detailed solving steps of the combination algorithm are given. Through the reasonable collocation of the high and low order Newton method, the calculation time is reduced and the efficiency of power flow calculation is effectively improved. At last, based on the IEEE AC system examples, the validity and correctness of the proposed method for solving power flow calculation are verified by comparing the results of basic case test, algorithm efficiency and system heavy load, etc. © 2020 The Institution of Engineering and Technology

Accurate prediction of short-term photovoltaic power generation will help the grid dispatching department to make reasonable arrangements to ensure the safe and stable operation of the power system, thereby increasing the proportion of new energy in the power system. In order to improve the accuracy of photovoltaic power generation prediction, a short-term photovoltaic power generation prediction model based on deep belief network with momentum factor is proposed. It focuses on the selection of training samples for short-term prediction models of photovoltaic power generation, short-term prediction models of photovoltaic power generation based on deep belief networks, and optimization of the model by adding dynamically adjusted momentum factors. The actual measured data of the Australian Desert Solar Research Center were used to verify the short-term prediction model of photovoltaic power generation. The simulation experiment results show that the short-term prediction model of photovoltaic power generation established in this paper can effectively characterize the various factors that affect photovoltaic power generation and the measured the complex nonlinear relationship between powers has high prediction accuracy.
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021.