高功率密度碳化硅MOSFET驱动器设计:集成有源米勒钳位的600kHz开关应用

With the accelerating application demand for new energy-based power systems, a energy storage system (ESS) optimization configuration method is proposed to fully consider the load demand and the uncertainty of photovoltaic (PV) output and improve the adaptability of the ESS configuration scale. Firstly, the Gaussian kernel function of the kernel method is used to map the temporal variation characteristics of the PV output scenario to establish a comprehensive evaluation index scenario for the uncertainty and temporal correlation of PV-Load. The iterative self-organizing data analysis algorithm (ISODATA) is used to optimize clustering of the comprehensive evaluation index scenario to generate PV-Load typical scenarios. A distribution network ESS configuration model is constructed with the objectives of minimizing voltage fluctuation indicator, line loss rate, and minimizing ESS investment cost. To solve the high-dimensional multi-objective functions in the model, an non-dominated sorting genetic algorithm-III (NSGA-III) was employed, and the optimal solution is selected by using the entropy weight method (EWM). Finally, the proposed method and model are analyzed and verified by case simulation, and the results show that the proposed model and method can comprehensively consider the actual operating characteristics of PV and load, and can effectively formulate the ESS configuration scale and operation strategy in the distribution network.