In order to improve the sampling efficiency of the non-sequential Monte Carlo simulation method, an improved hybrid method combining the analytical method and the significant Latin hypercube sampling method is proposed based on the idea of state space partitioning. The method partitions the system state space based on the determination of the significant state subspace to avoid sampling the zero-f...Show More
Due to the operation of regional networking, the scale of the power grid is becoming larger and larger, and a fault in the power grid needs to be located in the fault area timely and accurately. The models and structures of BP neural network and convolution neural network are analyzed. The training and test samples are constructed for a power grid model, and the BP neural network and convolution n...Show More
