Construction of Measurement Model for Ultimate Carrying Capacity of Medium Voltage Distribution Network Based on Genetic Neural Network

In order to avoid the overload fault of distribution network and ensure the safety and stability of power supply of distribution network, a measuring model of ultimate load capacity of medium voltage distribution network based on genetic neural network is established. The model takes the maximum off-grid power of medium voltage distribution network substation as the objective function, and considers the constraints such as the capacity constraints of substation main transformer, line transmission capacity and switch station capacity stipulated by N-1 power supply safety standard. The key hyperparameters of Long Short term memory (LSTM) neural network are optimized by genetic algorithm, and an improved GA-LSTM (Genetic Algorithm Long Short-Term Memory) neural network is obtained. Combined with the determined objective function and constraint conditions, the measuring model of the ultimate carrying capacity of medium voltage distribution network is established to measure the ultimate carrying capacity of medium voltage distribution network. The experimental results show that the model can effectively measure the quarterly ultimate load capacity of two different medium-voltage distribution networks. The measurement results are accurate and reliable, with high measurement sensitivity, and excellent performance in comprehensive application performance, which is of great significance for the operation and management of distribution network.