Breakdown characteristics and voltage calculations of large-size sphere-plane long air gaps

With the increase in voltage level, the size of the metal fittings used in the valve hall or DC field of HVDC converter station becomes larger and larger. The switching impulse discharge characteristics of the air gap formed by such fittings are quite different from the typical rod-plane gap. Therefore, according to the sphere-ground gap and sphere-net/ground gap in the valve hall of converter station, the switching impulse discharge tests under different gap distances are carried out. Based on the experimental data, the calculation model of large-size sphere-plane gap discharge breakdown voltage based on cuckoo search algorithm optimization granular support vector regression algorithm is established. The model divides most class data into particles by K-means clustering method, and then extracts information particles and minority class data sets to train support vector regression model. The cuckoo search algorithm is used to optimize the parameters of support vector machine to obtain the optimal parameters. The prediction results of this model are compared with those of cuckoo optimization support vector regression model and BP neural network model, and the results show that the proposed algorithm can effectively improve the generalization performance of the model on minority data.