Abstract
Air insulation strength relates closely to the electrostatic field distribution of the gap configuration. To achieve insulation prediction on the basis of electric field (EF) simulations, the spatial structure is characterized by a feature set including 38 parameters defined on a straight line between sphere electrodes. A support vector classifier (SVC) with particle swarm optimization (PSO) is used to establish a prediction model, whose input variables are those features. The EF nonuniform coefficient f of each sample gap is calculated and used for training sample selection according to the ranges off values. Trained by only 11-sample data, the PSO-optimized SVC model is employed to predict the power frequency breakdown voltages of 260-sphere gaps with a wide range of structure sizes. The predicted values coincide with the standard data given in IEC 60052 very well, with the same trend and minor relative errors. The MAPEs of the five predictions with different training sets are within 2.0%. The model is also effective to predict the breakdown voltages of Φ9.75-cm sphere-Φ6.5-cm sphere gaps, whose MAPEs are within 2.6%. The results demonstrate the effectiveness of the EF feature set and the generalization ability of the SVC model under the case of limited samples. This paper lays the foundation for estimating the dielectric strength of other air gaps with similar structures.
Highlights
Air is the most commonly used dielectric in electric power systems, and its strength is of vital importance for insulation design of many electrical equipment
This paper presents a feature set used for structural characterization of the sphere gap, and proposes a machine learning model based on particle swarm optimization (PSO)-optimized support vector classifier (SVC) for breakdown voltage prediction
1) The electric field (EF) feature set extracted along the sphere gap interelectrode path is effective to characterize its electrostatic field distribution, and these features can be used as input parameters of the PSO-optimized SVC model for sphere gap breakdown voltage prediction with a wide range of sphere diameters and gap spacings
Summary
Air is the most commonly used dielectric in electric power systems, and its strength is of vital importance for insulation design of many electrical equipment. Classical gas discharge theories including the Townsend theory [1] and the streamer theory [2], [3] are beneficial to explain air discharge phenomena theoretically Their interpretations of the discharge process have some differences and are applicable for different cases, but the common ground is that air gap breakdown results from out-of-limit of the electric field (EF) applied on the air dielectric. Wang: Feature Set for Structural Characterization of Sphere Gaps and the Breakdown Voltage Prediction systems and some key physical parameters are applicable to limited cases. It is still unclear whether these criteria are suitable for various gap arrangements [7]. This study is beneficial to provide reference for numerical simulation and intelligent prediction research of air gap discharge characteristics
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
