Intelligent Prediction of Transformer Loss for Low Voltage Recovery in Distribution Network with Unbalanced Load

In order to solve the problem of low voltage caused by unbalanced load in distribution network, a transformer loss intelligent prediction model under is proposed. Firstly, mathematical with an established. The zero-sequence impedance and neutral line current are calculated using Chaos Game Optimization algorithm (CGO), correctness proved actual data. Then, correlation among network input variables eliminated Principal Component Analysis (PCA), so number decreased. At same time, Sparrow Search Algorithm (SSA) used optimize initial weight threshold BP accurate based on PCA-SSA-BP Finally, compared Genetic optimized (GA-BP), Particle Swarm (PSO-BP) (SSA-BP), has been proven be data it helpful for low-voltage recovery network.