基于改进PSO-BP神经网络的变压器绕组热点温度预测研究

Prediction of Hot Spot Temperature of Transformer Winding Based on Improved PSO-BP Neural Network

文章提出了一种基于改进粒子群优化BP神经网络的变压器热点温度预测方法。改进后的变压器绕组热点温度预测方法是一种具有更强的收敛能力的粒子群优化算法。通过对变压器负载电流、顶层油温、底层油温以及环境温度等特征量的提取,同时将训练用试验样本数据对改进后的神经网络进行训练,将训练集数据与测试集数据按相应比例划分,从而达到对变压器绕组热点温度的精准预测。将预测结果与实测值进行对比,结果表明改进的PSO-BP神经网络算法的预测结果与实测值具有较好的一致性。

Prediction method of transformer hot spot temperature based on improved PSO-BP neural network is proposed.The improved method for hot spot temperature prediction of transformer winding is a particle swarm optimization algorithm with stronger convergence.Through the extraction of characteristic quantities such as transformer load current,top oil temperature,bottom oil temperature and ambient temperature,the improved neural network was trained with test sample data,and the data of training set and test set were divided in the corresponding proportion.Comparing the predicted results with the measured values,the results show that the predicted results of the improved PSO-BP neural network algorithm are in good agreement with the measured values.