【摘要】 针对变压器故障诊断精确度低的问题，本文提出了一种多策略改进的鲸鱼优化算法(MIWOA)优化随机配置网络(SCN)的变压器故障诊断模型。首先，对变压器冗杂繁多的原始故障数据进行核主成分分析(KPCA)降维处理，降低无效特征的影响；其次，利用Tent混沌映射、动态自适应权重和初级知识获取共享算法对鲸鱼算法(WOA)进行改进，提高其优化能力；然后，在SCN中引入L2范数惩罚项进行正则化处理，并使用改进后的MIWOA算法对SCN惩罚项系数C进行寻优求解，提高SCN分类精度和泛化能力；最后，将降维的数据输入到MIWOA-SCN故障诊断模型中，提高模型收敛速度。结果表明，本文所提出的模型诊断精度为93.1%,与WOA-SCN、GWO-SCN和PSO-SCN诊断模型相比，分别提高了6.89%、9.48%、14.65%,证明MIWOA-SCN诊断模型在变压器故障诊断上具有良好的诊断效果。更多还原

【Abstract】 To address the problem of low accuracy of transformer fault diagnosis, a multi-strategy improved whale optimization algorithm(MIWOA) is proposed to optimize the transformer fault diagnosis model of stochastic configuration network(SCN). First, the raw transformer redundant and extensive fault data are subjected to kernel principal component analysis(KPCA) to reduce the influence of invalid features. Secondly, the whale optimization algorithm(WOA) is improved by using tent chaos mapping, dynamic adaptive weighting and primary knowledge acquisition sharing algorithm to improve its optimization capability. Then, the L2 parametric penalty term is introduced in the SCN for regularization and the improved MIWOA algorithm solves the SCN penalty term coefficients C in an optimal way to improve the SCN classification accuracy and generalization ability. Finally, in order to accelerate the convergence speed of the model, degraded data are input into the MIWOA-SCN fault diagnosis model. The results show that the diagnostic accuracy of the model is 93.1%, which is 6.89% and 9.48% higher than the WOA-SCN, GWO-SCN, and PSO-SCN diagnostic models, respectively. This is 14.65% higher. This proves that the MIWOA-SCN diagnostic model has good diagnostic performance for transformer fault diagnosis.更多还原