The On-Load Tap Changer(OLTC) of transformer has a complex nonlinear relationship between the mechanical fault symptom and fault type,and the traditional BP neural network is used to diagnose with low accuracy,slowing convergence rate and easy to fall into local minimum value and so on.An Adaptive Genetic Algorithm (AGA) is proposed to optimize the BP neural network fault diagnosis.The weights and thresholds of BP neural networks based on adaptive genetic algorithm optimization,and then the optimized BP neural network is applied to the OLTC mechanical fault diagnosis.The simulation results show that the fault diagnosis model of BP neural network optimized by AGA algorithm is superior to the traditional BP neural network method.It can effectively improve the accuracy and speed of the mechanical fault diagnosis of OLTC.%由于变压器有载调压分接开关(On-Load Tap Changer,OLTC)机械故障征兆与机械故障类型之间有着复杂的非线性关系,采用传统的BP神经网络诊断具有准确率低、收敛速度慢和易陷入局部极小值等缺点,提出了一种自适应遗传算法(Adaptive Genetic Algorithm,AGA)优化BP神经网络的故障诊断方法.利用自适应遗传算法对BP神经网络的权值和阈值进行优化,将优化后的BP神经网络应用于OLTC机械故障诊断.仿真结果表明,AGA算法优化BP神经网络的故障诊断模型明显优于传统BP神经网络方法,有效地提高了OLTC的机械故障诊断精度和速度.
展开▼
