CAJ | 학술논문

提出了一种基于PSO-RBF神经网络的电机轴承故障诊断方法.针对RBF神经网络泛化能力方面的不足,利用PSO算法对RBF神经网络的参数进行优化,然后采用优化后的PSO-RBF神经网络对轴承的故障形式进行诊断.结果表明, PSO-RBF神经网络的分类效果较好,在故障诊断领域有很好的应用价值.
제출료일충기우PSO-RBF신경망락적전궤축승고장진단방법.침대RBF신경망락범화능력방면적불족,이용PSO산법대RBF신경망락적삼수진행우화,연후채용우화후적PSO-RBF신경망락대축승적고장형식진행진단.결과표명, PSO-RBF신경망락적분류효과교호,재고장진단영역유흔호적응용개치.
A new method based on PSO-RBF neural network for fault diagnosis on the motor bearing is presented. PSO algo-rithm is used to optimize parameters of RBF neural network for improving performance of generalization ability.The bearing di-agnosis can be identified by using optimized PSO-RBF neural network automatically.The experiment results show that PSO-RBF neural network on classification is quite effective. That method has the good function in fault diagnosis techniques.