论文部分内容阅读
为解决目前电动机故障诊断方法收敛慢、易陷入局部极小、准确性不足等缺点,提出了一种电动机故障诊断新方法,即将Levy Flight随机游走机制、人工蜂群算法和径向基神经网络结合的方法。首先通过改进的蜂群算法即随机游走蜂群算法(LFABC)对网络参数进行全局寻优,然后对网络进行监督训练,局部细化网络参数,最终得到故障诊断模型。经实例验证,本文所提方法能更快速、准确地实现电动机故障诊断,取得了较好的效果。
In order to solve the shortcomings of current motor fault diagnosis methods such as slow convergence, easy falling into local minima and lack of accuracy, a new method of motor fault diagnosis is proposed, which combines the Levy Flight random walk mechanism, artificial bee colony algorithm and radial basis neural network A combination of methods. Firstly, the improved bee colony algorithm called random walked bee colony algorithm (LFABC) is used to globally search the network parameters. Then the network is supervised and trained, and the network parameters are partially refined. Finally, the fault diagnosis model is obtained. The example validation, the method proposed in this article can be more quickly and accurately to achieve motor fault diagnosis, and achieved good results.