基于改进BP神经网络的矿用通风机故障诊断

Fault diagnosis of mine ventilator based on improved BP neural network

  • 摘要: 针对矿用通风机故障与征兆对应关系复杂的特点,提出一种用动态适应布谷鸟搜索算法优化BP神经网络并进行故障诊断的方法。利用动态适应布谷鸟搜索算法的全局搜索能力,求解神经网络的最优初始参数;然后对BP神经网络进行学习训练,得到最终的故障诊断模型。实例分析结果表明,该方法能有效地进行矿用通风机故障诊断,且具有收敛速度快、精度高的特点,对测试样本的诊断准确率达到了92.5%。

     

    Abstract: In view of characteristics of complicated correlation of mine ventilator failure and symptom, a fault diagnosis method using BP neural network optimized by dynamic adaptation cuckoo search algorithm was proposed. The optimal initial parameters of neural network are solved by using global search ability of dynamic adaptation cuckoo search algorithm. Then, the BP neural network is trained to obtain the final fault diagnosis model. The example analysis results show that the method can effectively achieve fault diagnosis of mine ventilator and has the characteristics of fast convergence and high precision, and the diagnosis accuracy of the test sample is 92.5%.

     

/

返回文章
返回