基于膜聚类算法的风机振动故障诊断

在工业环境下,风机振动故障常常需要人工诊断,诊断效率低,不易完成实时计算和在线分析判断。针对上述问题,提出了一种膜聚类算法可用于风机振动故障的在线智能诊断。该算法将膜计算的方法引入到聚类中,并采用概率模型更新种群的方法实现最佳聚类中心的寻优。算法首先在多个数据集上进行聚类实验,实验结果显示该算法克服了常规聚类算法聚类结果不稳定,聚类质量差的缺点。然后将其应用于风机振动故障在线诊断系统中进行仿真测试,结果显示所采用的方法能满足风机振动故障在线智能诊断要求,也可应用于其他各类设备的振动故障在线智能诊断。

Fault Diagnosis of Fan Vibration Based on Membrane Clustering Algorithm

In industrial environment, fan vibration fault often needs manual diagnosis, which is inefficient and difficult to complete real-time calculation and online analysis and judgment. To solve the above problems, a membrane clustering algorithm is proposed in this paper, which can be used for on-line intelligent diagnosis of fan vibration faults. The algorithm introduces the membrane computing method into clustering, and uses the probability model to update the population method to optimize the best clustering center. The algorithm first carries out clustering experiments on multiple data sets, and the experimental results show that the algorithm overcomes the shortcomings of the unstable clustering results and poor quality of the clustering algorithm. Then it is applied to the on-line diagnosis system of fan vibration fault. The results show that the method can meet the on-line intelligent diagnosis requirement of fan vibration fault and can also be applied to the on-line intelligent diagnosis of vibration fault of other kinds of equipment.