基于FA-BP神经网络模型的烘丝机设备故障诊断

Fault Diagnosis of Silk Drying Machine Based on FA-BP Neural Network Model

烘丝机作为烟丝生产的重要设备,其健康状态直接影响着烟丝的品质及生产效率。为了较早诊断出烘丝机故障,降低因故障带来的损失,提出了一种基于萤火虫算法(FA)优化BP神经网络的烘丝机故障诊断算法。首先,分析烘丝机的故障特征;其次,利用萤火虫算法寻优特性找到BP神经网络的最优权值和阈值,使故障诊断模型效果达到最佳状态。通过与SVM和BP神经网络模型进行对比分析,结果表明,使用FA-BP神经网络模型的烘丝机故障诊断准确率高达94.5%,诊断效果优于所对比的模型。

As an important equipment of tobacco production,the health status of silk drying machine directly affects the quality and production efficiency of tobacco.In order to diagnose the fault of the silk drying machine early and reduce the loss caused by the fault,a fault diagnosis algorithm of the silk drying machine based on BP neural network optimized byfirefly algorithm(FA)was proposed.Firstly,the fault characteristics of the silk drying machine were analyzed.Secondly,the optimal weighting factor and bias factor of BP neural network were found by using the optimization characteristics of thefirefly algorithm,so that the effect of the fault diagnosis model reached the best state.Compared with SVM and BP neural network model,the results show that the fault diagnosis accuracy of the silk drying machine using FA-BP neural network model is as high as 94.5%.The diagnosis effect is better than the compared models.