机构地区: 北京信息科技大学自动化学院
出 处: 《北京信息科技大学学报(自然科学版)》 2017年第4期75-79,共5页
摘 要: 针对水泥回转窑支撑装置的故障问题,提出一种基于小波包神经网络的故障诊断方法。对于传统方法分析故障数据分辨率差等问题,通过更为精细的小波包信号分解方法,结合BP神经网络对分类问题的优异处理能力,将采集到的信号进行小波包分解与重构,提取到信号的特征向量,作为BP神经网络的训练样本和检测样本,对不同的故障信号进行分类输出。实验表明,该方法对回转窑支撑装置的故障具有良好的诊断效果,最终识别正确率达到92.5%。 A fault diagnosis method based on wavelet packet neural network is proposed to solve the fault problem of cement rotary kiln support device. To solve the problem of poor resolution of fault data in traditional method,through more sophisticated wavelet packet signal decomposition method,BP neural network is combined on the classification problem of excellent processing capacity. The collected signals are decomposed and reconstructed by wavelet packet,and the eigenvectors of the signal are extracted.As the training samples and detection samples of the BP neural network,the different fault signals are classified and output. The experimental results show that this method has a good diagnosis effect on the failure of the rotary kiln support device,and the final recognition rate is 92.5%.