摘要
分形法诊断柴油机故障时,实际测取的柴油机振动信号中存在着许多干扰信息,利用改进的小波包分解和重构算法对原始信号进行降噪处理,再选择合适的嵌入维数,计算出的关联维数即能够反映系统的真实特性。实例分析表明,该算法准确可靠,可极大提高利用信号关联维数判断柴油机运行状态的准确性和可靠性。
In the fractal diagnosis of diesel fault, many noises exist in the measured vibration signals. The ameliorated wavelet packet analysis and reconstruction algorithm can de-noise these original signals. Then with proper embedded dimensions, the consequent correlation dimension can reflect the real condition of the system. As case study has shown its reliability, the algorithm therefore can greatly improve the accuracy and reliability of diagnosing the running state of diesels with signal correlation dimension.
出处
《农机化研究》
北大核心
2007年第11期170-172,共3页
Journal of Agricultural Mechanization Research
关键词
自动控制技术
故障诊断
试验
柴油机
振动信号
关联维数
小波包变换
auto-control technology
fault diagnosis
experiment
diesel engines
vibration signal
correlation dimension
wavelet packet transform
