内燃机变分模态Rihaczek谱纹理特征识别诊断
发布时间：2018-10-10 14:55
【摘要】：针对内燃机故障诊断中振动响应信号强耦合、弱故障特征的问题,提出一种基于内燃机振动谱图纹理特征提取的故障诊断方法。首先,为了清晰地刻画内燃机振动信号时频联合分布中的非平稳时变分量,将变分模态分解(VMD)与Rihaczek复能量密度分布方法有效结合,得到了时频聚集性好、无交叉项干扰的内燃机振动谱图像;针对VMD分解过程中的参数选取问题,提出将功率谱熵作为目标函数,对VMD的分解参数进行网格寻优,提高了VMD分解的自适应性。为了实现对内燃机振动谱图像的自动识别及故障诊断,提出了改进的局部二值模式(ILBP)方法,用来对振动谱图中蕴含的纹理信息进行分析,提取低维特征参量并采用最近邻分类器对内燃机不同工况的振动谱图像进行模式识别。将该方法应用于内燃机故障诊断实例中,结果表明该方法能有效提取内燃机振动信号中的微弱故障特征,实现内燃机故障的自动诊断。
[Abstract]:Aiming at the problem of strong coupling and weak fault characteristics of vibration response signals in internal combustion engine fault diagnosis, a fault diagnosis method based on texture feature extraction of internal combustion engine vibration spectrum is proposed. Firstly, in order to describe clearly the nonstationary time-varying components in the time-frequency joint distribution of internal combustion engine vibration signals, the variational mode decomposition (VMD) method and the Rihaczek complex energy density distribution method are effectively combined to obtain good time-frequency aggregation. In view of the problem of parameter selection in the process of VMD decomposition, the power spectrum entropy is used as the objective function to optimize the decomposition parameters of VMD in order to improve the self-adaptability of VMD decomposition. In order to realize the automatic identification and fault diagnosis of internal combustion engine vibration spectrum image, an improved local binary mode (ILBP) method is proposed to analyze the texture information contained in the vibration spectrum image. The low dimensional characteristic parameters are extracted and the nearest neighbor classifier is used to recognize the vibration spectrum images of internal combustion engines under different working conditions. The method is applied to the fault diagnosis of internal combustion engine. The results show that the method can effectively extract the weak fault characteristics from the vibration signal of internal combustion engine and realize the automatic fault diagnosis of internal combustion engine.
【作者单位】： 火箭军工程大学理学院;
【基金】：国家自然科学基金(51405498) 中国博士后基金(2015M582642)项目资助
【分类号】：TK407;TP391.41
本文编号：2262205
[Abstract]:Aiming at the problem of strong coupling and weak fault characteristics of vibration response signals in internal combustion engine fault diagnosis, a fault diagnosis method based on texture feature extraction of internal combustion engine vibration spectrum is proposed. Firstly, in order to describe clearly the nonstationary time-varying components in the time-frequency joint distribution of internal combustion engine vibration signals, the variational mode decomposition (VMD) method and the Rihaczek complex energy density distribution method are effectively combined to obtain good time-frequency aggregation. In view of the problem of parameter selection in the process of VMD decomposition, the power spectrum entropy is used as the objective function to optimize the decomposition parameters of VMD in order to improve the self-adaptability of VMD decomposition. In order to realize the automatic identification and fault diagnosis of internal combustion engine vibration spectrum image, an improved local binary mode (ILBP) method is proposed to analyze the texture information contained in the vibration spectrum image. The low dimensional characteristic parameters are extracted and the nearest neighbor classifier is used to recognize the vibration spectrum images of internal combustion engines under different working conditions. The method is applied to the fault diagnosis of internal combustion engine. The results show that the method can effectively extract the weak fault characteristics from the vibration signal of internal combustion engine and realize the automatic fault diagnosis of internal combustion engine.
【作者单位】： 火箭军工程大学理学院;
【基金】：国家自然科学基金(51405498) 中国博士后基金(2015M582642)项目资助
【分类号】：TK407;TP391.41
【相似文献】
相关期刊论文 前1条
1 刘钢铸;刘成;;百年内燃机的第三次革命——运用油氧共混技术是提高内燃机燃烧效率、减少有害尾气排放和综合治理雾霾污染的有效途径[J];科技创业月刊;2014年05期
，本文编号：2262205
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2262205.html
