Fault diagnosis based on wavelet package and Elman neural network for a hydraulic pump
-
摘要: 针对液压泵出口故障检测信号信噪比低、难以进行故障特征提取的特点,及 传统的BP网络进行故障诊断时网络学习具有收敛速度慢和学习、记忆不稳定的缺陷,提出了 一种将小波包变换和改进Elman神经网络相结合,进行液压泵故障诊断的新方法.利用具有 紧支结构的小波函数对信号进行分解,削减小波系数以滤除信号中的噪声;单支重构以有效 提取各频带的故障特征,并以频带能量作为识别故障的特征向量;应用改进的Elman神经网 络建立从特征向量到故障模式之间的映射,实现液压泵故障分类.试验结果表明,采用小波 包和改进Elman神经网络相结合的方法可有效的实现液压泵故障的诊断.Abstract: Considering the low signal-to-noise, faint failure characteristics of hydraulic pump, and slow convergence speed and the instability of BP network , a new fault diagnosis method based on the wavelet package analysis and Elman n eural network is presented. The wavelet package analysis is adopted to eliminate the noise in the actual signals and to extract the fault characteristics. Throu gh signal decomposition and single reconfiguration with wavelet package, the noi se can be eliminated from signals to strengthen the failure signal and to extrac t fault feature in every frequency bands effectively. Energy of various frequenc y bands acting as the fault feature vector is input into the improved Elman neural n etwork to realize the mapping between the feature vector and the fault mode. The experiment results verified the effectiveness of the proposed method in the hydraulic pump faul t diagnosis.
-
Key words:
- wavelet package analysis /
- Elman neural network /
- hydraulic pump /
- fault diagnosis
-
[1] Goumas S K, Zervakis M E. Classification of washing mac hines vibration signals using discrete wavelet analysis for feature extraction . IEEE Trans on Instrumentation and Measurement, 2002, 51(3):497-508 [2] 彭玉华. 小波变换与工程应用[M]. 北京:科学出版社,2002:116-117 Peng Yuhua. Wavelet transformation and engineering design[M]. Beijing:Sci ence Publishing House, 2002:116-117(in Chinese) [3] 胡昌华, 张军波. 基于MABLAB的系统分析与设计——小波分析[M]. 西安: 西安电子科技大学出版社, 2001:209-211 Hu Changhua, Zhang Junbo. System analysis and design based on Matlab—Wavele t analysis[M]. Xi’an:Xi’an Electronic Technology Publishing House, 2001:209 -211(in Chinese) [4] 王少萍. 液压泵计算机辅助测试及故障诊断研究 . 北京:北京航空航天大学自动控制系, 1994 Wang Shaoping. Computer aided test and fault diagnosis to hydraulic pump . Beijing:Dept of Automation Control, Beijing University of Aeronautics and Astro nautics, 1994(in Chinese) [5] 宋友,柳重堪,李其汉.基于小波包分解的早期碰摩故障诊断研究[J]. 北京航空航天大学学报,2003, 29(1):87-90 Song You, Liu Chongkan, Li Qihan. Early rub-impact fault diagnosis based on wavelet packet decomposition[J]. Journal of Beijing University of Aeronautics and Astronautics, 2003,29(1):87-90(in Chinese) [6] 翁海娜,房建成,杨功流,等.基于调频高斯小波变换的导航传感器故障诊断[J]. 北京航空航天大学学报,2002, 28(4):477-480 Weng Haina, Fang Jiancheng, Yang Gongliu, et al. Fault diagnosis for naviga tional sensors based on modulated gaussian wavelet transformation [J]. Journal of Beijing University of Aer
点击查看大图
计量
- 文章访问数: 3317
- HTML全文浏览量: 148
- PDF下载量: 970
- 被引次数: 0