Fault Detection of Hydraulic Vane Pumps Using Multiple Variables in an Adaptive Neuro-Fuzzy Inference System
Health monitoring of hydraulic vane pumps is currently achieved through oil analysis. Monitoring pump operating parameters to detect faults on-line could improve condition based maintenance techniques. Hancock 1 developed a simple on-line hydraulic vane pump fault detection system. Two parameters, vertical and horizontal vibration signals, were decomposed using wavelet packet analysis. Packets containing signal features distinguishing normal and failed pump operation were entered into an adaptive neuro-fuzzy inference system (ANFIS) for pump health classification. The results showed that this technique has potential to reliably classify pump health.Here the fault detection system is expanded to use multiple pump parameters for fault detection. Fault detection rates were comparable to the original study, demonstrating that using one sensitive parameter for ANFIS decision making is as effective as developing a more complex structure with multiple pump parameter inputs.
Fault detection condition based monitoring hydraulic vane pump wavelet analysis adaptive neuro-fuzzy inference
K.M. Hancock Q. Zhang
Department of Agricultural and Biological Engineering, College of Agriculture, Consumer, and Environmental Sciences, University of Illinois Urbana-Champaign, Urbana, Illinois, USA
国际会议
北戴河
英文
234-236
2007-06-06（万方平台首次上网日期，不代表论文的发表时间）