A data-driven early micro-leakage detection and localization approach of hydraulic systems

Leakage is one of the most important reasons for failure hydraulic systems. The accurate positioning leakage great significance to ensure safe and reliable operation For early stage leakage, pressure circuit does not change obviously therefore cannot be monitored by sensors. Meanwhile, changes frequently due influence load state switch, which further reduces accuracy localization. In work, a novel Bayesian networks (BNs)-based data-driven localization approach multi-valve systems proposed. Wavelet transform used signal noise reduction BNs-based leak model identify location leakage. A normalization developed improve robustness model. system with eight valves demonstrate application proposed micro-leakage detection approach.