Detection and Isolation of Leakage and Valve Faults in Hydraulic Systems in Varying Loading Conditions, Part 1: Global Sensitivity Analysis

Abstract

Model-based condition monitoring methods are widely used in condition monitoring. They usually rely on ad hoc approaches to verify the system model and then best practices are reported to detect the given set of faults. This first part of a two-piece paper introduces a generic Global Sensitivity Analysis-based approach that can be applied systematically to verify the model parameter sensitivities used for the model-based fault detection. The case study is a generic servo valve-controlled hydraulic cylinder with unknown loading condition which is then systematically analyzed with Global Sensitivity Analysis. The method shows valuable insight into systematic model verification and resulting fault detection in terms of showing the dominant sensitivity of the nominal flow rate and nominal pressure difference, and the exact sensitivities of 0-1 dm3/min external and internal leakages on cylinder chamber pressures and velocity. In the second paper, an Unscented Kalman Filter-based Fault Detection and Isolation scheme for leakage and valve faults of a generic servo valve-controlled hydraulic cylinder is devised and fault patterns are presented.