Worn piston identification method for an axial piston pump

The research article 'A new leaky piston identification method in an axial piston pump based on the extended Kalman filter' has been published in Elsevier journal Measurement.

Abstract

Volumetric losses are essential to ensure proper lubrication of moving components in an axial piston pump (APP). However, amplification of these volumetric losses can be observed when one or more pistons of the APP degrade due to friction and contaminated fluid. This amplification of volumetric losses due to a worn piston is often called a piston leak. The latter disturbs the output pressure signal and considerably reduces the efficiency of the pump. It also generates significant vibrations that can lead to the resonance of the pump structure. In this context, it is necessary to implement a diagnosis tool to identify the worn piston among the others. This will allow effective maintenance interventions by changing only the worn piston instead of all pistons.

This paper presents a new approach based on the physical model of the pump to identify the worn piston from the healthy ones. It begins by modelling the dynamic comportment of the pump in a nonlinear state model. Then, the extended Kalman filter (EKF) is adapted to estimate pressure in piston chambers. This estimation gives the possibility to observe the pressure into each piston chamber and then allows the identification of the worn piston. The proposed approach is validated on an APP test rig. The obtained results prove the efficiency of the proposed approach in identifying the worn piston.

Access the complete article on ScienceDirect.