Predictive maintenance and fault diagnosis of hydraulic gear coupling of a boiler feed pump unit

Abstract

Increasing demand for reliability and performance of today’s most complex machines, and maintaining high productivity without sacrificing product quality have made it imperative for maintenance engineers to devise newer strategies in maintenance of plant and machinery. One of such strategies is condition monitoring, which has emerged as one of the most powerful technology in maintenance engineering. Condition monitoring and diagnostic engineering is a novel concept which enables us to detect in advance any incipient failure with ease and confidence in any part of a dynamic system before such failure trigger – off various types of failure mechanisms, which in turn may render the whole system uneconomical, unreliable, unhealthy, unsafe and even lethal. The present work highlights an experimental investigation to monitor the vibration behavior of hydraulic gear coupling, which is a part of Boiler Feed Pump train of a large utility Thermal Power Plant.  The coupling is supported by 4 Bearings. Tri-axial measurements are made at the bearing supports for 12 months. Displacement and Velocity are measured along Horizontal, Vertical and Axial directions. The experimental data is plotted on Time domain for graphical analysis to ease viewing of vibration signals. Based on the experimental data, faults are diagnosed using ISO – 2372 standards and causes are predicted. It is observed that the front and rear bearings of Input and output shafts of coupling are experiencing excess vibration. The work is concluded by suggesting remedial measures to ensure vibration intensity, at the said points, within the safe limits.