Knowledge-Based Approaches to Fault Diagnosis: A Practical Evaluation of the Relative Merits of Deep and Shallow Knowledge

Abstract

This paper argues that knowledge-based systems are an appropriate tool for the diagnosis of faults in complex devices and that both deep and shallow knowledge have their part to play in this process. The successful implementation and evaluation of the two diagnostic knowledge-based systems, named DIPLOMA and MIDAS, have shown that such systems are an appropriate tool for the diagnosis of faults in complex hydromechanical devices and that they make a beneficial contribution to the business performance of the host organization. Furthermore, it has been demonstrated that the most effective and comprehensive knowledge-based approach to fault diagnosis is one that incorporates both deep and shallow knowledge, so that the distinctive advantages of each can be realized in a single application.

Model-based reasoning for fault diagnosis of twin-spool turbofans

“…The governing reasons are that this approach consists of the simulation of the specific real-world system and then draws a conclusion about the system by reasoning from first principles. Moreover, the deep-knowledge approach is always based upon the structural and functional knowledge of the device under diagnosis [3]. For the case of a gas turbine model, the interconnecting component parameters are non-linearly related to each other in such a way that the structural and functional knowledge of the aero-engine is explicit, which is then arranged in such a way that the computational process becomes much easier vis-a Á-vis the various structural changes, leading to the formation of the aero-thermodynamic model.…”

Degradation analysis of grinding machine spindle systems based on complexity

“…6,7 Spindle systems are one of the most important components in machine tool, which consist of the spindle, bearings and spindle box. The main function of spindle systems is to transmit the motions and forces during manufacturing.…”

“…The condition detection of the critical components or systems of the devices has attracted extensive attention in the academic research and practical engineering in recent decades and has made great achievements, such as the condition detection of the shaft and bearings of device, 1,2 the steam turbine, 3 the coolant system of device 4,5 and hydraulic system of device. 6,7 Spindle systems are one of the most important components in machine tool, which consist of the spindle, bearings and spindle box. The main function of spindle systems is to transmit the motions and forces during manufacturing.…”