A framework for modeling fault propagation paths in air turbine starter based on Bayesian network

Abstract

Any minor fault may spread, accumulate and enlarge through the causal link of fault in a closed-loop complex system of civil aircraft, and eventually result in unplanned downtime. In this paper, the fault propagation path model (FPPM) is proposed for system-level decomposition and simplifying the process of fault propagation analysis by combining the improved ant colony optimization algorithm (I-ACO) with the Bayesian network (BN). In FPPM, the modeling of the fault propagation path consists of three models, namely shrinking model (SM), ant colony optimization model (ACOM), and assessment model (AM). Firstly, the state space is shrunk by the most weight supported tree algorithm (MWST) at the initial establishment stage of BN. Next, I-ACO is designed to improve the structure of BN in order to study the fault propagation path accurately. Finally, the experiment is conducted from two different perspectives for the rationality of the well-trained BN’s structure. An example of practical application for the propagation path model of typical faults on the A320 air turbine starter is given to verify the validity and feasibility of the proposed method.

Key fault propagation path identification of CNC machine tools based on maximum occurrence probability

“…The algorithm idea is that if there is a link from page A to B, then page A assigns a PR value to B, and page B accepts the PR value from A. The calculation process is iterated based on equation (7).…”

“…Accurate and reasonable fault propagation analysis can provide an important foundation for identifying key fault propagation paths in the system, thus can provide theoretical support for maintenance strategies and fault analysis [3]. The current methods for fault propagation analysis include Petri net method [4], cellular automata [5], topological network models based on complex network theory [6,7],…”

“…Cellular automata have been widely used in the fields of power grids, complex circuits, information propagation, but its considerations of changes in system structure and basic components are not comprehensive [5]. For complex network theory methods, the small world network model is the most classic, which is widely used due to it close to real networks most [6,7]. Although this method compensates for the shortcomings of traditional "experiential knowledge based" modeling, it fails to fully consider the functional coupling relationship between components within the system, resulting in biased analysis results.…”