Aircraft air conditioning system health state estimation and prediction for predictive maintenance

• DOI: 10.1016/j.cja.2019.03.039
• Publication Date: 2019-05-30T03:24:26Z
• Authors: Jianzhong SUN; Fangyuan WANG; Shungang NING
• Keywords: Prognostics; Condition Monitoring; Condition-Based Maintenance; Airplane

The vast potential of system health monitoring and condition based maintenance on modern commercial aircraft is being realized through the innovative use Airplane Condition Monitoring System (ACMS) data. However there are few methods addressing issues failure prognostics predictive for Air Conditioning (ACS). This study developed a Bayesian approach using ACMS data ACS. First, index characterizing ACS state inferred from multiple sensor signals driven method. Then dynamic linear model proposed to describe degradation process prognostics. inference formulas carried out estimation prediction. applied passenger fleet with recorded one year. analysis case shows that method can produce satisfactory results, where all precursors identified in advance, relative errors time prediction made when just entering warning stage less than 8%. would allow operators proactively plan future maintenance.