Modeling study demonstrates optimized maintenance timing and risk thresholds for industrial air compressors, indicating improved cost efficiency over fixed-interval schedules.
Key Points
To develop a quantitative condition-based maintenance decision framework that eliminates reliance on subjective threshold settings and fixed-interval schedules for rotating machinery.
Constructed a baseline life model using historical failure data via the Weibull proportional hazards model, incorporating high- and low-pressure discharge temperatures as time-dependent condition covariates.
Optimized maintenance scheduling to minimize unit time maintenance costs subject to operational availability constraints, establishing dual upper and lower decision thresholds.
Formulated an early warning mechanism based on the rate of change in the decision index to detect rapid mechanical degradation.
Balanced long-term maintenance cost minimization with short-term sensitivity to rapid degradation anomalies.
Established quantitative, data-driven condition thresholds that successfully replace experience-dependent fixed-interval maintenance strategies in gas transmission station air compressors.