Improvement of Unit Maintenance Scheduling for Power Producers

Abstract

This paper presents a novel unit maintenance scheduling (UMS) model for a power producer, to maximize its potential benefits whilst considering the influence of maintenance outage on generating units. Firstly, the Weibull distribution is applied to simulate the unexpected unit failure probability for long-time running. Its corresponding parameters can be estimated from historical data by solving a least-squares-fitting problem via Gauss-Newton iteration method. Based on the failure analysis, two economic indexes are introduced to assess the impacts of UMS on generating units, which represent the expected revenue loss for maintenance outage and expected risk for continuous operating respectively. Then, the producer could schedule its outage plan following a certain sequence, determined by the expected unit economic loss. Compared with current models, the proposed model gives not only the optimal outage period but also an acceptable maintenance interval of each unit. Finally, numerical examples on a five-unit producer are utilized to demonstrate the usefulness of the proposed scheme.