Dynamic preventive maintenance strategy for MMC considering multi-term thermal cycles. (March 2020)

• Record Type: Journal Article
• Authors: Feng, Fei; Yu, Juan; Yang, Zhifang; Dai, Wei; Zhao, Xingpan; Kamel, Salah; Wang, Jianxiao; Fu, Guobin
• Is Part Of: International journal of electrical power & energy systems. Volume 116(2020)
• Publication Date: 2020-03
• Subjects: Dynamic preventive maintenance -- Modular multi-level converter -- Maintenance economy -- Multi-term thermal cycling -- Reliability; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2019.105560
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A dynamic PM strategy for MMC is proposed to reduce PM costs with acceptable reliability. This method optimizes varied redundant SMs and PM interval in each operational stage. The impact of operational conditions is considered based on multi-term thermal cycles MMC. Long-term variations of wind speed and air temperature create more damages in MMC. Abstract: With the integration of large-scale renewable energy, preventive maintenance (PM) of modular multi-level converter (MMC) has become a critical technique for guaranteeing the secure operation of long-distance transmission systems. However, existing PM strategies result in errors and extra costs because of ignoring the time-varying operational conditions (e.g. wind speed, air temperature and power current alternating) and constant redundant sub-modules (SM). To reduce PM-related costs with an acceptable level of reliability considering different operational conditions, a dynamical PM strategy is proposed. Compared with the traditional PM strategy, adjustable redundant SMs in every operational stage and PM interval are added to dynamically coordinate reliability and PM-related costs of MMC. The reliability model of MMC proposed in this paper considers the impact of different operational conditions based on multi-term thermal cycles of components and operating characteristics of MMC. Monte Carlo method is used to estimate the reliability of MMC, and the genetic algorithm is applied to optimize the PM interval and the number of redundant SM in every operational stage. Case studies show that the PM results vary with different operational conditions. Besides, compared with existing strategies, the proposed dynamic PM strategy could reduce the PM-related costs with an acceptable level of reliability by dynamically adjusting the PM interval and redundancy on every stage.