The work "Design and Selection of High Reliability Converters for Mission Critical Industrial Applications: A Rolling Mill Case Study" was accepted for publication in the IEEE Transactions on Industry Applications, March 2018. THE authors Victor Nazareth Ferreira and Allan Fagner Cupertino of Electrical Engineering at UFMG), Anderson Vagner

Rocha (CEFET-MG), Heverton Augusto Pereira (Federal University of Viçosa), Seleme Isaac Seleme Junior and Braz de Jesus Cardoso Filho (PPGEE-UFMG) received the confirmation of acceptance on March 30, 2018.In this work the authors call attention to important aspects in the design and selection of large power converters for mission-critical industrial applications. In this type of application, the cost of a single drive fault can exceed the total installation costs of the entire system. The work focuses on applications of the metal industry, particularly steel mills which, in the metallurgical industry, is a forming process widely used to enable high production and control of the final product. In large mills, important requirements such as rapid dynamic response and high overload capacity are achieved thanks to the high power of medium voltage converters. Among the available converter types, the three-level neutral point clamped converter (3L-NPC) is currently the most widely used. Commercial versions of this converter do not traditionally employ any fault tolerance strategy for their internal devices, so in case of failure event, the lack of proper redundancy strategy takes the entire system early to an unscheduled shutdown. One possible reason for this scenario is the lack of a methodology that allows for solid decision making in the design and selection phases of the application. To fill in part of this gap, the authors present a simple selection criterion to be applied by project or field engineers. Two converters are compared to the standard solution: the fault-tolerant version of the active NPC converter (previously proposed by TESLA laboratory researchers) and the Triple-Star Bridge Cells Modular Multilevel Converter. A case study is conducted with data from a large steel industry in south-east Brazil.