Phd Ahmed ZAMA

“Modeling and Control of Modular Multilevel Converters (MMCs) for HVDC applications”

Common understanding today is that the challenges to develop the SuperGrid are huge at many levels (e.g.: political, societal, economical, financial, scientific, technical…). It is also commonly accepted that the SuperGrid will require novel technologies (breakers, cables, converters…) and operating principles (e.g.: transition from a more passive essentially AC-based power system to a more active AC/DC-based one). This thesis deals with the development of a key actor for SuperGrid technology, that is, AC/DC converters.

Recently, the Modular Multilevel Converter (MMC) has gained more and more importance for different applications, particularly for High Voltage Direct Current (HVDC). This topology, firstly proposed by Lesnicar and Marquardt in 2003, has many advantages compared to other Voltage Source Converter (VSC) topologies. As the name suggests, the topology is modular and easily scalable in terms of voltage levels: excellent AC output voltage and DC voltage can be easily increased by adding new cells. Moreover, thanks to the redundant cells, the MMC can also manage its internal degraded modes which are important technical and economic issues.

The development of a complex structure such as the MMC, requires a detailed analysis of its state variables and a depth understanding of its behavior when integrated. This necessitates developments of accurate models (detailed and/or simplified) and performant control systems. In 2014, when this work has started, many studies from academia as well as industry involving modeling and control of Modular Multilevel Converter have been carried out. In this thesis, these two aspects have been reviewed, analyzed and improved by proposing new models based on several implementation methods, in addition to develop innovative control algorithms in order to fulfill industrial requests by allowing fast MMC dynamics. Real time simulation as well as experimental tests have been carried out to validate our findings.

More informations