GRID-INTERACTIVE DFIG-BASED WIND ENERGY SYSTEMS FOR CONTROLLED POWER FLOW MANAGEMENT

Abstract

The future power grids will thus be multi-converter and multi-source systems, particularly with the massive insertion of RESs and storage devices interfaced with power-electronic converters, and the current trend favouring electricity as energy vector. This concerns especially MGs, building or eco-district networks, and future smart grids (at distribution level and even at transportation level). Electric power systems, integrating MGs, must therefore be partially or completely re-designed in order to integrate these new intermittent sources that could negatively affect quality of power supplied due to their very low short-circuit power. Moreover, the development of power-electronic interfaces can potentially induce unstable operation due to dynamic and harmonic interactions generated by the converters. Research problems associated with MGs concern consequently the development of optimal control strategies allowing maximizing power quality, securing supply, managing intermittence and its impact on the system by using for instance storage devices, maximizing efficiency, performance, reliability, and costs associated with the system. The main difficulties result from the necessity to understand interactions and couplings between components within these systems, and then establish static, dynamic, and harmonic models that allow taking into account the incense of weather conditions or load control strategies. In autonomous MGs, the stability and control issues are among the main challenges due to low inertia, uncertainties, and intermittent nature of DERs. Maintaining frequency and voltage deviations within their predefined range is still critical in autonomous operation mode. High-frequency ESSs for which more complex advanced control structures are needed for providing the MGs with desirable performance specifications in a wide range of operating conditions despite multiple constraints such as unexpected disturbances and model uncertainties.