Global advanced control strategy for Modular Multilevel Converter integrated in a HVDC link. (4th January 2018)

• Record Type: Journal Article
• Authors: Ayari, Mohamed; Belhaouane, Mohamed Moez; Guillaud, Xavier; Braiek, Naceur Benhadj
• Is Part Of: International transactions on electrical energy systems. Volume 28:Number 4(2018)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2018-01-04
• Subjects: bilinear model -- LQR approach -- Modular Multilevel Converter (MMC) -- energies and DC voltage control -- polynomial control; Electric power -- Periodicals; Electric power systems -- Periodicals; Electrical engineering -- Periodicals; 621.3
• DOI: 10.1002/etep.2511
• Languages: English
• ISSNs: 2050-7038

Abstract

Summary: Modular Multilevel Converter (MMC) is becoming a promising converter technology for high‐voltage direct current transmission systems due to its high modularity, availability, and power quality. It is a multi‐input‐multi‐output nonlinear system. The control system for MMC is required to simultaneously achieve multiple control objectives. Existing control strategies for MMC are complex and the controller parameter design is not straightforward for the nonlinear systems with highly coupled states. In view of this, a steady‐state model for the MMC is developed on bilinear deviation state‐space model around a working point. Based on linear quadratic regulator and least squares methods, a nonlinear polynomial feedback law is designed to simultaneously control the grid and differential currents, and the global stored energy and energy balancing between total upper and lower arms. To generate the optimal current references, a multivariable linear quadratic controller is used to regulate the total energy per leg, energy difference between each upper and lower arms, and the DC bus voltage. The proposed high‐level controller depicts an original advanced control structure of MMC converter. The performance of the proposed strategy for a detailed model of 400‐level MMC is evaluated using simulations in MATLAB/SIMULINK/SPS software environment.