Multi-objective predictive control of cascaded H-bridge multilevel inverter based grid integrated PV based distributed generation system with improved power quality features

Rise in the distributed generation (DG) systems to cater to the continuously growing energy demand and provide a sustainable alternative to the conventional system of energy generation, presents additional challenges of grid synchronization, harmonic compensation, active and reactive power control, voltage transients, islanding etc. in the power generation system. This paper introduces a novel approach to controlling photovoltaic (PV) inverters through the use of model predictive control (MPC) as the main control strategy. The proposed model predictive current controller for grid-connected cascaded H-bridge multilevel inverters (CHBMLI) is designed to minimize the computational effort required to select the optimal switching vector. This approach improves active power flow control, harmonic mitigation, reactive power compensation, and capacitor voltage balancing of DC-link capacitors. The comprehensive control scheme for CHBMLI eliminates the need for additional active power filters (APFs), which are commonly employed to enhance power quality at the grid connection point in distributed generation (DG) systems. In this paper, two different DC-DC converters are utilized to track the DC link voltage, ensuring maximum power output from the PV system. The P&O optimization technique is employed to achieve this maximum power. In order to improve the output power quality, the load active (both fundamental and harmonic) and reactive power are estimated in every sampling time by sensing different electrical parameters of the load. Based on the harmonic and reactive power requirement of the load, the reference signal has been and hence the optimal switching pulses are generated through the MPC controller. With reference to this the load harmonic as well as reactive power requirement of the load have been supplied by the Inverter thereby grid side power quality gets improved. Furthermore, an optimal switching sequence is selected through the proposed controller, out of the available redundant voltage vector of the CHBMLI to control the input DC link voltage of individual H-Bridge. The proposed control scheme's efficacy has been evaluated through simulations performed in MATLAB. Additionally, its performance has been verified through practical experiments using a laboratory-created model of the Photovoltaic Distributed Generation (PVDG) system.