Multivector Model Predictive Power Control for Grid Connected Converters in Renewable Power Plants

Model-based predictive power control (MPPC) is a well-known and useful technique for control of electric drives and renewable energy generation systems. However, this strategy relies on the knowledge of accurate system models and parameter values, and would otherwise lead to tracking errors in applications because of inevitable parameter uncertainties. Also, step delays in MPPC implementations have to be compensated to eliminate errors. This article proposes a predictive control application to control active and reactive powers exchanged between the grid and the grid side converter (GSC) interfacing renewable energy sources such as wind farms or photovoltaic. The proposed MPPC minimizes the power tracking error based on a proposed cost function and the control system output is the voltage reference for the electronic converter that is converted into switching pulses by the modulation stage. The proposed control strategy is designed to eliminate tracking errors and has fixed switching frequency while featuring a fast-dynamic response, low current THD, and low computational burden. The method has been evaluated using MATLAB/Simulink environment and an experimental 5-kW grid-connected voltage source converter. Finally, the proposed MPPC method has been critically compared to the previous MPPC approaches.

Automated summary

This article proposes a model-based predictive power control (MPPC) method for controlling the power exchange between the grid and the converter in renewable energy generation systems. The method features fast dynamic response, low current total harmonic distortion, and low computational burden.