Improved direct model predictive control for variable magnitude variable frequency wave energy converter connected to constant power load

Archimedes wave swing (AWS) is the first direct-drive wave energy converter (WEC). The AWS is coupled to a linear permanent magnet synchronous generator (LPMSG) and generates a variable magnitude, variable frequency voltage. This paper presents an improved version of direct model predictive control (DMPC) which called improved DMPC (I-DMPC) for a hybrid system containing WEC, supercapacitor (SC) energy storage system (ESS), and constant power load (CPL). This method is implemented based on discrete incremental model of generator and modified objective function, therefor steady-state error is reduced. Simulation results in the MATLAB/SIMULINK environment show the proposed control can maintain stability of the system under different CPL demand and systems faults and improve robustness and fault-tolerant capability of the system in regular and irregular waves. The Comparing results with the conventional DMPC indicate, that by using I-DMPC, d-axis and q-axis errors under regular wave can be reduced 37.5% and 50%, respectively. Also, under irregular wave, I-DMPC reduce the peak of q-axis errors from 73A to 1.5 A when loss of one leg occurred.