Modulated Model Predictive Control of Multisource Inverters With Flexible Power Distribution
Dehong Zhou, Zhao Zhang, Zewei Shen, Jianxiao Zou
Abstract
The multisource inverter (MSI) is a high-efficiency topology for hybrid electric vehicles due to its advanced merits of single-stage power conversion without employing dc–dc converters. However, the control scheme for the MSI is challenging because of its asymmetrically distributed voltage vectors and extra control objective of power distribution between the energy sources. In view of this, this article proposed a modulated model predictive control (M <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> PC) to realize flux, torque, and power distribution control of MSI simultaneously. In the proposed M <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> PC, the MSI is considered as a whole to reduce the evaluation options and the computational burden of the model predictive control. The optimal voltage vector is calculated by the M <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> PC and decomposed into the decoupled subinverters to achieve the desired power distribution. Finally, the decomposed vectors are fed into the modulation stage directly in the subinverters for gating signal generation. With the proposed M <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> PC, motor drive control and power distribution are easily realized simultaneously without a heavy computation burden under the asymmetric vector diagram. Besides, a one-switching-time vector is selected in one subinverter to enhance the power conversion efficiency of the MSI. Experimental results verify the effectiveness and performance of the proposed method.