Efficient model predictive control with optimal duty cycle for power converters

Abstract

Conventional Finite Control Set Model Predictive Power Control (FCS-MPPC) has been in constant development, especially in the two-level AC/DC converter, because it has many advantages such as good steady state performance, quick dynamic response and flexibility in the definition of control objectives. However, single voltage vector applied in one control period is hard to reduce the power ripples to a minimal value. In order to resolve this issues, the concept of duty cycle control was introduced in FCS-MPPC, which helps to achieve further power ripple reduction. Although better steady state performance is acquired, the control complexity and computational burden are increased when selecting the best voltage vector. In view of this, this paper presents a efficient MPPC(E-MPPC) with optimal duty cycle. The method called E-MPPC only requires one prediction to find the best voltage vector and its duration is determined base on the principle of error minimization of converter voltage vector. Different from prior FCS-MPPC, the computation load of E-MPPC with optimal duty cycle is greatly reduced while the control performance is not affected. Furthermore, the relationship between E-MPPC and FCS-MPPC with duty cycle control is studied and it is found that in some cases, the both methods are completely equivalent. Experimental results on two-level AC/DC converter are presented to confirm good performance and application of the proposed method.