Improved Continuous Control Set Model Predictive Control for Three-Phase CVCF Inverters: Fuzzy Logic Approach

Abstract

This paper presents an improved continuous control set model predictive control (CCS-MPC) scheme using a fuzzy logic approach (FLA) for three-phase constant voltage constant frequency (CVCF) inverters with an output LC filter. The proposed method can enhance the output voltage quality by using an optimal predictive control law (OPCL). Additionally, the proposed method corrects the input voltage commands through the FLA-based online gain tuning aiming to satisfy the hardware constraint. Thus, the proposed FLA-based CCS-MPC can guarantee an acceptable output voltage with a faster transient response, smaller steady-state error (SSE), and lower total harmonic distortion (THD). Moreover, a Kalman filter-based load current observer is used to significantly improve the reliability and robustness. Comparative experimental studies with another CCS-MPC method are conducted on a prototype three-phase CVCF inverter using TI TMS320F28335 DSP to validate the improved performance under the step-changes of linear/unbalanced/nonlinear loads with uncertain parameters. INDEX TERMSContinuous control set model predictive control (CCS-MPC), fuzzy logic approach (FLA), optimal predictive control law (OPCL), three-phase constant voltage constant frequency (CVCF) inverter.

A Novel Predictive Control Method with Optimal Switching Sequence and Filter Resonance Suppression for Two-Stage Matrix Converter

“…It should be stated that the CMPC may obtain a similar performance with the VMMPC in aspects of control objectives, by using a much higher sampling frequency than that in the VMMPC [18,22]. However, since predictive controllers are commonly demanding in terms of computational resources, it is not always possible to increase the sampling frequency in practical implementations.…”

“…The proposed VMMPC strategy employs two valid adjacent vectors 1 , 2 and one zero vector 0 or two zero vectors (if necessary). It should be noticed that the sampling frequency is much higher than the fundamental frequency, and thus, the control objective error can be approximated by using a time-weighted value [22,23]. It should be noticed that the sampling frequency is much higher than the fundamental frequency, and thus, the control objective error e i can be approximated by using a timeweighted value d ij e ij [22,23].…”

“…Owing to this weakness, variable switching frequency and a broad harmonic spectrum are produced, which lead to unexpected ripples in the system and affect the system's power-quality performance [18]. To overcome these problems, the modulated model predictive control (MMPC) strategy is introduced [19][20][21][22][23], where the switching frequency is kept constant and more than one switching states are determined and applied by the two independent cost functions of the rectifier and the inverter in an appropriately fixed switching instant. In [19], the MMPC strategy was proposed for shunt active filters, where a cost function ratio for output vectors-based modulation algorithm was included in the MPC.…”

“…In [21], the MMPC was proposed and applied for the modular multilevel converters in voltage source converter-high voltage dc systems, where modulated switching signals with a fixed switching frequency were applied to minimize the ac line current ripples. In [22], a continuous control set model predictive control (CCS-MPC) scheme was proposed for the constant voltage constant frequency inverters, in which a fuzzy logic approach corrected the input voltage commands to obtain a faster transient response and smaller steady-state error. In [23], a comparative study of the FCS-MPC and the CCS-MPC for induction machine in terms of their design and performance was conducted, in which the behaviors were assessed by applying reference and disturbance steps to the system in different operational modes.…”

Enhanced Fault-Tolerant Robust Deadbeat Predictive Control for Nine-Level ANPC-Based Converter

“…Continuous control set MPC (CCS-MPC) and deadbeat MPC (DB-MPC) are two different types of MPC that offer a constant switching frequency and reduced steadystate error [14], [15]. The first method requires complex mathematical formulations and a digital platform with high computing power, and in some cases solving the control problem offline is inevitable.…”