The Finite control set model predictive control (FCS-MPC) is recently introduced to control inverters without the modulation stage. The absence of the modulation stage gives an unpredictable performance of the control system. In this paper, the performance of FCS-MPC is assessed by comparing with PID control which is based on Scalar Pulse Width Modulation (PWM). The two control techniques are applied for load voltage regulation of the autonomous four-leg voltage source inverter (FLVSI). Practically, the predictive control requires a large number of calculations, resulting in high computation time and delay. In this paper, a new finite control set model predictive voltage control (MPVC) algorithm is proposed to predict the load voltages for 15 switching states instead of 16 switching states for reducing the computation time required for the control algorithm. Moreover, the algorithm is optimized by removing the repeated computations and the delay is compensated using the two-step prediction horizon principle. An accurate discrete-time state-space model of the autonomous FLVSI with output LC-filter is used for predicting the load voltages considering the neutral inductance and damping resistance of the LC filter. A simple PID control scheme with decoupled feedforward voltage and current loops is used in the DQ0 reference frame, while MPVC operates in the ABC reference frame. The simulation and experimental results are used to show the full assessment of the MPVC. The prominent outcomes show the ability of the proposed MPVC algorithm to provide high power quality under unbalanced and non-linear load conditions with high stability and robustness.
Among the variety of modern approaches to the mathematical description of the power quality indicators during the processes of transmission, distribution, conversion and calculation of the ac electric power, the representation of three-phase models in the form of a purely imaginary quaternion located in a separate subspace of the four-dimensional hypercomplex space allows, in relation to the generally accepted method of analyzing linear circuits, for example, symmetrical components with the selection of a direct, reverse and zero phase sequence for the fundamental harmonic, to take into a more complete account the features of energy consumption, especially in the presence of distortion in the modified forms of harmonic signals. In addition, the division of the quaternion into scalar (real) and partial (imaginary) makes it possible to significantly simplify the subsequent analytical processing of synthesis of a power converters control signals for active filtering and power supply of autonomous loads of an arbitrary type, including a single-phase configuration, by extracting from its composition individual components responsible for both the amplitude-phase asymmetry and the nonlinearity of the characteristics.
The main algorithmic principles of organizing control structures as part of three-phase systems of various functional purposes, as a rule, are based on the conversion of reference signals and current values of measured currents and voltages into state coordinates obtained by rotating the three-dimensional space plane by a given angle. At the same time, the calculated ratios for the numerical determination of the initial variables transformed by rotation in the quaternion basis are a function of only four kinematic parameters, which, other things being equal, leads to a simplification of the control law in relation to the traditional vector-matrix approach using nine direction cosines with six connection equations. In this regard, this paper is devoted to the applied problems of implementing linear transformations by E. Clarke and R.H. Park in terms of four-dimensional hypercomplex numbers, in compliance with the additional requirement of the invariance of scalar quantities after the transition.
In real inverters' operations, it is essential to insert delay time in the pulses provided to the inverter switches to protect the DC link against the short circuits. From this situation, the dead time phenomenon is introduced that causes undesirable performance and distortion of the output signal. Previously, researchers have proposed various schemes for compensating or eliminating dead-time. In this paper, a new dead-time elimination (DTE) scheme is proposed with a guarantee algorithm to eliminate dead-time and overcome the issues produced at the zero-currents-crossing point (ZCC). This method does not require additional hardware or filters to determine the polarity of the output current, and its principle is very simple to implement. The developed DTE method completely removes the dead-time issues on the magnitude and phase of the output voltage, and avoid the problems which can be induced around the ZCC. The results confirm the effectiveness and safety of this method.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.