Power quality in electrical systems is a basic requirement that should be guaranteed. Nonetheless, the increase of power-electronics-based applications and nonlinear loads have induced harmonic distortion leading to a degradation of the power quality; therefore, effective methods to measure and suppress such distortion need to be developed. The main contributions of this paper are: the development of a novel fast-converging optimal estimator to determine on-line the harmonic components of a time-varying signal, which can be used for analysis and control purposes; the usage of the estimator to generate reference signals for an optimal control system in order to suppress the harmonics components; and the experimental validation of the proposed methodology as a feasible solution for industrial applications. The proposal considers practical assumptions such as the signal to be analyzed is corrupted with noise and contains time-varying harmonics (no a steady-state condition is assumed). As an application of the harmonic estimator, two experimental tests are carried out: firstly, the optimal determination of the harmonic components of the current waveform in a nonlinear load, and secondly, the usage of harmonics information to mitigate them through of an active power filter, which is controlled by using an optimal trajectory tracking control scheme.
10 July 2019 in IEEE Transactions on Circuits and Systems I: Regular Papers
The modeling and control problem for a grid-connected photovoltaic (PV) power electronic system, which includes a dc/dc boost converter, an inverter and a filter are considered. A linear complementarity (LC) dynamic model of the PV system allows the design of a model predictive controller (MPC). Dynamic models of the subsystems are obtained and merged in order to represent the whole PV system in a compact and comprehensive LC model, which is valid for all operating modes of power converters and PV cells involved in the energy conversion process. A finite-control-set MPC problem is formulated as a mixed-integer quadratic program subject to the dynamic LC model and pulse width modulators. The minimization of an objective function aimed at tracking dc voltage and grid current references provides directly the commands for the switches of the boost converter and inverter. Numerical results show the effectiveness of the proposed strategy for maximum power point tracking and synchronization to the grid under dynamic scenarios characterized by variations of the solar irradiance.
In this paper, a recently proposed approach for modeling switched networks (SNs) and power electronic converters is presented. Actually, typical modelings of SNs such as switched models (SMs) and the modified nodal analysis (MNA) exhibit certain drawbacks for simulating switching. For example, the SMs, which consider the switches to be ideal, require to obtain a set of dynamical models that describes all circuit topologies and need to calculate the switching instants of electronic devices (EDs) (diodes, thyristors, transistors). Thus, SMs do not consider a complete switched model that calculates by itself the natural switching instants of EDs and that considers their continuous and discontinuous conduction. For models based on the MNA, the analysis adds nonlinear functions of the voltage-current characteristics of EDs. However, this consideration involves iterative processes, which implies a high computational cost. In this paper, mixed and classical linear complementarity systems (LCSs) are validated to form complete switched models that include continuous and dicontinuous conduction of EDs without requiring iterative processes. It is shown that LCSs are useful to compute dynamical and steady-state responses of SNs. In particular, it is seen that the steady-state solution can be obtained in a direct way without requiring iterations and a consistent initial condition, which is a goal in power systems analysis. As example, a controlled buck-boost dc/dc converter is simulated to show the potential of the complementarity framework.
A systematic method for constructing models in the complementarity framework from a bond graph is proposed. Bond graphs with and without storage elements in derivative causality are considered. The proposed method allows the study of switching systems represented by a bond graph model of fixed causality. The proposed methodology allows the complementarity framework to be exploited in different engineering areas by using bond graphs. The idea of representing a unidirectional switch with a model that is essentially the same as a diode is presented. By employing a similar representation for diodes and switches, the modeling and simulation of power switching converters are simplified and become more intuitive. Two application examples are shown. A non-inverting buck-boost converter and a zeta converter with an element in derivative causality are simulated.