Search

Finite Control Set Model Predictive Control (FCS-MPC) is an alternative to conventional linear controllers in power converters due to its attractive properties such as its conceptual simplicity, flexibility, direct consideration of non linearities as well as constraints, and fast control response. This last feature imposes computational constraints and short prediction horizons are therefore needed. In fact, in many applications an horizon one FCS-MPC controller is used. However, using FCS-MPC with a short prediction horizon in topologies that exhibit non-minimum phase behavior such as active-front-end rectifiers, dc-dc boost converters, and Z-source converters, among others, may lead to control issues. This work studies the origin of these control issues using nonlinear control theory considering a dc-dc boost converter as a case study. It is shown that the difficulty to directly control the converter output voltage using short horizon FCS-MPC is the resulting unstable internal dynamic, as the short horizon FCS-MPC controller acts like an input-output linearizing controller. To solve the problem of the unstable internal dynamic, a short horizon FCS-MPC controller based on input-state linearization is proposed allowing correct operation of the system. Experimental results show the feasibility of the proposal.

Finite control set model predictive control can be applied to a power converter if there is an accurate existing model of the converter. The best results will be achieved if and only if the parameters and variables that make up the system are properly estimated. If this is not the case, the predictions made using these strategies may be erroneous and cause problems, such as steady-state error with respect to the assigned desired references. This paper presents a predictive control strategy with integral action that compensates for the differences between the estimated model and the inverter with the objective of achieving a zero steady-state error without requiring external loops or state observers. The proposed strategy is tested on a single-phase Z-source inverter so as to evaluate the error in both the ac and dc controlled variables with respect to their references to their cosigns. The experimental results confirm that the proposed strategy achieves a zero error in the steady state while maintaining the fast dynamic response of the classic predictive control.

Finite Control Set Model Predictive Control is an emerging alternative for the control of power converters. Its main advantages are its conceptual simplicity, fast control response, flexibility and direct consideration of non linearities as well as constraints. Due to computational constraints, short prediction horizons are usually desired in power converters. However, control issues may appear when applied in topologies that exhibit non-minimum phase behavior such as active front end rectifiers and boost dc-dc converters. This work proposes the use of the concept of statically equivalent outputs in order to control these systems with the shortest prediction horizon. The proposed predictive controller partially inverts the plant dynamics, determining an auxiliary output by means of nonlinear control tools. Theoretical and simulated results show the feasibility of the proposal.

Future distribution grids will require to integrate renewable energies in higher penetration levels than we are used to today in conventional power systems. To make this possible, key aspects of their operation such as reliability and robustness must been in safe levels. To accomplish the latter, two different technologies have been proposed: (a) Virtual Synchronous Machines and (b) Smart Transformers. Despite all the benefits that each one of these technologies provide, solid-state transformers have not been very successful in operating as smart transformers. Then, hybrid transformers were proposed as a more feasible solution and with the capability of providing ancillary services to the distribution grid. In this paper, a new type of hybrid transformer that incorporate features of a virtual synchronous machine is proposed. The proposed transformer is capable of emulating virtual inertia as an ancillary service, and, at the same time, providing other basic services such as reactive power injection and voltage compensation. Simulation results provide evidence of the advantages of using this type of hybrid transformers in distribution grids.

Finite control set model predictive control (FCS-MPC) has been widely investigated in recent years due to its ability to handle optimization problems with multiple control objectives in a diverse variety of systems. Moreover, its direct implementation in digital-based systems has made it an attractive strategy in static power converter applications. However, its characteristics such as variable switching frequency and spread harmonic spectrum limit the use of standard MPC due to power losses, audible noise, steady-state performance, and resonances. To mitigate these problems and extend the FCS-MPC applications to new areas, this paper proposes a new hybrid predictive control scheme, capable of achieving a harmonic spectrum distribution similar to that obtained with a pulse-width modulation scheme. The proposed strategy is based on a system model to generate an optimization, and, at the same time, an input restriction in the cost function of the standard FCS-MPC. This new approach is validated through experimental tests carried out in a grid-connected Cascaded H-bridge inverter.

In this paper, an operation scheme for a single-phase grid-connected quasi-impedance source inverter (qZSI) is proposed. The proposal indicates that, to achieve a more stable operation for the microgrids, the qZSI must emulate the inertia of a real synchronous machine (SM) through a virtual synchronous generation (VSG) strategy. The proposed VSG strategy generates the setpoints for the ac and dc sides of the converter. These references are taken by a direct model-based predictive control (MPC) scheme that is responsible for controlling the variables on both sides of the inverter. For implementing the MPC scheme, the model of the qZSI will allow predicting the variables that must be controlled. Simulations are shown to demonstrate the feasibility of this proposal. In the simulations a desirable performance of the converter will be observed, controlling the power injection to a single-phase grid as if this converter were an SM.

Cascade H-bridge multilevel (CHB-ML) inverters are an excellent alternative to supply power to the ac micro-grids since they have high reliability and achieve an acceptable quality of voltage at their output terminals. On the other hand, the standard direct model predictive control (MPC) scheme is a very intuitive strategy to control this type of converters, but in general, has a steady-state error when operates with low sampling frequencies and/or if the parameters of the predictive model are different in the real system. In this paper, a CHB-ML inverter that injects power to a micro-grid using a new improved direct MPC scheme is proposed. The proposed strategy eliminates the steady-state error in the MPC operating with low sampling frequencies, also maintaining a correct operation when occurring a change in some parameter of the system. Simulated results will demonstrate the feasibility of the proposal.

Multi-cell topologies based on three-phase input / single-phase output power converters have the drawback of dealing with an important oscillatory second harmonic current in the DC link. This requires the use of a large DC link capacitor in order to minimize the DC link voltage ripple. Different control schemes have been presented to reduce the second harmonic generated for the H bridge inverter aiming the design of the DC link capacitor without oversizing it. However, this is achieved at expense of much distorted AC supply currents and increased modulating signals amplitudes. In this paper, these problems are accurately characterized with the objective of mathematically identify the limitations of those schemes.

Design criteria are addressed to select the parameters of an asymmetric multilevel shunt active power filter based on a Cascaded H-Bridge topology. The proposed approach is meant to simultaneously compensate both fundamental reactive power and current harmonics. The presented methodology allows to systematically design the main parameters of the power converter, particularly the inductive filter, the DC capacitors and the DC voltage level of the power modules. The procedure allows to consider not only a single compensation condition but the entire operating region. The obtained results are supported with mathematical derivations and proved through simulated waveforms.

Multilevel converters are widely used in medium voltage applications. One of the most popular topologies in this field is the 18 pulses diode rectifier, which uses a mu Iti pu I so transformer that realizes a current harmonics cancelation that are generated by the three phase diode rectifiers. In this work an efficiency comparison of the aforementioned topology and a Multi-Cell Active-Front-End (AFE) Rectifier operating with Finite Control Set — Model Predictive Control is performed. This efficiency comparison is focused on the rectifier losses divided into two types. The first one is the semiconductors switching and conduction losses. The second one evaluates the transformer losses, which are studied under two criteria: copper losses (Standard IEEE C57.10-1998) and the K — Factor (Standard IEEE / ANSI C57.110). The results show that the diode based rectifier topology has lower semiconductors losses, while the topology based on AFE rectifiers has lower transformers losses and a less complex design. Indeed, a reduction of a 7.2% in copper losses is achieved in the transformer secondary with respect to a multi-pulse transformer secondary.

This work presents a new control scheme based on a finite states Model Predictive Control aimed to fix the switching frequency and to reduce the spectrum dispersion on the output variables of pulse width modulation (PWM) static converters. These topics are particularly important in the operation of power converters where passive filters present resonant modes and in the switching losses reduction. Contrary to standard model predictive control algorithms, the proposed scheme allows for obtaining time-continuous outputs from the optimization process. Thus, the advantages of modulated schemes, like a defined spectrum and fixed switching frequency, are inherited. The tradeoff is a slight reduction on the dynamic response, in comparison with previous schemes using direct non-modulated outputs. Simulated and experimental results show the proper performance of the proposed scheme.

In the last decades, finite control set-model predictive control (FCS-MPC) has been extensively investigated. This control strategy uses the system model to determine the optimal input that minimizes a predefined cost function. This procedure is performed by evaluating the admissible inputs of the converter model. In order to achieve a proper performance of the control strategy, a small sampling time must be used. In practice, the computational effort and processing time are critical factors that depend directly on the number of admissible states of the topology. With the objective of reducing the execution time needed by the algorithm, it is possible to take advantage of the high degree of parallelism that can be obtained from a field-programmable gate array (FPGA). This paper deals with the implementation of FCS-MPC in an FPGA XC3S3500E to control a power cell of a cascaded half bridge (CHB) converter. The design is divided into subparts, called modules, which can be projected to control multiple cells using only one chip. In addition, with the aim to restrict the use of hardware to the available resources, the document presents a series of considerations related to the algorithm and its implementation. This paper presents a detailed guide with emphasis on the design and outlining of each module in order to reduce the developing times for researchers in the area of power converters.

An asymmetric inverter features different dc-link voltages for each power cell and depending on the dc voltage ratios, there are binary, trinary, etc., asymmetric topologies. In particular, the trinary one can provide a waveform with $3^{n}$ levels, where n is the number of cells connected in cascade. Hence, for $n = 3$ , an output voltage with 27 levels can be obtained, ensuring a low total harmonic distortion (THD) and reduced $dv/dt$ . However, when the load consumes active power, some power cells of the inverter can operate in regeneration mode, complicating the topology and control. To overcome this problem, a new modulation method for asymmetric inverters is presented. It features an ac output voltage of 13 levels for $n = 3$ , and ensures unidirectional power flow in every power cell. This new modulation method also provides a voltage waveform with low $dv/dt$ and reduced THD (5.7%). To confirm the approach, power flow and sensitivity analysis to changes in the dc voltage ratios are performed. Finally, simulation and experimental results are provided.

Multi-cell converters allow the use of AC motor drives for medium and high voltage level applications using low voltage semiconductors. This type of converter has a multi-pulse transformer which performs the cancellation of the low frequency harmonics generated by the three-phase diode rectifiers. Despite this advantage, the multi-pulse transformer is bulky, heavy, expensive, and must be designed according to the number of power cells, making the transformer topology dependent. This work proposes to reduce the complexity of the transformer in multi-cell converters using Active Front End rectifiers in combination with a Finite Control Set - Model Predictive Control strategy. This is achieved by the strategy as it emulates the multi-pulse transformer harmonic cancellation thanks to the predictive algorithm and the generation of input current references with a phase shift angle for each Active Front End rectifier.

This work presents a new control scheme based on a finite states model predictive control aimed to fix the switching frequency and to reduce the spectrum dispersion on the output variables of static converters while regulating the sate variables as required by the references. These topics are particularly important in the operation of power converters including filters with resonant modes and in the overall efficiency due to the switching losses. The proposed scheme allows obtaining continuous outputs from the optimization process of MPC, thus exploiting the advantages of modulated schemes like known spectrum and fixed switching frequency. Simulated results show the proper performance of the proposed scheme, without significant reduction on the dynamic response in comparison to previous schemes using direct, non-modulated outputs.

Multilevel inverters are the alternative for medium voltage applications. Within the inverters types there are symmetric and asymmetric topologies. The asymmetric inverters have different DC voltage values. The most common topology is when the different cells are implemented in cascade arrangement, where the DC voltage are in multiples of 3, obtaining an AC voltage with 3n = 27 levels (n = 3 cascaded inverters). This topology provides a load voltage with low harmonic content, THD

The asymmetric multi-cell inverters are an alternative to provide an AC voltage with low THD. Considering a voltage DC ratio 1∶ 3∶ 9 for three inverters connected in series an AC voltage featuring 27 levels is obtained. The most common modulation used in asymmetric inverters is the Nearest Level Modulation (NLM) which has the drawback of bringing some cells into regeneration for some values of the modulation index, regardless of the load conditions. This paper presents a new modulation strategy for this asymmetric inverter that forces all the cell to have identical power flow direction. This strategy is based on the fact that the fundamental voltage amplitude of each inverter is required to be always greater than or equal to zero. To confirm the proposed technique, an active power flow analysis is performed as well as simulated and selected experimental results are included.

The main drawback of a Multi-Cell topology based on Current Source Converters is the oscillating power drained by the single-phase inverters from the DC link which requires a large DC reactor in order to avoid a significant variation of the DC current. In this work, a control scheme is proposed in order to compensate the oscillating power, avoiding to oversize the DC reactor at the expense of increasing the distortion of the input currents in each power cell and the instantaneous peak of the modulation index. However, the overall input currents present very low distortion and the instantaneous peak of the modulation index can be reduced by properly designing the multi-pulse transformer in terms of voltage ratio and phase shift. Both, the input currents distortion and the effect in the modulation index are mathematically characterized. Simulation result illustrates the effectiveness of the proposed control approach.

Multi-cell topologies based on three-phase input / singlephase output power modules have the drawback of dealing with an important oscillatory power in the DC link. This requires the use of a large DC link capacitor in order to minimize the DC link voltage ripple. Instead, this paper proposes a control scheme based on predictive control aimed to compensate the oscillatory power that allows the design of the DC link capacitor without oversizing it. However, this is achieved at expense of much distorted AC supply currents and amplified modulating signals. Such effects of the proposed compensating technique based on predictive control are mathematically characterized. The results show that the multi-pulse transformer guarantees distortion free overall AC currents. Simulated results show the feasibility of the proposed scheme.