Search

Finite Control Set Model Predictive Control (FCS-MPC), thanks to its simple and powerful concept, has been applied widely to control the multilevel converters. FCS-MPC allows the control of several objectives and can incorporate restrictions, even nonlinear, into the mathematical model of the system under consideration. However, FCS-MPC needs to evaluate all the switching states of the multilevel converter to find the best switching state to be applied in the next sampling time. For higher-level multilevel converters, this iterative action requires computational capacity that is far beyond the digital controller’s capacity in the current market. This paper proposes a new predictive geometric pre-filtering strategy to reduce the iterations and computational burden without affecting the dynamic performance of FCS-MPC. This method consists of a novel pre-filtering stage that uses the predictive model of the system and geometrical properties to find the sector where the reference vector is located and evaluates few vectors that constitute the optimal sector. The proposed method is experimentally validated using a four-level three-cell flying capacitor converter with 512 voltage vectors, obtaining a 64% reduction in the computational burden, while achieving excellent electrical performance indices and maintains the high dynamic performance of the standard FCS-MPC.

Multilevel grid-tied inverters have been widely introduced for integrating photovoltaic (PV) inverters for several output power ranges. Among the existing multilevel inverters (MIs), the flying capacitor (FC) based topologies are preferred. However, PV systems are subjected to leakage currents due to the existing parasitic capacitance of PV panels and the common mode voltages in PV inverter topologies. Hence, multilevel FC based commonly grounded PV inverters can provide efficient solutions by merging the inherent MIs benefits with total elimination of leakage currents. Therefore, this paper presents model predictive control (MPC) method for commonly-grounded single phase five-level FC PV inverter. The proposed MPC method provides a multi-objective controller for capacitor voltages control in addition to high quality sinusoidal current injection into the grid. The modelling, design, and simulation results of the proposed MPC method are provided in this paper.

A power conversion system based on a diode rectifier, three-level (3L) boost converter and 3L neutral-point clamped (NPC) inverter is a cost effective and reliable solution in contrast to the back-to-back connected NPC converter for the medium-voltage permanent magnet synchronous generator-based wind turbines. This paper investigates a novel predictive control scheme to operate the 3L boost converter and an NPC inverter with fixed switching frequency, high-speed tracking, and accurate locking in steady state. A dual decoupled control loop consisting of deadbeat current control and modulated model predictive current control has been developed for the 3L boost converter and NPC inverter, respectively. The controller for three-level boost converter performs dual functions such as neutral-point voltage control and inductor current control. The controller of NPC inverter regulates the grid currents in stationary frame for the simultaneous control of active and reactive powers.

Model predictive control (MPC) has emerged as a promising approach to control a modular multilevel converter (MMC). With the help of a cost function, the control objectives of an MMC are achieved easily by using an MPC approach. However, the MPC has several technical challenges and issues including the need of accurate system models, computational complexity, and variable switching frequency operation and weighting factor selection, when it comes to the control of an MMC. In the past few years, several research studies are conducted to address some of the challenges and issues in an MPC and developed several model predictive algorithms for an MMC. In this paper, the importance of each challenge and its impact on the system performance is discussed. Also, the MMC mathematical models used in the implementation of MPC are presented. Furthermore, some of the popular MPC algorithms are discussed briefly, and their features and performance are highlighted through case studies. Finally, summary and future trends of MPC for an MMC are presented.

Multilevel converters are the most suitable and efficient candidates for high-power and medium-voltage industrial applications such as motor drives, traction, ship propulsion, regenerative conveyors, wind and photovoltaic energy conversion systems, and power quality conditioners. This book chapter reviews a model predictive control scheme for current control, balancing of dc capacitor voltages, switching frequency reduction and common-mode voltage mitigation in multilevel diode-clamped converters. The continuous- and discrete-time models of control variables of multilevel converters are formulated in terms of the switching states. A multiobjective cost function with suitable weighting factors is designed incorporating the control objectives for the multilevel converter. The cost function minimization is used as criteria for choosing the best switching states which would be applied to the converter during the next sampling interval. The feasibility of the model predictive control scheme is verified by the MATLAB simulation results on three- to six-level diode-clamped converters.

Model predictive control (MPC) is a promising tool to achieve multiple control objectives of a modular multilevel converter (MMC). This book chapter presents direct model predictive control (DMPC) and indirect model predictive control (IMPC) schemes for an MMC. The DMPC and IMPC schemes use sampled data models of an MMC to predict the control variables such as output current, circulating current, and submodule (SM) capacitor voltage. In the DMPC scheme, these predictions are performed for all possible switching states of an MMC. The minimization of error between the reference and predicted control variables is fulfilled through a single cost function. The optimal switching state corresponding to minimum cost value is selected and applied to the MMC. On the other hand, the IMPC combines a predictive algorithm with a classical voltage balancing strategy to control an MMC. The predictive algorithm is designed to control the output current and circulating current of an MMC. These control requirements are fulfilled through a cost function, which is evaluated for all possible voltage levels of an MMC. The optimal voltage level corresponding to the minimum cost value is selected and applied to the voltage balance strategy. The voltage balance strategy ensures voltage balance among the SMs in an arm based on the information of optimal voltage level, the instantaneous value of SM capacitor voltage, and arm current direction. Finally, an optimal switching state is generated and applied to an MMC. The simulation studies are conducted to verify the performance of DMPC and IMPC schemes for an MMC.

In an attempt to decrease the cost of energy, increase the energy conversion efficiency, reliability, power density, and comply with the stringent grid codes, the power electronic converters have emerged in a rigorous manner. In contrary to the standard practice of using multilevel converters for high-power and medium-voltage power conversion systems such as wind energy, the current industry also uses multilevel converters for low-voltage, low-power, high-performance, high-efficiency, high-power-quality applications such as photovoltaic energy. This chapter provides an overview of multilevel converters used in wind and photovoltaic energy systems. From the market-based survey, the most successful multilevel converter configurations are addressed along with few promising topologies available in the literature. For wind energy systems, the multilevel converters are classified into four categories: back-to-back connected converters, passive generator-side converters, multiphase generator-side converters, and direct ac-ac converters. The central, string, and multistring inverter configurations for photovoltaic energy systems are also discussed with respect to the configuration and operation.

An LC-filtered neutral-point clamped (NPC) converter is commonly used in distributed generation, uninterrupt- able power supplies and dynamic voltage restorers. A multiobjective modulated modelpredictive voltage control scheme is proposed to control the load voltages and neutral-point voltage of an LC-filtered NPC converter. In this work, the load voltages and converter currents are controlled simultaneously to improve the load voltage quality. An exact discrete-time model of power conversion system is used to predict the future behavior of state variables. These predictions are evaluated by a triple-objective duty-cycle optimized cost function. The three stationary voltage vectors corresponding to the optimal cost function are synthesized by the space vector modulation. The performance of the proposed control scheme is verified through MATLAB simulation results with linearand nonlinear loads. The results indicate fixed switching frequency operation, fast transient response, and low steady-state errors.

This paper investigates novel control methods for the squirrel-cage induction generator based variable speed wind power generation system composed of back-to-back connected two-level voltage source converters. A modulated model predictive torque control method is designed by combining the operating principles of direct torque control, predictive torque control and space vector modulation to effectively control the electromagnetic torque and stator flux control of an induction generator. Similarly, a modulated model predictive current control method is designed for the grid-side converter to control the active and reactive powers injected to the three-phase grid. The proposed control methods employ duty-cycle optimized cost functions for fixed switching frequency operation, fast dynamic response and low steady-state errors during a wide-dynamic range operation. Simulation results are presented with a 750-kW wind power generation system to validate the effectiveness of proposed control methods.

A three-phase two-level voltage source inverter is the most widely adopted configuration for the grid-integration of large-scale photovoltaic energy systems. To obtain high energy conversion during a wide dynamic range and to comply with the stringent grid codes, faster and high-performance nonlinear controllers are required. This paper develops a novel modulated model predictive current control method to control the photovoltaic central inverter for maximum energy yield and reactive power control. The developed control method estimates the future values of stationary-frame grid currents and assesses them by a duty-cycle optimized cost function. Three stationary voltage vectors corresponding to the low cost value are then synthesized by the space vector modulation stage to produce gating signals for the central inverter. The proposed control method ensures constant switching frequency, low steady-state errors and fast transient response simultaneously during rapid environmental changes. To verify the feasibility of the developed control method, simulation studies are conducted on a 1-MVA photovoltaic system, under solar irradiance, temperature and reactive power changes.

In this paper, an efficient modulated model predictive current control of back-to-back connected neutral-point clamped (NPC) converter in direct-driven surface-mounted permanent magnet synchronous generator-based variable-speed wind energy conversion system is presented. The proposed control scheme fulfils the wind energy requirements such as maximum power point tracking, control of neutral-point voltage and grid reactive power regulation through the control of generator and grid currents with fast transient response, smooth steady-state and fixed switching frequency operation simultaneously. The three voltage vectors corresponding to the optimal cost functions of NPC rectifier and inverter are synthesized by the space vector modulation with active control of dwell times to balance the DC capacitors voltage. To evaluate the proposed control scheme, dynamic simulation results are presented for a 3 MW and 3000 V wind turbine under varying wind speed conditions.

In this paper, novel modulated model predictive toque and power control schemes are proposed for machine- and grid-side converters, respectively in the gearless permanent magnet synchronous generator-based wind turbines. The state variables such as generator torque and flux, and grid active and reactive powers are predicted and evaluated by two independent duty-cycle optimized cost functions. The stationary voltage vectors corresponding to the optimal duty cycles are then synthesized by the space vector modulation to achieve fixed switching frequency operation for the back-to-back connected voltage source converter. With the proposed control schemes, the state variables are controlled precisely with fast transient response and low steady-state errors to accomplish the control goals of wind turbines during a wide-dynamic range operation. The feasibility of the proposed schemes is verified through simulation results using a 750-kW wind turbine system.

In photovoltaic energy systems, a two-stage power conversion system composed of boost converter and three-phase voltage source inverter improves the energy conversion efficiency. A novel deadbeat current control (DBCC) for the boost converter and a modulated model predictive current control (M2PCC) for the grid-side converter are proposed to obtain constant switching frequency operation. The DBCC regulates the inductor current for the maximum power point tracking, while the M2PCC controls the stationary-frame grid currents such that the grid active and reactive powers are regulated with excellent power quality. The proposed control strategies ensure less steady-state error and fast transient response under a wide dynamic range. To verify the feasibility of the proposed controls, simulation studies are conducted on a 12-kW system at different irradiance, temperature and grid reactive powers.

A novel modulated model predictive current control for a two-level four-leg voltage source inverter is proposed in this paper. The proposed scheme is designed by combining the operating principles of traditional predictive control with the three-dimensional space vector modulation to enhance the output power quality, while producing fixed switching frequency operation and fast dynamic response for the four-leg inverter. In contrast to the classical predictive control which applies the switching state corresponding to optimal cost function, the proposed controller uses three active vectors and a zero vector to optimize the cost function. Finally, the switching states for four-leg inverter are produced by a modulation stage containing nine-segment switching pattern similar to the classical three-dimensional space vector modulation. The proposed control scheme is validated through simulations considering three-phase unbalanced reference currents and loads.

This paper presents an efficient modulated model predictive current control for direct driven wind energy system composed of surface-mounted permanent magnet synchronous generator and back-to-back connected voltage source converter. The proposed control method fulfils the generator-side control requirements such as maximum power point tracking, and grid-side control objectives such as DC-link voltage control and power factor correction. These objectives are achieved through the regulation of generator and grid currents with fast transient response, smooth steady-state and fixed switching frequency operation simultaneously. The proposed controller predicts the future behavior of generator and grid currents using quasi-exact discrete-time models and eight voltage vectors, and then evaluates them by two independent cost functions. Finally, the switching sequence is designed by the space vector modulation using three stationary voltage vectors corresponding to the optimal cost function. The performance of the proposed method is validated through the MATLAB simulations using a 750-kW wind energy system at different wind speeds and grid reactive powers.

A simple and efficient model predictive control technique for a six-phase permanent magnet synchronous generator-based wind energy system with fixed switching frequency is presented in this paper. The power converter interface to the grid features a dual diode-bridge rectifier, followed by a three-channel (3C) boost converter and a two-channel grid-tied voltage source inverter (2C-VSI). The proposed control technique is divided into two decoupled and independent control loops: the first corresponds to a deadbeat current control for the 3C-boost converter, while the second is a modulated model predictive current control for the 2C-VSI. The maximum power point tracking is achieved through the regulation of inductor currents of 3C-boost converter, whereas the 2C-VSI is in charge of grid active and reactive power control with excellent power quality. The proposed control techniques ensure fixed switching frequency and interleaved operation for the 3C-boost converter and 2C-VSI under a wide dynamic range, leading to less steady-state errors and fast transient response with effective distribution of power among the channels. To evaluate the proposed control technique, dynamic simulation results are presented for a 1.5 MW commercial wind turbine under varying wind speed conditions.

In this paper, a novel modulated model predictive voltage control method is proposed for a two-level four-leg inverter with an output LC filter. To decouple the dynamics of state variables in an output LC filter and to improve the load voltage quality, the capacitor voltages and inductor currents are controlled simultaneously. The proposed controller predicts the state variables using the discrete-time model of LC-filtered four-leg inverter for 16 voltage vectors and evaluates them by a dual-objective cost function. The active and zero voltage vectors corresponding to the lowest cost value are realized by the three-dimensional space vector modulation, resulting in fast transient response, low steady-state error, and harmonic components centered around the fixed switching frequency. The proposed method's feasibility is confirmed through MATLAB simulation studies under balanced/unbalanced reference voltages and nonlinear loads. The performance comparison of proposed controller with the classical predictive voltage control is presented considering the transient and steady-state operating conditions.

The even-order harmonics have detrimental effect on the grid compared with the triplen harmonics. In this paper, a novel modulated model predictive current control strategy capable of eliminating even-order harmonic components from the two-level inverter output voltage/current waveforms is proposed. The proposed control strategy combines the features of fast dynamic response from the classical model predictive control and fixed switching frequency operation from the linear control. The proposed control strategy regulates output currents through the duty cycle based cost function optimization. The optimal voltage vectors are used to form a P-type and N-type switching sequences depending on the sector number and generate the gating signals for the inverter. The performance of the proposed control strategy is validated through MATLAB simulations.

In permanent magnet synchronous generator (PMSG) based wind turbines, the back-to-back connected voltage source converter (VSC) can be replaced by a power converter based on diode rectifier, boost converter, and a grid tied VSC to decrease the cost and improve reliability. To accomplish fixed switching frequency operation, a deadbeat current control (DBCC) for the boost converter and a modulated model predictive current control (MMPCC) for the grid tied VSC is proposed in this paper. The DBCC regulates the inductor current for the maximum power point tracking, while the MMPCC controls the stationary-frame grid currents such that the grid active and reactive powers are regulated with excellent power quality. The proposed control method ensures less steady-state error, fast transient response and fixed switching frequency under a wide dynamic range. The simulation studies are conducted on a 750-kW direct driven surface-mounted PMSG wind turbine at different wind speeds and reactive powers to verify the feasibility of the proposed control method.

The permanent magnet synchronous generator (PMSG) is dominantly used in the present wind energy market. Reflecting the latest wind energy market trends and research articles, this study presents a survey on important electrical engineering aspects for PMSG-based megawatt-level wind energy conversion systems (WECSs). A comprehensive analysis on power converter topologies for wind turbines (WTs), grid integration of wind farms, digital control schemes, fault-ride-through compliance methods, and future trends is presented. The updated market share, technology trends, WT products information, in-depth technical analysis, and promising research works highlighted in this study will help the reader to understand the state-of-the-art and emerging technologies for PMSG-based WECS.

The indirect model predictive control (I-MPC) is one of the reduced computational predictive strategies, used to control the modular multilevel converter (MMC). This approach operates at higher switching frequency, which is not desirable for high-power applications. This study proposes an integrated solution for MMC by combining predictive control with the classical energy balancing approach. To implement the predictive algorithm, a detailed three-phase MMC model is presented. The three-phase model includes the zero sequence voltage to reduce the switching frequency of submodules. In addition, the output power quality is enhanced, while operating at reduced switching frequency. The performance of integrated approach is experimentally validated on a laboratory prototype under balanced and unbalanced conditions. In addition, the performance of integrated approach is compared with the existing methodology in terms of output current ripple, switching frequency, computational complexity, and total harmonic distortion.

Submodule (SM) capacitor voltage ripple is one of the major concerns in modular multilevel converters (MMCs). Capacitor voltage ripple leads to the double-frequency circulating current (CC) in legs, thereby resulting in a cascading effect of increased peak value of the arm current, semiconductor device stress, and power losses in MMCs. In this study, a model predictive control (MPC) with common-mode voltage (CMV) injection is proposed to minimize capacitor voltage ripple and the magnitude of CC. A discrete-time mathematical model of the MMC with CMV is presented to predict the future behavior of the control variables. The injection of CMV guarantees arm voltage balancing without CC control and long-term stability of MMC without large capacitors. The dynamic and steady-state performances of MPC with CMV injection are verified on an MMC with three-level flying capacitor SMs. A performance comparison between the proposed approach and the conventional MPC is also presented. Simulation and experimental studies show that CMV injection significantly reduces the capacitor voltage ripple and the CC in legs. The proposed approach also improves output voltage and current waveform quality.

This chapter deals with the low???voltage (LV) and medium voltage (MV) passive generator???side (PGS)???converters???based Type 4 PMSG WECS and the digital control using model predictive control (MPC). The control system uses only feedback signals and does not require the model and parameters of PMSG. The model of inductor current and DC???link capacitors voltage is presented in terms of boost converter parameters and switching states. The multilevel boost converters are used in conjunction with multilevel diode???clamped inverters to increase WECS operating voltages. The feedback signals measured variables from the generator side, DC link, and grid side are provided to the intermediate DC or DC converter and GSC control systems. Two individual control loops can be developed because of the decoupled nature of the MSC and GSC. The control objectives can be interchanged between the MSC and GSC. The DC or DC boost converter can handle the regulation of the net DC???bus voltage, whereas the GSC can perform MPPT operation. The grid???side NPC inverter controls the reactive power and net DC???bus voltage through the regulation of decomposed grid currents.

This chapter discusses how continuous time (CT) and discrete time (DT) are two alternative frameworks that are used in the modeling of variables. Generator‐ and grid‐side currents are regulated and fed into the grid in a controlled manner by properly regulating the turn‐on and turn‐off statuses of the switching devices in the power converter. The objective of discretization methods is to minimize the discretization error to an acceptable minimum value and to involve simple discrete mathematical expressions and calculations. Exact discretization methods are further classified into pole‐zero matching methods and invariant approaches, which include impulse invariant, step invariant (ZOH), and ramp invariant methods. Discretization methods are selected according to the specific application and nature of control variables. The forward and backward Euler methods provide implicit and explicit solutions, respectively. The combination of these two methods obtains a better solution in DT. The bilinear transformation is also known as the Tustin or trapezoidal method.