Search

This paper focuses on the analysis and design of two multilevel–multicell converters (MMCs), named 3-phase 5-Level E-Type Multilevel–Multicell Rectifier (3Φ5L E-Type MMR) and 3-phase 5-Level E-Type Multilevel–Multicell Inverter (3Φ5L E-Type MMI) to be used in microgrid applications. The proposed 3-phase E-Type multilevel rectifier and inverter have each phase being accomplished by the combination of two I-Type topologies connected to the T-Type topology. The two cells of each phase of the rectifier and inverter are connected in interleaving using an intercell transformer (ICT) in order to reduce the volume of the output filter. Such an E-Type topology arrangement is expected to allow both the high efficiency and power density required for microgrid applications, as well as being capable of providing good performance in terms of quality of the voltage and current waveforms. The proposed hardware design and control interface are supported by the simulation results performed in Matlab/Simulink. The analysis has been then validated in terms of an experimental campaign performed on the converter prototype, which presented a power density of 8.4 kW/dm3 and a specific power of 3.24 kW/kg. The experimental results showed that the proposed converter can achieve a peak efficiency of 99% using only silicon power semiconductors.

Power-Hardware-in-the-Loop (PHIL) platforms are gaining increasing attention as a valid solution for testing of converters operating in power electronics systems. When the Device-Under-Test (DUT) is represented by a power converter designed to supply a Variable Speed Drive (VSD), electrical machine emulation must be performed by the PHIL. However, emulation of electrical machines imposes several requirements to the design of the PHIL test platform. Indeed, the emulable range of the machine operating points is constrained by the choice of the DC-bus voltage and of the coupling network. Furthermore, coupling network sizing strongly affects the stability of the emulation. In this paper, the analytical relation between the coupling network, the emulator DC-bus voltage and the machine operating range is described. A design procedure is then derived for both the DC-bus voltage and the coupling network. Afterwards, a stability analysis is carried out to determine the stability characteristic of the PHIL emulation as a function of the coupling network choice and of the main parameters of the DUT control loop. The FPGA implementation of the PHIL is then described. Simulation and experimental results show the capability of the proposed implementation, with respect to both transient and steady state machine emulation.

Repetitive Control (RC) enables inverters to inject, at fundamental frequency, high quality sinusoidal current into the grid or to provide low THD voltages in off-grid applications. However, digital RC, which can achieve zero steady state error for each periodical signal with known period, is not able to track or reject variable frequency periodic signals. In fact, digital RC weakness is related to performance degradations when it deals with variable frequency references. In other words, existing RC algorithms require an integer number of samples every period, while in some industrial applications the number of samples per period could be a non-integer time-varying value. Therefore, RC would lead to a significant quality degradation of the provided outputs when the grid frequency or the off-grid desired frequency changes. According to that, a newly RC structure with adaptive decoupled sampling is proposed to work with a constant delay line, regardless of the desired output frequency in both grid-tied and stand-alone operation.

A novel fault detection and ride-through method for the three-phase 4-Leg Inverter (3$\Phi$4L Inverter) used in off-grid applications is addressed in this paper. Open circuit and short circuit faults of the power devices located in the $3 \Phi 4 \mathrm{~L}$ Inverter have been theoretically analyzed. Hence, the fault detection and converter protection strategy has been implemented using a combination of a resonant controller in the $\alpha \beta$ frame and a repetitive observer. The proposed strategy has been at first validated by means of simulation results performed in MATLAB/Simulink environment. The fault detection and ride-through algorithm has been then implemented in FPGA using LabVIEW environment and full experimental verification are performed on the prototype of the three-phase 4-leg inverter.

This paper proposes a 3-phase 5-Level Flying E-Type Topology for photovoltaic applications, the converter is able to balance the voltage across the flying capacitors thanks to the modulation technique. The flying capacitor voltages are self-balanced without using any additional balancing circuits or closed-loop controls. Operation modes and characteristics of the proposed converter are presented, as well the voltage balancing approach based on the sinusoidal pulse width modulation scheme is explained. The analysis of the proposed converter has been validated through the digital-switching model realized in Matlab/Simulink environment. Preliminary results show the effectiveness of the proposed converter to balance the DC-bus voltage. Experimental tests on the converter prototype are illustrated.

Solar photovoltaic (PV) is one of the most favourable renewable energy types today, the use of which has grown throughout the world, and much more is expected in the future due to the continuous price drop. It is important to further develop renewable systems to generate green energy, which results in the reduction of air pollution and in enabling a future sustainable society for the next generation of people. Accordingly, new topologies of the multilevel converter have been proposed and investigated for PV applications to achieve high efficiency and high-power density. This study presents the analysis and design of the symmetrical three-phase E-type inverter (S-3Φ7L E-type). The proposed topology is directly derived from the T-type converter and exhibits seven voltage levels at the output offering interesting performance in terms of power semiconductors current and voltage stress. The hardware design process of the converter has been carried out to enhance the efficiency conversion. Finally, theoretical results are carefully verified by testing the converter prototype, featuring a peak efficiency of 99%.

Today, for several economic and environmental reasons, Electric Vehicles (EVs) have a notable impact over the automobile world market. Accordingly, the amount of charging stations with reduced charge time is increased. The design aspects and the realization of the isolated Four-Port bidirectional Dual Active Bridge DC-DC Converter (4P-DAB) for fast charging applications are described in this paper. The 4P-DAB can handle different sources and loads thanks to the connection between four full bridges to the four windings of the high-frequency (HF) transformer. The design procedure and the power devices selection are carried out. The conversion efficiency as a function of the output power for different control strategies is presented using the Silicon-Carbide (SiC) power semiconductors. Simulation and preliminary results are verified by realizing the switching model of the 100 kVA 4P-DAB in MATLAB/Simulink.

This paper focuses on the modelling of the current ripple in a Power-Hardware-in-the-Loop (PHIL) emulator of an isotropic Permanent Magnet Synchronous Machine (PMSM). In such applications, one of the main challenges is to replicate the current behavior when the inductance that is installed in the PHIL differs from that of the emulated machine. This is because the power emulator must emulate the current ripple at the switching frequency of the Device Under Test (DUT). A theoretical analysis is carried on deriving the equations that describe the current waveform in the PHIL. An expression for the current error between the PHIL system and the emulated one is derived and the parameters that affect the error are highlighted. Mathematical expressions for the control actions needed to bring the current error to zero are proposed and the technical limitations in their application are discussed. Simulation results confirm the effectiveness of the proposed analysis.

Number of electric vehicles (EVs) continues to expand rapidly. This leads to significantly increase the amount of charging stations with reduced charge time. This paper focuses on design and realization of the isolated Four-Port bidirectional Dual Active Bridge DC-DC Converter (4P-DAB) to be used in fast charging applications. Each port of the 4P-DAB can manage different sources and loads thanks to the full bridge converter. A magnetic node is created connecting all the ports through a high-frequency multi-winding transformer. The efficiency conversion as a function of the output power has been carried out using Silicon-Carbide (SiC) power semiconductors. Different control strategies have been applied to improve the performance of the converter. Simulation and optimization results are verified by realizing the switching model of 100 kVA 4P-DAB in MATLAB/Simulink.

Power electronic converters have a heavily influence in our lifestyle, as they are used today in several applications, such as in-home and office appliances, electromedical apparatus, industry machine tools, as well as in electrical energy transmission and distribution systems. Power electronic converters are based on using fast-switching power semiconductors, which allow the desired shaping of the output electrical quantities waveforms, however, these power devices are also the most fragile component of the power conversion apparatus. Thus, to ensure safety and reliability of the power conversion system, a fault detection method must be suitably embedded in the control system. When a failure occurs, the fault detection and protection system become the most important function of the converter control unit. This article is devoted to the theoretical analysis of all the possible faults occurring in the three-phase five-level E-type converter (3Φ5L E-type converter) and based on such an analysis, a suitable fault management strategy is proposed. Fault effects are initially analyzed on the long-term time scale, from 10 μ s up to 100 ms. The fault effects, the detectability, and the safety shutdown actions are identified and summarized. The analysis is covered by simulation results performed in MATLAB/Simulink. Following the theoretical analysis, a control algorithm has been implemented in field-programmable gate array using LabVIEW. Finally, experimental tests have been performed on a prototype of the proposed multilevel converter in order to validate the proposed approach to management of the envisaged fault conditions.

The last few decades have seen a significant increase in the number of electric vehicles (EVs) for private and public transportation around the world. This has resulted in high power demands on the electrical grid, especially when fast and ultra-fast or flash (at the bus-stop) charging are required. Consequently, a ground storage should be used in order to mitigate the peak power request period. This paper deals with an innovative and simple fast charging infrastructure based on supercapacitors, used to charge the energy storage system on board electric buses. According to the charging level of the electric bus, the proposed fast charging system is able to provide the maximum power of 180 kW without exceeding 30 s and without using DC&ndash, DC converters. In order to limit the maximum charging current, the electric bus is charged in three steps through three different connectors placed between the supercapacitors on board the bus and the fast charging system. The fast charging system has been carefully designed, taking into account several system parameters, such as charging time, maximum current, and voltage. Experimental tests have been performed on a fast charging station prototype to validate the theoretical analysis and functionality of the proposed architecture.

New sensorless observers (i.e., from stator currents/voltages measurements only), to be included into a simple observer-based sensorless control for the tracking of non-definitely zero speed references in nonsalient-pole surface permanent magnet synchronous machines (PMSMs), are proposed. They are obtained through a “minimum distance” modification from recently presented position-sensorless observers (i.e., from rotor speed and stator currents/voltages measurements only). The rotor speed estimate is here directly provided by the phase locked loop (PLL)-based third-order steady-state linear kalman filter (SSLKF) that has been previously used to mitigate the distortions on the estimated position. Experimental results illustrate the effectiveness of the proposed approach in a speed sensor fault-tolerant scenario.

This paper deals with the real time implementation of an Implicit Model Predictive Control strategy applied to a 3-phase Voltage Source Inverter (VSI) in stand-alone applications. Solving the optimization problem online with a sufficiently short sampling time allows to directly control the output voltage while respecting the desired constraints. Due to the resulting complex calculations and the related time-constraints, the implementation has been carried out on an industrial-grade FPGA to take advantage of the true parallel execution capabilities.

This paper deals with a newly conceived control topology based on the Repetitive action in 4-leg VSI for stand-alone Ground Power Unit (GPU) applications. In such mode of operation, a dedicated controller has to regulate the inverter output voltages, which are measured at the output of the power filter. The proposed Repetitive Control can provide the required harmonic compensation capabilities that are mandatory to comply with the Standards especially in the aircraft field, when balanced and unbalanced, linear and non-linear loads have to be fed at the fundamental frequency of 400Hz.

This paper proposes a constant switching frequency Finite Control Set Model Predictive Control (FCS-MPC), formally Modulated FCS-MPC or M2PC, for a 4-leg inverter having an Active Split DC-bus on the fourth leg. The great advantage of MPC over linear control schemes is the very fast transient response it is capable to produce; it also can handle general constrained nonlinear systems with multiple inputs and outputs in a unified and clear manner. These features are highly valuable in power electronic converters used to supply the electrical utility loads in micro-grids. However, one of the main drawback of the MPC is its variable switching frequency, above all in system with accurately tuned output power filters (i.e. switching traps), which is the case when stable voltage waveforms with very low harmonic content are required. The proposed investigation relates with the application of a constant switching frequency variant of the MPC to a 4-leg inverter with a specifically tuned filter to assure high quality voltage supply even in case of non-linear and unbalanced loads.

This paper deals with a synchronous demodulation scheme to be used to obtain the speed and position from resolver position sensor. Proposed algorithm is intended to be implemented on a FPGA to provide the elaborated information for very low latency control loops. Moreover, resolver excitation circuit has been simplified working directly with a square wave signal. Resolver frequency behavior has been also taken into account due to the non-sinusoidal excitation.

1Simple adaptations of the latest design ideas in [2] suggest the existence of a new observer to be used for sensorless speed control of (nonsalient-pole surface) permanent magnets synchronous machines. The rotor speed measurement-assumed to be measured in [2] by a low-resolution transducer providing a single pulse per revolution — is here replaced by the output of a PLL (Phase Locked Loop) block that is based on a third-order Steady-State Linear Kalman Filter (SSLKF). Experimental results concerning the new observer are provided.

The paper deals with a high efficiency boost rectifier for high rotational speed Gen-Set applications. It is tailored around a permanent magnet synchronous generator (PMSG) featuring an open end winding configuration. According to such a configuration the generator stator windings are connected on one side to a three level T-type unidirectional rectifier, and on the other side, to an auxiliary two-level inverter. The multilevel rectifier is operated at the fundamental frequency, while, the two-level inverter is driven by a standard sine PWM technique, in order to suitably shape the currents produced by the electric generator. The entire system is equivalent to a six-level multilevel converter. As the main stream power is handled by the multilevel rectifier operating at low switching frequency, a higher efficiency is achieved compared to a similar multilevel rectifier driven through a PWM technique, even considering extra losses due to the auxiliary inverter. The last can be supplied through an active source or a floating capacitor, and plays a key role in balancing the voltage across the DC Bus capacitors of the T-type rectifier, making unnecessary additional circuits. Suitable control solutions have been implemented considering either an active source, either a floating capacitor. Simulation results confirm the effectiveness of the proposed configuration.

This paper deals with the gain scheduling feature for zero-phase-shift direct repetitive control (dRC) implemented in a FPGA system. Proposed control structure is intended to be used in ultra-low THD, fast transient response stand-alone generating applications. Moreover, implementation aspects of the dRC and gain scheduling algorithm are highlighted with reference to an industrial-grade μProcessor-FPGA control system and its graphical programming capabilities.

This paper proposes a Predictive Control, formally Dead-Beat (DBC), for a four-leg inverter having an Active Split DC-bus on the fourth leg and LC filters on phase-to-neutral outputs. Such a configuration permits to reduce the voltage ripple on the neutral point connected to inverter grounding. As only few control techniques have been investigated for Active Split DC-bus, the paper proposes to investigate the performance of DBC, which has been widely used for other power electronics applications. The main advantage of DBC over the classical PI or Resonant controller is that no tuning is required for control loop, while obtaining very fast transient response as well it can handle general constrained nonlinear systems with multiple inputs and outputs in a unified and clear manner. These features are highly valuable in power electronic converters used to supply the electrical utility loads in micro-grids. However, one of the main drawback of the DBC is the limited capabilities on harmonics compensations required when supplying unbalanced and non-linear loads. The paper presents continuous-time and discrete-time models of DBC applied to a four-leg VSI with Active Split DC-bus, highlighting the performance through simulation results as well as experimental tests.

This paper deals with a dedicated load-adaptive phase compensation algorithm to be used in repetitive control (RC) based stand-alone four-leg voltage source inverter (VSI). The plant model is achieved; its inherent modifications according to the operating point are highlighted and used to properly adapt the RC structure. Modification of the RC parameters is described to obtain the desired phase compensation capabilities achieving a zero-phase-shift condition at each harmonic. This allows us to increase the gain of the repetitive controller at high-order harmonics, thus yielding a better VSI output voltage with strongly reduced total harmonic distortion and faster dynamic response. As a consequence, the VSI output voltages are almost independent from the loads to be fed, and the four-leg VSI with the proposed zero-phase-shift direct RC is an ideal candidate to supply sensitive loads in microgrid, in particular for stand-alone applications.

Three-phase four-leg inverters are commonly employed to supply unbalanced loads avoiding the use of a transformer. A suitable output filter is necessary to improve the harmonic content of the output voltage and current waveforms. To this aim, different kind of filters, presenting both symmetrical and asymmetrical phase-to-phase and phase-to-neutral behavior, can be used. The paper discusses and analyzes the effects of different carrier-based modulation techniques, based on the independent control of the fourth leg from that of the other phases, with reference to the different filter topologies. The comparison, performed through extensive simulations and experimental tests, have demonstrated that in case of the asymmetrical filter utilization, such as the active split dc-bus (also known as split-link ), the conventional continuous and discontinuous space-vector modulations produce a significant harmonic content in the output voltages even at no-load condition. Moreover, the analysis is extended considering the interaction between the modulation strategy and load characteristics as linear, nonlinear, balanced and unbalanced loads. The problem is discussed in detail and a specific design procedure for the active split dc-bus is presented.

Voltage-source inverters in microgrid stand-alone applications need the accurate control of the filter output voltages. Most of the time, supplied loads are not constant along daily based period; they change the total power consumption, the power factor, the current harmonic content of nonlinear equipment, and the power balancing among the phases. Consequently, the control dynamic behavior is affected, and the effectiveness of harmonic controllers is reduced. In this paper, the very sinusoidal shaping of inverter output voltage is achieved through a multiple resonant controller, which is adaptively tuned online according to the output power filter and load behavior. Resonant compensators are placed in the loop to achieve low total harmonic distortion of output voltage waveforms.

As required by international standards, the distributed energy generators, connected to the grid by an inverter, have to detect an islanding condition within a suitable time interval. In this paper, a phase-locked loop (PLL), based on a third-order prediction-correction filter, is proposed to implement an islanding detector with reduced detection time. Such a feature is obtained using the estimation of the grid angular frequency and acceleration provided by the PLL with a negligible time delay. The proposed approach is implemented on an industrial grade DSP and validated through the experimental comparison among different detection methods, such as rate of change of frequency (ROCOF) and Slip Mode frequency Shift (SMS). The combined use of ROCOF and SMS is also illustrated and discussed.

Grid-connected and stand-alone applications need inverters that accurately control the filter output voltages. Nowadays, resonant controllers (RCs) represent a good tradeoff between implementation issues and control performances. However, the unknown nature and size of supplied loads affect the dynamic control behavior and may reduce the effectiveness of the control system. A new structure of RC is introduced with the purpose of providing three degrees of freedom and allowing us to select independently the controller’s magnitude, width, and phase. In order to fully take advantage of the proposed RC form, the online tuning procedure based on the output filter analysis is also proposed. According to that, several resonant compensators can be placed in the loop to achieve a better output voltage waveform (i.e., lower total harmonic distortion).

Three-phase four-leg inverters are commonly employed to supply unbalanced loads avoiding the use of a transformer. A suitable output filter is necessary to improve the harmonic content of the output voltage and current waveforms. To this aim, different kind of filters, presenting both symmetrical or asymmetrical phase-to-phase and phase-to-neutral behavior, can be used. The paper discusses and analyses the effects of different carrier-based modulation techniques, based on the independent control of the fourth-leg from that of the other phases, with reference to the different filter topologies. The comparison, performed through extensive MATLAB/Simscape simulations and experimental tests, have demonstrated that in case of an asymmetrical filter utilization, such as an Active Split DC-Bus (also known as Split-Link), the conventional continuous and discontinuous Space-Vector modulations produce a significant harmonic content in the output voltages. A detailed problem description and a specific design procedure is presented and validated by experimental tests, allowing a substantial improvement of the inverter performances.

Both the static and the dynamic end effects of the linear induction motor (LIM) cause an increase of the difficulty in controlling with good performance the main electro-mechanical characteristics of the machine, in particular the thrust and the linear speed. These effects are related to asymmetries in the inductor structure with respect to the rotating counterpart, in both the longitudinal and in the transversal direction. This paper propose a technique for the compensation of the LIM static end effects by a synchronous controller in conjunction with a neural adaptive filter, exploited to extract the inverse sequence current component from the inductor current signature. This permits to compensate the static-end effects and thus to cover the static asymmetries of the LIM with respect to the rotating induction motor (RIM).

This paper deals with the EMI (Electromagnetic Interference) issue related to the power electronic converters of a hybrid vehicle. The effects of different configurations of the power supply system that feeds all the power conversion units are investigated in order to minimize the overall conducted EMI. In particular, two different power supply distribution networks (centralized and distributed) and related EMI filters architectures have been analyzed and compared in terms of conducted emissions and EMI filter sizes and efficiency.

Electromagnetic interference (EMI) has become in recent years a great concern to deal with, in every kind of application and especially in power electronics. As a consequence of international and national standards requirements, the insertion of an EMI filter in power electronic converters assembly is today a common practise. In this paper the EMI characterization of the power conversion system that manages the power flows between the generation and storage units of a hybrid vehicle is presented. With the purpose of reducing conducted emissions and designing a proper EMI filter, the overall noise has been separated into its common mode (CM) and differential mode (DM) components. Experimental results underline the importance of considering CM and DM noises separately in order to minimize the filter components count and size.

The use of voltage source inverters (VSI) in microgrid stand-alone applications needs the accurate control of the filter output voltages. Resonant controllers represent a good tradeoff between implementation issues and control performance. However, the un-known nature and size of connected loads, from the control point of view, affect the dynamic control behavior and may reduce the effectiveness of the harmonic controllers. The proposed multiple resonant controller is on-line tuned according to the designed filter and load behavior which may change during the converter operation. In this manner, several harmonic compensators can be placed in the loop to achieve better output voltage waveform, (i.e. lower total harmonic distortion).

Single-phase voltage source inverters are commonly used for interfacing small distributed generation units to the grid. Requirement in terms of THD, reliability and power factor correction is getting tightening day by day, so it is necessary to implement efficient and reliable algorithms that are capable to adapt control parameters to the varying grid operating conditions, in order to assure low current distortion, as well high efficiency and reliability. This paper deals with the investigation of both configuration topology and control algorithm with the purpose of achieving high performance in terms of efficiency as well THD of the output current, despite the non ideal grid characteristics. The proposed inverter uses a second order generalized integrator based phase-locked loop, a single loop current control based on a grid frequency tuned proportional-resonant controller, and a damped output LC filter in cascade with a switching frequency tuned LC-series trap-filter.

""This paper deals with an adaptive control method to be used in the dc-link voltage controller of diesel-electric power supply units being arranged with both permanent magnet generator and three-phase boost-rectifier. The proposed control method is based on the suitable manipulation of the overall system transfer function to achieve an adaptive closed form expression for the parameters of the dc-link voltage controller. Such an approach allows overcoming the inherent non-linear behavior of the system, so that the overall system dynamic performance is not affected by the system load conditions. Moreover, a current feed-forward technique based on the estimation of the load current will be introduced to improve the control stiffness to load unbalances.""

This paper presents a complete design methodology for the sensorless vector control of permanent magnet synchronous machine (PMSM) motor drives in fan-type applications. The proposed strategy is built over a linear asymptotic state observer used to estimate the PMSM back-EMF, and a novel tracking controller based on a phase-locked loop (PLL) system, which by synchronizing the estimated and actual d-q frames estimates the rotor speed and position. The paper presents the complete derivation of all associated control-loops, namely state observer, tracking controller, d-q axes current regulator, speed controller, an anti-saturation control loop - which provides inherent operation in the flux-weakening region, and all corresponding anti-windup loops. Detailed design rules are provided for each of these loops, respectively verified through time-domain simulations, frequency-response analysis, and experimental results using a 300 Vdc 3.5 kW PMSM PWM motor-drive, validating both the design methodology and the excellent performance attained by the proposed control strategy

Recognition of technical specification for electric drives thermal sizing is quite challenging in both transportation applications and distributed generating systems. In particular, components of electric drives for hybrid farm equipment, like tractors as well large combines, are often overstressed because of either high ambient temperature or overload working condition. The electric drive components are mainly chosen with reference to average power and energy considerations; however, overload operating conditions can occur for time durations not established in advance, as well quite unpredictable thermal conditions are associated to environment related to diesel engines in hybrid configurations. For this reason, it is convenient to provide the control system of the electric drive unit with a sort of thermal regulation of the electric and electronic equipment, which can be achieved by adjusting some operating conditions, as dc-link voltage, modulation strategy and switching frequency, in order to reduce thermal stress on either power electronic devices or electrical machines windings.

The paper deals with a fast and reliable algorithm for position and speed estimation using low-resolution position sensors in PMSM driven boost-rectifier for distributed generating units. The proposed algorithm does not require a high performance control platform making it suitable for industrial implementation. Different solutions have been analyzed considering also the available peripherals in mixed-signal Digital Signal Processors and Microcontrollers.

Conventional PV power converters are usually based on configurations that include DC-DC boost converter stage, voltage source inverter stage, passive filter stage and grid interface isolation transformer, whose presence implies less efficiency and higher costs, in particular for medium size power PV converters. Thereby, many transformer-less applications have been proposed in the last years and in particular the Neutral Point Clamped inverter is growing as a reference topology. In this paper a transformer-less NPC inverter prototype is described, some discussions on control strategy and phase-locked loop approach are provided; complete experimental results are also given.

The proposed study intends to investigate the benefits of using multilevel converters in low-voltage fed applications at high fundamental frequency as high-speed electric drives, with particular reference to the energy recovering from gas exhausts in automotive systems. Comparison of a conventional two-level sine wave converter and a neutral point clamped (NPC) converter has been developed and investigated. Experimental results have been achieved by means of a prototypal realisation of a three-phase three-level NPC converter.