Fiche individuelle

The future multi-terminal direct-current (MTDC) grid will require the interconnection of point-to-point high-voltage (HV) DC links with different specifications such as DC voltage level, system grounding configuration and HVDC technology. To adapt these differences, it is obligatory for DC/DC converters to interconnect HVDC links. Additionally, they are capable of providing supplementary functionalities as they are highly controllable devices. In this article, a primary virtual resistance DC voltage controller associated with DC/DC converter is proposed for managing DC grid voltages of the interconnected HVDC grids, increasing the reliability of the system. The commonly known topology, Front-to-Front Modular Multilevel Converter (F2F-MMC) is adopted for DC/DC converter. Time-domain simulations are performed using EMTP software for validating the controller behaviour under power disturbances and large events of loss of one converter in a MMC-based MTDC system. The converters are modelled using reduced order modelling (ROM) methodology. Apart from this, dynamic studies have been carried out using a linear state space model for small-signal stability analysis of a HVDC system integrating DC/DC converter with a virtual resistance DC voltage controller. The results are examined through parametric sensitivity analysis.

Energy management in modular type converters constitutes a key aspect of their operational stability. This paper introduces a full energy management structure for the Extended Overlap-Alternate Arm Converter (EO-AAC) ensuring both equal energy distribution across all six stacks and the maintaining of ripple-free DC current during steady state. The performance of the control structure against active power step events is validated by detailed simulations using EMTP-RV software. Moreover, the full energy management allows the EO-AAC to have an equivalent controllability to Modular Multilevel Converter (MMC) through only two conducting stacks in overlap mode ensuring the power balance. From this observation, it is demonstrated that the use of control strategy like the virtual capacitor to support HVDC system, originally designed for MMC, is possible. Gathering all these controllers leads to a general conclusion which is the dynamic equivalence between EO-AAC and MMC.

This paper presents an evolution of control systems of Modular Multilevel Converters (MMCs) focusing on the internal voltages and currents dynamics. MMCs have passive components that create extra dynamics compared to conventional VSCs. Some control schemes that do not consider these internal dynamics may still stabilize the system asymptotically thanks to the linearisation in the modulation step. However these control schemes are less robust because they are prone to poor damped oscillations on the dc side of the converter. The MMC circuit and energy relationships are presented in this paper. Along with a gradual development of the energy based control, the important roles of each internal dynamics are clearly demonstrated. Experimental results are presented to show the impacts of the linearisation in the modulation step on the system behaviour.

Modular Multilevel Converter (MMC) is an established technology for HVDC or Multi-Terminal DC (MTDC) systems, due to its advantages over classical Voltage Source Converters (VSCs) such as two or three level VSCs. To achieve a full control of all state variables, it is essential to implement energy-based method in which a cascade control loop is employed to regulate all state variables including ac and differential currents, and stored energy within MMC arms. In addition to normal operation control, dc Fault Ride Through (DC-FRT) capability of the MMC is a crucial and challenging control issue especially for overhead line HVDC system where non-permanent dc fault occurrence is statistically more probable. Furthermore, the main problematic technical obstacle to develop HVDC/MTDC grids is the lack of mature dc fault protection. Since conventional control in normal operation cannot be employed in case of dc fault, an efficient control strategy is indispensable. The principal objectives of the novel control methodology are to (i) obtain DC-FRT capability, (ii) decay short circuit current to zero, (iii) secure the MMC through leg and arm energy balancing, (iv) support ac grid as a Static Synchronous Compensator (STATCOM) and (v) resume normal operation after dc fault clearance. The simulation results verify the validity of proposed control strategy to fulfill the abovementioned objectives in dc fault operation of the hybrid MMC.

This paper presents the modelling of the DC-DC Multilevel Modular Converter (DC-DC MMC) with half-bridge Sub-Modules (SM) and the control based on the inversion of its model. The DC-DC MMC structure presents many advantages such as its modularity, the absence of capacitors on the high DC bus voltage and a very low switching frequency due to the large number of SMs. This topology also preserves the intrinsic disadvantages of the MMC as the complexity of modelling and controlling due to the large number of semiconductors and state variables to control. The control of this converter cannot be symmetrical due to the interconnection of the two parts by an internal AC grid. The control strategy of one part of the DC-DC MMC uses the conventional control scheme with currents controls and stored energy control. The second one uses the energy control and produces the waveform of the three-phase internal AC bus voltage linking the two parts of the converter. The explicit control for the generation of internal AC voltages guarantees the correct operation of the converter even in a critical DC voltage dip on one or the other DC buses. Thus, it avoids the need of a DC circuit breaker or the use of full bridge MMC sub-modules. The validity of the proposed control is verified by simulation using Matlab-Simulink.

The M2DC exploits the interleaving between the three legs of an MMC to realize a promising uninsulated DC/DC converter to interconnect HVDC grids. This paper details a current and energies decoupled model of the M2DC. The major idea proposed in this paper is focused on the full energy control generating optimal current references to minimize the internal currents magnitude. The energy sum and difference models are fully detailled. Both current and energy control loops are based on the model inversion principle in order to control all the state variables. The proposed control is based a dynamic control developed with the model inversion principle associated on an optimization of the current magnitude deduced from a quasi static analysis. All dynamics of the system are then explicitly controlled, which guarantee a good dynamic behavior during the transient. Therefore, current and energy controls are presented in details. Simulation results show the dynamic behavior of the converter for various operating points.

Grid-forming inverters usually use inner cascaded controllers in order to regulate the output AC voltage and the converter output current. Yet, at the power transmission system level where the power inverter bandwidth is limited (i.e.; low switching frequency), it is difficult to tune its controller parameters to achieve the desired performances because of the control loops interactions. In this paper, a direct AC voltage control based state-feedback control is applied. Its control gains are tuned using the linear quadratic regulator. In this paper, a sensitivity analysis is proposed in order to choose the right cost factors that allow the system to achieve the imposed specifications. Conventionally, the system based on direct AC voltage control has no restriction on the inverter current. Hence, in this paper, threshold virtual impedance has been added to the state-feedback control in order to protect the inverter against overcurrent. The robustness of the proposed control is assessed for different short circuit ratio using small signal stability analysis, then, checked in different grid topologies using time domain simulations. An experimental test bench is developed in order to validate the proposed control.

The present paper deals with the post-fault synchronization of a voltage source converter based on the droop control. In case of large disturbances on the grid, the current is limited via current limitation algorithms such as the virtual impedance. During the fault, the power converter internal frequency deviates resulting in a converter angle divergence. Thereby, the system may lose the synchronism after fault clearing and which may lead to instability. Hence, this paper proposes a theoretical approach to explain the dynamic behavior of the grid forming converter subject to a three phase bolted fault. A literal expression of the critical clearing time is defined. Due to the precise analysis of the phenomenon, a simple algorithm can be derived to enhance the transient stability. It is based on adaptive gain included in the droop control. These objectives have been achieved with no external information and without switching from one control to the other. To prove the effectiveness of the developed control, experimental test cases have been performed in different faulted conditions.

The dq impedance stability analysis for a grid-connected current-control inverter is based on the impedance ratio matrix. However, the coupled matrix brings the difficulties to derive its eigenvalues for the analysis based on the General Nyquist Criterion. If the couplings are ignored for simplification, the unacceptable errors will present in the analysis. In this paper, the influence of the couplings on the dq impedance stability analysis is studied. For taking the couplings into account simply, the determinant-based impedance stability analysis is used. The mechanism between the determinant of the impedance-ratio matrix and the inverter stability is unveiled. Compared to the eigenvalues-based analysis, only one determinant rather than two eigenvalue s-function is required for the stability analysis. One Nyquist plot or polemap can be applied to the determinant for checking the right-half-plane poles. The accuracy of the determinant-based stability analysis is also checked by comparing with the state-space stability analysis method. For the stability analysis, the coupling influence on the current control, phase-locked loop and the grid impedance are studied. The errors can be 10% in the stability analysis if the couplings are ignored.

This article deals with DC voltage dynamics of Multi-Terminal HVDC grids with energy-based controlled Mo\-dular Multilevel Converters (MMC) adopting the commonly used power-voltage droop control technique for power flow dispatch. Special focus is given on the energy management strategies of the MMCs and their ability to influence on the DC voltage dynamics. First, it is shown that decoupling the MMC energy from the DC side by controlling the energy to a fixed value, regardless of the DC voltage level, causes large and undesired DC voltage transient after a sudden power flow change. Second, the Virtual Capacitor Control technique is implemented in order to improve the results, however, its limitations on droop-based MTDC grids are highlighted. Finally, a novel energy management approach is proposed to improve the performance of the later method. These studies are performed with detailed MMC models suitable for the use of linear analysis techniques. The derived MTDC models are validated against time-domain simulations using detailed EMT MMC models with 400 sub-modules per arm.

The Modular Multilevel Converter (MMC) is a power electronic structure used for high voltage adjustable speed drives applications as well as power transmission applications and high-voltage direct current (HVDC). MMC structure presents many advantages such as modularity, the absence of a high voltage DC bus and very low switching frequency. It presents also some disadvantages such as modeling complexity and control due to the large number of semiconductors to control. The objectives of this paper are to present the methodology to design a laboratory MMC converter and its control system. This methodology is based on an intensive used of real-time simulation, to develop and test the control algorithm is proposed. This MMC prototype must be as realistic as possible to a full scale MMC, with a large number of SM (i.e. 640kV on the DC side, a rated power of 1GW and 400 sub-modules). A control hardware integrating distributed processors (one for each arm) and a master control is presented. The protocols to validate sub-modules, arms and the converter are explained.

The modular multilevel converter (MMC) is be- coming a promising converter technology for HVDC transmission systems. Contrary to the conventional two- or three-level VSC-HVDC links, no capacitors are connected directly on the dc bus in an MMC-HVDC link. Therefore, in such an HVDC link, the dc bus voltage may be much more volatile than in a conventional VSC-HVDC link. In this paper, a connection between the dc bus voltage level and the stored energy inside the MMC is proposed in order to greatly improve the dynamic behavior in case of transients. EMT simulation results illustrate this interesting property on an HVDC link study case.

Today, industry has not fully embraced the matrix converter solution. One important reason is its high control complexity. It is therefore relevant to propose a simpler but efficient modulation scheme, similar as three phase VSI modulators with the well-known symmetrical carrier-based ones. The modulation presented in this paper is equivalent to a particular Space Vector Modulation (SVM) and takes into account harmonics and unbalanced input voltages, with the same maximum voltage transfer ratio (86%). The aim of this work is to propose a simple and general pulse-width-modulation method using carrier-based modulator for an easier matrix converter control. Furthermore, a simple duty cycle calculation method is used, based on a virtual matrix converter. Finally, simulations and experimentations are presented to validate this simple, original and efficient modulation concept equivalent to matrix converter SVM.

Turn-on performance of a reverse blocking insulated gate bipolar transistor (RB-IGBT) is discussed in this paper. The RB-IGBT is a specially designed insulated gate bipolar transistor (IGBT) having ability to sustain blocking voltage of the both polarities. Such a switch shows superior conduction but much worst switching (turn on) performances compared to a combination of an ordinary IGBT and blocking diode. Because of that, optimization of the switching performance is a key issue that makes the RB-IGB not well accepted in the real applications. In this paper the RB-IGBT turn-on losses and reverse recovery current are analysed for different gate driver techniques and a new gate driver is proposed. Commonly used conventional gate drivers do not have capability for the switching dynamics optimization. In contrast to this, the new proposed gate driver provides robust and simple way to control and optimize the reverse recovery current and turn-on losses. The collector current slope and reverse recovery current are controlled by the means of the gate emitter voltage control in feed-forward manner. In addition, the collector emitter voltage slope is controlled during the voltage falling phase by the means of inherent increase of the gate current. Therefore, the collector emitter voltage tail and the total turn on losses are reduced, independently on the reverse recovery current. The proposed gate driver was experimentally verified and the results presented and discussed.

This paper presents an MTDC test case integrating a DC/DC converter where the converter is working with a DC voltage controller and participating in the DC voltage management system. The influence of voltage-controlled DC/DC converter is studied by introducing power disturbances in the MTDC system.

The development of multi-terminal DC (MTDC) networks has various challenges as interconnecting grids of different voltages and grounding schemes, DC grid protection and power flow. DC/DC converter has emerged as the solution for interconnecting HVDC links with different specifications. In this paper, the Front-to-Front Modular Multilevel Converter (F2F-MMC) topology is adopted for DC/DC converter, which can act as a firewall between the healthy and faulty grid during DC faults. Along with this, DC/DC converters when operated in DC voltage control mode can provide supplementary functionalities such as participating in DC grid voltage management and increasing the reliability of the system. In this study, the F2F-MMC converter is operated with a virtual resistance DC voltage controller integrated into an MTDC system. A pole-to-pole DC fault is applied on the MTDC grid and the influence of the virtual resistance controller associated with the DC/DC converter is studied for re-establishing the power flow after DC faults.

The Modular Multi-Level DC-DC Converter (M2DC) is an attractive non-isolated DC-DC converter topology for HVDC grid. In order to carry out MTDC grid stability studies, the development of reduce order models of converters is necessary. This article first presents the M2DC converter. Then, the reduce order model will be developed in the second part. The development of the control of this model will be carried out in the third part. Atlast, the comparison of the reduce order model and its control with the average arm model will be performed in the later section of the paper.

The Modular Multilevel DC Converter is an attractive non-isolated topology to interconnect High Voltage DC Links. This paper presents the interaction among control, component design and efficiency of this converter. The impact of the two degrees of freedom on the design and the efficiency is analyzed.

This work focuses on the steady-state analysis of three types of MMC based STATCOM. For a given STATCOM rating, Double-Star Full Bridge, Single-Star Full Bridge and Single-Delta Full Bridge have been compared in terms of design and losses. In this approach, the number of submodules is chosen according to the voltage and current ratings of semiconductor devices while the submodule capacitor value is obtained by following an energy storage criterion to maintain the submodule voltages within an acceptable voltage range.

Modular Multilevel Converters for High Voltage Direct Current (MMC_HVDC) is a hopeful technology for future smart grid and offshore applications. This modular configuration allow to output almost an ideal sinusoidal voltage which permit to prevent the harmonic injected to the power system; therefore, there is no need to use a large filters in the system. Furthermore, its allows to support a high voltage rating. Thus, it is appropriated for High Voltage Direct Current (HVDC) transmission systems applications. This paper studies the operation of a point to point MMC HVDCsystem simulated with its associated loop controller.

The question of grid forming control is very different depending on the connection to a low voltage or high voltage grid. In case of higher power application, the low switching frequency may induce some stability issues. This question has been studied and some solutions have been proposed through new inner current and voltage control tuning methods. However, the possible interactions between the inner and the outer controls have not been discussed yet. Actually, in large power system, the phasor modeling approximation is used in order to ease the analysis and reduce the time computations. It assumes a good decoupling between the control loops, which allows neglecting the inner loop dynamics. This paper investigates the effectiveness of this assumption by taking some examples of tuning methods proposed in the literature and showing the ability of each method to guarantee the decoupling between controllers. In this paper, small-signal analysis tool, participation factors and parametric sensitivities are used.

The MMC is the solution today to connect HVDC grids to the current HVAC grid. This paper proposes to evaluate the impact of Zero Sequence Voltage Injection variants, which until now, have not been extensively studied. Such techniques can involve, for example, a reduction of the SM capacitor value, the number of SM requirement and converter losses. The paper presents MMC current model, control and highlights the implication of the zero-sequence voltage. Grid current control structure with the introduction of the zero-sequence voltage is presented in different techniques. These modulation schemes are compared through two main quantities in MMC, the energy requirement defining the SM capacitance value and the power losses.

Because of the throng of control strategies based Voltage Source Converters (VSC) recently proposed in the literature; their classification and characterization are becoming a trending topic. The high similarities of the proposed control strategies may lead to confusions and a misunderstanding of vocabulary. Therefore, this paper seeks first to highlight the possible features fulfilled by power converters in a large power system. The combination of these features is used to classify power converters. Furthermore, power converters can be seen by a power transmission system operators as black boxes, and they may have the same inputs and outputs, which makes their characterizations more difficult. This paper looks to show that only the fundamental nature of the source has an influence on the system dynamic behavior, thus, power converter can be characterized from their transient behavior in response to grid disturbances.

The modular multilevel converter (MMC) is recognized for many advantages over conventional 2-level or 3-level voltage source converters (VSC). One of the features is that it is possible to control the stored energy in the distributed submodule (SM) capacitors. The aim of this paper is to propose a methodology to develop the control system implemented on a 21-level MMC mock-up. This control must be first validated in a real-time (RT) simulation environment with an accurate MMC model representing the behaviour of the physical MMC mock-up. Then it is performed in the hardware-in-the-loop (HIL) simulation before the implementation. The power system is emulated in a real time simulator from Opal-RT Technologies, which is connected to an external control system programmed in a dual-core microcontroller. The developed HIL platform facilitates the tests of the control and protection system in all possible scenarios.

This paper deals with the conception and the development of a detailed EMT Model for MMC based on experimental results obtained from a mock-up. The main purpose is to illustrate how to exploit the performances of EMT simulation tools to develop a detailed model that represents accurately the behaviour of a physical MMC. According to step-by-step identification of the MMC element parameters, the idea is to perform a systematic method, which allows expanding an accurate EMT model considering the behaviour of the prototype and its environment. The first part depicts the MMC topology and the modelling approach of the Half-bridge Sub-Module (SM) using a detailed IGBT-based model. The second part of the simulation model conception concerns both control levels such as high-level and low-level controllers. The last part of the EMT model conception involves the modelling of measurement process, ADC (Analogue Digital Converter), sensors dynamics, the communication delays and especially the quantization effect. Finally, the obtained results from the final detailed EMT model is compared to the experimental behaviour for different active and reactive powers operating points in order to prove the effectiveness and the capability of the EMT modelling to reach a detailed and accurate model.

In the near future, power converters will be massively introduced in transmission grids due to renewable energy sources and high voltage direct current (HVDC) increase. Voltage Source Converter (VSC) control laws assume that Synchronous Generators (SGs) build a stiff AC voltage which allows the synchronization of converters. This is one of the major reasons that limit the high integration of currentsource converters in transmission grid. This constraint is no longer relevant when power converters operate as a voltage source based on the grid-forming concept. This concept uses an inner cascaded PI controllers in order to regulate the output AC voltage. However, it is difficult to tune its controller parameters for stable operation in grid-connected mode. This paper proposes an alternative state-feedback control with integral compensator based linear quadratic regulation (LQR) in order to ensure a stable operation and to get a better AC voltage transient and good decoupling between reactive and active power. The proposed control will be fully analyzed and compared to conventional methods.

The virtual synchronous machine concept (VSM) has been developed initially to reproduce the synchronous machine stabilizing effect by providing inertia with the emulation of swing equation, whereas droop control is developed initially to ensure load sharing and has no inertia. An introduction of a low pass filter to droop control has been motivated to filter the active power measurement and ensures a time decoupling with the inner control loops, whereas, this low-pass filter can also provide inertia to the system. This functionality is limited due to its negative impact on the active power dynamic. This paper proposes an analysis of the conventional droop control by showing its limitations and proposes an improved inertial droop control that allows providing the inertia to the system and ensures a good dynamic behavior of the active power at once in simple manner, and without modifying the load sharing capability. The results obtained are compared to the conventional method (Droop control and VSM) in various topologies in order to show the relevance of the proposed method.

The use of DC transmission is particularly advantageous for long-distance transmission and interconnection of asynchronous AC networks. Several converter topologies can be used for HVDC. Multilevel Modular Converters (MMCs) are the most favored given their technological advantages over other converters topologies. Due to their industrial maturity, they have become essential for all AC / DC conversion. So far, they have always been studied with a voltage source on DC side. However, when the converter is equipped with DC breaker, a series inductor is associated to limit current variations. This has consequences in terms of modeling and control determination. This article aims to propose a modification of the control law in order to take into account this inductor. To facilitate the control organization, the Energetic Macroscopic Representation (EMR) is used.

From the origin of the grid, energy has been delivered to electrical loads mainly by synchronous machines. All the main rules to manage the grid have been based on the electromechanical behavior of these machines which have been extensively studied for many years. Due to the increase of HVDC link and renewable energy sources as wind turbine and PV, power converters are massively introduced in the grid with a fundamentally different dynamic behavior. Some years ago, they were connected as simple power injector. Then, they were asked to provide some ancillary services to the grid, in the future, grid forming capability will be required. Even if gridforming converters had been extensively studied for microgrids and offshore grids, it has to be adapted to transmission grid where the topology may be largely modified. This paper presents an algorithm for calculating the controller parameters of a gridforming converter which guarantee a stable behavior for many different configurations of the grid.

The interconnections of offshore wind farms have raised the interest of Multi-Terminal DC (MTDC) grids with voltage source converters, specifically with the Modular Multilevel Converter (MMC) topology. For controlling the DC bus voltage, the droop control strategy proves to be one of the most effective since it allows a shared effort between the different converters connected to the DC grid. Furthermore, these MTDC grids may likely result in multivendor schemes, where each converter could have different control strategies among them. Specially on the way that the internal energy of the MMC is managed, which may have an important impact on the DC voltage dynamics. This paper proposes a simplified model to represent the dynamic behavior of the DC grid including converters which highlights the droop gain and the energy management of the converter as main influential parameters. The performance of this model was assessed by comparisons with EMT simulations on a representative case study with different parameters.

DC/DC converters are necessary in HVDC networks to adapt different voltage levels, different configurations or to participate in power flow control. This paper focuses on a recent DC/DC converter topology, called Modular Multilevel DC Converter (M2DC), which uses direct DC/DC connection without galvanic separations. A general analysis of the topology is firstly discussed. Afterwards a decoupled mathematical model, reducing the control complexity is proposed. To improve converter’s performances and limit AC constraints on each leg, parameters adjust and inductor sizing are explored. Finally, simulation results are presented to validate the proposed decoupled model and the control. The impact of optimized parameters are shown on AC components values.

The modular multilevel converter (MMC) is the most accepted solution to connect a HVDC grid to an AC transmission grid. The Alternate Arm Converter (AAC) is another promising structure since it allows a DC short-circuits blocking capability similarly to the Full Bridge MMC while having a small impact on the power losses. Its footprint is smaller to the MMC since the needed number of modules is closer to 50% and the SM Capacitors are about three times smaller. The AAC is a hybrid structure between a 2 level VSC converter and an MMC one. Elements hampering the development of the AAC are its complexity to model and control, in particular the opening procedure of the director switches (DS) since these DS are directly connected in series to the arm inductance. This paper proposes a fast method to control the opening of the DS at zero current. The first part is focused on the instantaneous model and current control of the converter and AAC. The second part is focused on the opening method of the DS without generating overvoltage in the converter and taking into account the technical parameters of the various elements of the AAC. Finally, simulation results validate the DS opening control of the AAC converter.

The modular multilevel converter (MMC) is one of the available solution to connect a HVDC grid to an AC transmission one. The Alternate Arm Converter (AAC) is a promising structure since it allows a DC short-circuit blocking capability as the Full Bridge MMC while having equivalent losses to the half-bridge MMC. Its footprint is smaller than the MMC since the needed number of modules is closer to 50% and the SM Capacitors are about three times smaller. The AAC is a hybrid structure between a 2 level VSC converter and an MMC one. The main drawback of the AAC is the complexity of its control. This paper presents first, the instantaneous model and current control of the AAC. The second part is focused on the energetic model of the AAC and its control. Finally, simulation results validate the quality of the proposed control.

Modular multilevel converters (MMCs) offer several advantages, such as high scalability and power quality, which are convenient for high-voltage DC transmission systems. The MMC structure contains high number of sub-modules (SMs) per phase that imposes high requirements on the control system. Therefore, it can be implemented in two control levels: the control of the switches that includes the balance of hundreds of voltage on the elementary SMs, the higher level control whose aim is to control the currents, power and energy in the system. This paper focusses on the latter control, which reveals interesting challenges in the management of the MMC stored energy. To illustrate this, the present paper proposes an overview of the different ways of controling the MMC energy. The impact of these energy based control strategies are verifierd by EMTPRV time domain simulation and their comparison highlights that the chosen control may have large influence on the MMC dynamics, MMC losses and exchange of the energy between MMC and DC link.

The Modular Multilevel Converter (MMC) has enhanced the feasibility of Multi-Terminal DC grids (MTDC). For controlling the DC bus voltage in the MTDC grids, the droop control is the most promised technique. This paper evaluates the dynamic impact of the way of controlling the MMC on the MTDC grids. Results are compared with a simplified model that highlights the key elements for the dynamic behavior of the DC bus voltage, the droop parameter and the equivalent DC bus capacitor.

This paper proposes an advanced control strategy for Modular Multilevel Converters (MMC) integrated in Multiterminal DC grid. In this present work, a three terminal MMC-MTDC system connecting onshore AC systems with an offshore wind farm is setup. Firstly, the voltage droop control associated to the conventional cascaded controllers for MMC stations is studied, the dynamic behavior of the DC voltage is analyzed and some drawbacks are outlined. In order to improve the dynamic behavior of the controlled DC bus voltage and the stability of MTDC system, an optimal multivariable control strategy of each MMC converter is proposed and integrated in a voltage droop controller strategy. The designed advanced controller allows to improve the overall DC grid stability and to reach the droop values designed on static considerations with acceptable dynamic behavior. By means of numerical simulations in EMTP-RV software, it is shown that the proposed control strategy performs well the stability of MTDC grid with 400- level model for MMC compared with the classic existing control methods.

The Modular Multilevel Converter (MMC) represents the recent development among the diverse available topologies of VSC and is allegedly the most suitable solution for converters in HVDC transmissions. This paper presents an Average Value Model (AVM) of the MMC that still includes the characteristic internal dynamics that are non-existent in traditional 2-level VSCs. This AVM and its control are described and linearized in order to obtain a state-space model of the MMC that can easily be used as a subsystem for multi-terminal HVDC (MTDC) grids. A case study showing a 401-level MMC-based HVDC link simulated in the EMTP-RV software validates the proposed state-space representation of the MMC.

This study presents control strategies of offshore wind farms with dc series collector systems and modular multilevel converter based high voltage direct current (MMC-HVDC) transmissions. Series connection of dc wind turbines enables the collector system to establish the HVDC transmission voltage level with neither transformer nor centralized converter platform, indicating huge savings on offshore wind farm investment. However, severe wind variation imposes overvoltage on the wind turbines and may induce a cascade of failures of other series connected units. This paper first explains the wind farm design constrains bound by overvoltage limitation of wind turbine units. Then two kinds of control strategies, classical control and proposed HVDC global control, are presented to ensure every unit operating within safety limit. Wind farms with both control strategies are simulated. Finally, the two kinds of strategies are compared in terms of technical feasibilities, economical benefits and annual energy production rates. The results demonstrate that DC series offshore wind farms with proposed HVDC global control strategy imply competitive economic advantages of future integration of remote energy generation.

to ensure the operation of each wind turbine under a safe voltage level while no power production is lost in the wind farm. This paper first introduces the whole wind generation system. Then the steady state analysis of the dc series offshore wind farm and the overvoltage phenomenon are described, which lead to the proposal of the global control strategy. The detailed realization of the global control strategy by use of the MMC is given after a short recall of the MMC arm average model.

The modular multilevel converter (MMC) is the most promising solution to connect HVDC grids to an HVAC one. The installation of new equipment in the HVDC transmission systems requires an economic study where the power losses play an important role. Since the MMC is composed of a high number of semiconductors elements, the losses estimation becomes complex. This paper proposes a simulation-based method for the losses estimation that combines the MMC averaged and instantaneous model in a modular way. The method brings the possibility to compare performances for different modules technologies as well as different high and low level control techniques. The losses characteristics within the MMC are also discussed. The passive losses are taken into account for the first time.

This paper presents an innovative Low Losses Modulation of the Matrix Converter. Its principle is to minimize the cumulated switched voltage in each cell by increasing and decreasing progressively the output potential of each matrix cell. This modulation can be applied to all conversion matrices previously proposed for the matrix converter modulation and can be easily implemented with a triangular carrier based modulator adapted for matrix converter cells. The proposed modulation involves the use of all vectors present in the matrix converter space vector approach, including the 6 rotating vectors. Simulations verify the losses decreasing, but also reveal a significant reduction of the output voltages THD, and a very slightly increasing of the input currents THD.

Modular Multilevel Converters (MMC) are becoming increasingly popular with the development of HVDC connection and, in the future, Multi Terminal DC grid. A lot of publications have been published about this topology these last years since it was first proposed. Many of them deal with converter control methods, other address the method of estimating losses. Usually, the proposed losses estimation techniques are associated to simple control methods For VSC (Voltage Sources Converters) topology, the losses minimization is based on the limitation of the RMS currents values. This hypothesis is usually extended to the control of MMC, by limiting the differential currents to their DC component, without really being checked.

Modular Multilevel Converters (MMC) are becoming increasingly popular with the development of HVDC connection and, in the future, Multi Terminal DC grid. A lot of publications have been published about this topology these last years since it was first proposed. Few of them are addressing explicitly the two different roles that are held by this converter in a HVDC link: controlling the power or controlling the DC voltage level. Most of the time, the DC-bus voltage is supposed to be constant. In an HVDC link, this corresponds to the substation which controls the power. This paper addresses the cases when the voltage is regulated by the converter and presents the different ways of voltage control.

Modular Multilevel Converters are becoming increasingly popular with the development of HVDC connection and, in the future, Multi Terminal DC grid. A lot of publications have been published about this topology these last years since it was first proposed. Few of them are addressing explicitly the 2 different roles that are held by this converter in a HVDC link: controlling the power or controlling the DC voltage level. Moreover, for a given function, different ways of controlling this converter may be considered. This paper proposes an overview of the different solutions for controlling the MMC and proposes a methodology to synthesize the control architecture.

This papers deals with the Modular Multilevel Converter (MMC). This structure is a real breakthrough which allows transmitting huge amount of power in DC link. In the last ten years, lots of papers have been written but most of them study some intuitive control algorithms. This paper proposes a formal analysis of MMC model which leads to the design of a control algorithm thanks to the inversion of the model. The Energetic Macroscopic Representation is used for achieving this goal. All the states variables are controlled to manage the energy of the system, avoid some instable operational points and determine clearly all the dynamics of the different loops of the system.

Matrix (MC) and indirect matrix (IMC) converters are direct three-phase to three-phase power converters providing variable frequency and amplitude control of their output voltage. These converters are compact solutions which can be used on industrial adjustable speed drive applications for induction motors. This paper deals with the comparison of the matrix and indirect matrix converter silicon losses for classical industrial applications with constant RMS current load (similar to a constant motor torque). The indirect matrix converter control is extracted from the matrix converter modulation so as to ensure identical instantaneous modulation. Furthermore, the chosen modulation strategy reduces the IMC losses by allowing zero-current switching on the IMC rectifier stage. This paper presents a new reduced losses modulation adapted for indirect converter based on a modified matrix converter modulation which reduces switching voltage step level during a pulse width modulation period. The losses simulations shows that the power losses peak value is about 20% smaller for the matrix compared to the indirect converter. Hence, the matrix cooling system can be significantly reduced compared to the indirect one. The modified modulation increases the number of gate drivers required from six to twelve in the indirect converter rectifier side in comparison with the classical DPWM modulation, but allows to obtain a 14% decrease of its average losses compared to the classical modulation.

Matrix and two stage matrix converters are direct three-phase to three-phase power converters providing variable frequency and amplitude control of their output voltage. These converters are compact solutions that can be used on industrial adjustable speed drive applications for induction motors. The two stage matrix converter control is extracted from the matrix converter modulation in order to assure the same instantaneous modulation. This paper deals with the comparison of silicon losses for both converters used in classical industrial applications with constant RMS current load, similar to a constant motor torque. Results show that the direct solution is best for efficiency criteria.

This paper proposes a novel method to adapt classical matrix converter modulations to the indirect matrix converter one. This method allows splitting the 3x3 instantaneous matrix of the matrix converter into two instantaneous matrixes 3x2 controlling the rectifier and the inverter of the indirect converter. The proposed method is based on a complete analysis of the classical space vector modulations of the matrix converter. This method is validated by Matlab-Simulink® simulations.

Afin de réaliser des études de stabilité des réseaux MTDC, le développement de modèles d’ordre réduit des convertisseurs s’avère nécessaire. Cet article présente un modèle réduit du convertisseur DC/DC modulaire multiniveaux (M2DC) : une topologie de convertisseur DC-DC non isolée attrayante pour le réseau HVDC. Cet article présente d’abord le convertisseur M2DC. Dans un second temps, le modèle réduit sera développé. Le développement de la commande de ce modèle sera effectué dans la troisième partie. Enfin, la comparaison du modèle réduit et de son contrôle avec le modèle moyen des bras sera effectuée dans la dernière section de l’article.

Le convertisseur DC-DC Modulaire Multi-niveaux (M2DC) est une topologie attrayante de convertisseur DC-DC non isolée pour les réseaux haute tension à courant continu (HVDC). Cet article présente dans un premier temps, le modèle et la commande du M2DC. Une étude de l’utilisation des degrés de liberté phares est menée dans un second temps pour réaliser une analyse de leurs impacts sur le dimensionnement des éléments du convertisseur tels que les condensateurs de sous modules ainsi que sur les pertes liées aux semi-conducteurs

L'utilisation de la transmission DC est particulièrement avantageuse pour la transmission à longue distance et l'interconnexion des réseaux AC asynchrones. Plusieurs topologies de convertisseur peuvent être utilisées pour le HVDC. Les convertisseurs modulaires multiniveaux (MMC) sont les plus favorisés étant donné leurs avantages technologiques par rapport aux autres topologies de convertisseurs. Du fait de leur maturité industrielle, ils se sont imposés maintenant pour tous les convertisseurs AC/DC à transistors de forte puissance. Jusqu'ici, ils ont toujours été étudiés avec une source de tension côté DC. Or, lorsqu'ils sont équipés de DC breaker, on associe une inductance en série pour limiter les variations de courant. Ceci a des conséquences en terme de modélisation puis de détermination de la commande. Cet article a pour objectif de proposer une modification de commande afin de prendre en compte cette inductance.

Cet article présente la modélisation et la commande basée sur l’inversion de ce modèle, du Convertisseur Modulaire Multiniveaux DC-DC (MMC DC-DC) en demi-ponts. La structure DC-DC MMC présente beaucoup d’avantages tels que sa modularité, l’absence de condensateur sur le bus DC haute tension et une fréquence de commutation très faible étant donnée le grand nombre de SM. Elle conserve aussi les inconvénients intrinsèques du MMC comme la complexité de modélisation [2] et de contrôle [3] dû au grand nombre de semi-conducteurs et de variables d’état à contrôler. La stratégie de contrôle utilise le schéma de contrôle classique avec contrôle d'énergie et contrôle de puissance pour une partie du MMC DC-DC. La seconde partie utilise le contrôle d'énergie et génère la forme d’onde de la tension du bus triphasé AC liant les deux parties du convertisseur. Le contrôle explicite de la génération de l’onde de tension AC permet de garantir le bon fonctionnement du convertisseur même en cas de creux de tension DC critique sur l’un ou l’autre des bus DC et ainsi évite la nécessité de disjoncteur DC ou d’utiliser un MMC Full bridge. La validité du contrôle proposé est vérifiée par simulation à l’aide de Matlab-Simulink.

L’objectif de cet article est d’analyser le fonctionnement d’une nouvelle topologie de convertisseur DC/DC haute tension (DC/DC Modular Multilevel Converter) dédié au transport de l’énergie électrique HVDC (High Voltage Direct Current). Le contrôle du convertisseur est premièrement discuté au travers d’un modèle mathématique découplé. Le modèle proposé permet de réduire la complexité du contrôle du convertisseur. Afin d’améliorer les performances du convertisseur, le dimensionnement du convertisseur est exploré dans un second temps. Enfin, des résultats de simulation illustrent le fonctionnement et les performances du nouveau convertisseur DC/DC basé sur le modèle découplé proposé.

Le Convertisseur Modulaire Multiniveaux (MMC) est une structure d’électronique de puissance utilisée dans des applications de variation de vitesse des machines électrique haute tension mais aussi des applications de transport de l’électricité à très haute tension et courant continu. La structure MMC présente beaucoup d’avantages tels que sa modularité, l’absence de bus DC haute tension et une fréquence de commutation très faible étant donnée le grand nombre de SM. Elle présente aussi des inconvénients comme la complexité de modélisation [2] et de contrôle [3] dû au grand nombre de semi-conducteurs à contrôler. Cet article a pour objectifs de présenter le dimensionnement d’un convertisseur MMC de laboratoire ainsi que son système de contrôle, le plus réaliste possible d’une structure échelle une, avec un grand nombre de SM. Le dimensionnement de ce dernier prendra en compte les contraintes et les caractéristiques son fonctionnement dans un réseau HVDC 640kV 1GW. Une architecture de contrôle, contraintes par le nombre de de sous module, sera présentée. Les protocoles de validation des sous modules, d’un demi-bras puis du convertisseur seront présentés.

Le système étudié dans cet article est un convertisseur Modulaires Multi-Niveaux. Dans une première partie, l’utilisation de la Représentation Energétique Macroscopique (REM) permet de mettre en évidence les couplages importants qui existent au sein de ce système. L’inversion du modèle conduit à une architecture générale de la commande amenant à avoir autant de correcteurs que les variables d’état. Avec la solution proposée, il est possible de contrôler chaque tension de condensateur équivalent. Dans une seconde partie, la méthodologie permettant de déterminer le diagramme PQ du convertisseur MMC est présentée. Quelques points de fonctionnement aux limites de ce diagramme sont validés par simulation. La maitrise des dynamiques ainsi que la connaissance du diagramme PQ sont des étapes nécessaires pour l’intégration de ce type de convertisseur dans un réseau alternatif.

Cet article présente une amélioration des performances d’un convertisseur matriciel par utilisation de degrés de liberté naturellement accessible au niveau de la matrice de conversion. Ces améliorations sont réalisées à partir d’un modulateur simple et synthétique, basé sur l’introduction d’un convertisseur virtuel. On présente tout d’abord une méthode de généralisation de la matrice de conversion obtenue avec une modulation classique. Cette matrice est modifiée afin d’induire la modification de la phase de roue libre. Un choix approprié est effectué et on réalise alors l’étude des pertes silicium du convertisseur. Les performances du convertisseur utilisant la modulation proposée et celle utilisée classiquement dans la littérature sont comparées. La méthodologie de calcul des pertes silicium est présentée ainsi que la validation fonctionnelle de cette nouvelle modulation par des relevés expérimentaux réalisés sur un prototype laboratoire.

Cet article propose un modulateur simple et synthétique adapté au convertisseur matriciel, basé sur l’introduction d’un convertisseur virtuel. Le but de ce travail est de construire une Modulation de Largeur d’Impulsion (MLI) simple et générale pour le contrôle des convertisseurs matriciel industriel. La solution proposée est équivalente à une modulation vectorielle (SVM) particulière. Cette solution prend en compte les harmoniques et le déséquilibre des tensions d’entrée et permet de générer un ratio en tension maximale identique aux modulations classiques (86%). Cet article présente tout d’abord la méthode de calcul de la modulation moyenne d’un convertisseur virtuel puis la modulation naturelle associée au convertisseur matriciel, l’ensemble étant validé par des relevés expérimentaux.

Cet article présente le fonctionnement des onduleurs Z-source. Ils utilisent un réseau d’impédance pour coupler l’onduleur à la source de tension continue. Ce réseau d’impédance est constitué d’une structure L C hybride croisée. Il permet à l’onduleur d’amplifier la tension de sortie grâce à une commande spécifique, ce qui le rend équivalent à la mise en cascade d’un hacheur survolteur avec un onduleur classique. L’étude du principe de fonctionnement ainsi qu’une comparaison avec une structure classique équivalente montreront les limites de cette structure récemment introduite.

The method involves performing synchronization and phasing steps permitting to implement control voltage vector at frequency and a turning voltage vector defined by exact connection between three input phases (u, v, w) and three output phases (a, b, c) of a speed variator. The control voltage vector is amplified beyond 87 percent of amplitude of input voltages (Vun, Vvn, Vwn) and until reaching amplitude of the turning voltage vector, where the input phases are connected to an alternative voltage source applying the input voltages between the input phases and the output phases. An independent claim is also included for a direct matrix converter type speed variator comprising a control unit for amplifying a control voltage vector.

The invention relates to a control method implemented in a variable speed device of the matrix converter type, comprising: three input phases (u, v, w) connected to an AC voltage source and three output phases (a, b, c) connected to an electrical load, nine two-way current and voltage electronic switches (fau, fav, faw, fbu, fbv, fbw, feu, fev, fcw) intended to be individually controlled in order to connect an output phase to any one of the input phases, the operation of switching the switches of the converter obeying a duty cycle matrix for obtaining an output voltage at the load, said duty cycle matrix including a zero phase, the method including a step of suppressing the zero phase in the duty cycle matrix and a step of positioning a new zero phase in the duty cycle matrix so as to minimize the switching losses and the common-mode voltages

The invention relates to a control method implemented in a matrix converter variable speed drive, comprising: three input phases (u, v, w) connected to an alternating voltage source and three output phases (a, b, c) connected to an electric charge, nine bidirectional current and voltage electronic switches (fau, fav, faw, fbu, fbv, fbw, fcu, fcv, fcw) distributed among three switching cells (A, B, C) and to be individually controlled for connecting an output phase to any of the input phases, wherein the switching between the switches of the converter is contingent upon a real matrix of duty cycles used for obtaining an output voltage towards the charge, the real matrix of duty cycles being determined from a virtual matrix (Mv) of a virtual converter (20) comprising three switching cells (A’, B’, C’), one of the switching cells of the virtual converter being permanently blocked.

Après avoir réalisé une thèse en électronique de puissance, j’ai intégré l’équipe Réseau du L2EP en 2011. Cette équipe s’intéresse aux problématiques émergeant de la décentralisation des sources de production et la production intermittente à base d’Énergies Renouvelables. Ces sources intermittentes, sont de nature DC pour les panneaux solaires par exemple ou alternatifs à fréquence variable pour les éoliennes par exemple. Elles nécessitent une interface active d’électronique de puissance. Un axe de recherche de l’équipe s’intéresse sur les aspects dynamiques et la stabilité du réseau avec un fort taux de pénétration de source connectée via des interface d’électronique de puissance. Elle s’intéresse aux technologies de conversions de systèmes connectés au réseau, leurs contrôles aussi bien local (rapproché et éloigné) que global (réglage primaire, secondaire du réseau). Cet axe est au croisement des compétences en électronique de puissance et sur les réseaux électriques. Une forte activité de cette décennie dans cet axe a été réalisée dans le contexte du transport en haute tension continu (HVDC) basé sur des câbles souterrains (ou sous-marins pour les applications off-shore). Cette technologie propose une alternative intéressante au transport en haute tension alternatif pour de longue distance pour des raisons aussi bien technologique qu’économique. Cette évolution doit beaucoup à l’avènement du Convertisseur Modulaire Multiniveau (MMC). Le besoin en recherche dans ce domaine porte sur les points suivants : • Les topologies de convertisseur HVDC/HVAC, leurs dimensionnements et leurs contrôles • L’étude de l’intégration de ces topologies dans les réseaux HVAC et HVDC, la supervision et l’interaction dynamique entre ces réseaux • L’étude en mode défaillant, aussi bien coté composant que défaillance système comme un court-circuit. Pour répondre à ces enjeux scientifiques, j’ai développé au cours de cette décennie des outils de compréhension, de dimensionnement des topologies de convertisseur pour le réseau. Cette compréhension passe par le contrôle de la topologie pour terminer par son intégration et son exploitation dans les réseaux. La synthèse de mes travaux de recherche sur le développement de convertisseurs modulaires et multiniveaux pour les réseaux électriques et se décompose en quatre chapitres. Le premier chapitre illustre une méthodologie permettant de faire évoluer les topologies de convertisseurs statiques de la basse tension vers la haute tension et la haute puissance. Le second chapitre scientifique synthétise la méthodologie de modélisation dynamique pour aboutir à une architecture de commande. Le troisième chapitre scientifique présentera une méthodologie de dimensionnement des condensateurs des sous-modules et d’estimation des pertes et du rendement. L’interaction avec le contrôle adopté sera mise en évidence. Le dernier chapitre scientifique se focalisera sur le principe de développement de prototypes de convertisseurs modulaires et multiniveaux développé lors de cette décennie sur la plateforme du L2EP. Pour terminer, une dernière partie sera l’occasion de conclure et d’évoquer la poursuite des travaux de recherche, en présentant un axe décomposé en différentes pistes, à court, moyen ou long terme.

Mots-clés : Electronique de Puissance, Variateur de vitesse, Convertisseurs statiques, Convertisseurs AC-AC, Convertisseurs matriciels, Modulation de largeur d’impulsion, Commande rapprochée, Commutations. Résumé : La variation de vitesse des machines électriques est une application très porteuse de l’électronique de puissance. La solution de conversion la plus répandue consiste à connecter en cascade deux convertisseurs statiques et d’effectuer une double conversion (AC/DC/AC). Une autre solution, faiblement répandue dans l’industrie, effectue une conversion directe AC/AC pour piloter des machines électriques triphasées à partir d’un réseau alternatif triphasé. Ce mémoire présente une synthèse sur les solutions de conversion directe ainsi que sur les stratégies de modulation des convertisseurs matriciels et matriciels ‘‘double étage’’. Une modulation scalaire généralisée est développée et associée à un modulateur par porteuse, équivalente à la modulation vectorielle et applicable aux différents convertisseurs directs triphasés. Cette généralisation a permis de mettre en avant des modulations à nombre de commutation réduit par période de découpage, et de mettre en évidence qu’une solution particulière réduit les pertes et améliore le comportement électromagnétique du dispositif comparativement aux modulations traditionnellement utilisées. La présentation des contraintes réelles (commande rapprochée des interrupteurs, les protections ainsi que le filtrage) a été abordée et a été utilisée pour développer une maquette laboratoire. Les stratégies de modulation ont été implantées expérimentalement et valident l’étude théorique. Enfin, un fonctionnement direct à la fréquence réseau, non moduler, donc à faibles pertes, est proposé. Key words: Power electronics, adjustable speed drives, static converters, AC-AC converters, Matrix converters, Pulse width modulation, Transistor control, Switching. Abstract: In the power electronics field, the adjustable speed drives is a growing application for electric motors control. The most common conversion solution is to connect in series two static converters and perform a double conversion (AC/DC/AC). Another solution, hardly proposed by industry, uses a direct AC/AC conversion. This thesis aims to make a direct conversion solutions and matrix converters and ‘‘two stage’’ matrix converters modulation strategies synthesis for the purpose to control three-phase electric motor with a three phase input network. This synthesis has developed a generalized scalar modulation, combined with a carrier wave modulator, equivalent to the space vector modulation and applied to matrix converters and the ‘‘two stage’’ matrix converters. Some attention has been done to reduce the switching number during the modulation period. The generalization allows to propose a modified modulation which reduces the losses and improves the electromagnetic performance compared to the traditional modulations used for these kind of converters. The practical constraint (switches control, protection system and filtering) are discussed and has been used to develop a laboratory prototype. The modulations strategies have been implemented experimentally and validate the theoretical study. Finally, a direct function mode with an equal frequency between the input and output network is proposed, without modulation and therefore low losses. In the last part, a particular operation mode is then introduced, permitting the transient operation between the modulated conventional mode to the direct mode without modulation.