Mohamed Trabelsi

Modular Multilevel Converter Circulating Current Reduction Using Model Predictive Control

Modular multilevel converter (MMC) is a promising new topology for high-voltage applications. The MMC is made of several identical submodules. For proper operation, each submodule can be considered as a controlled voltage source where capacitors voltage should be maintained at a certain level. Besides, the minimization of the circulating current, which does not flow to the load, is crucial for achieving stable and efficient operation of the MMC. The interrelations among the load current, circulating current, and capacitor voltages complicate the MMC control. This paper aims to achieve stable and balanced voltage and current control with reduced circulating current in various operating conditions. The proposed control uses weighted model predictive control based on a normalized cost function to select the inverter switching patterns, which control the load current, while minimizing voltage fluctuation and circulating current. The weighting factors were selected based on minimizing the load-current total harmonic distortion (THD) and circulating current. The analysis is conducted on a low-power case study of single-phase four-cells MMC with possible extension to higher number of cells. The low-power three-level prototype is designed and built to validate this proposed method. Theoretical analysis, simulation, and experimental results are presented and compared. Parameter sensitivity analysis was also conducted. They all confirm the effectiveness of the proposed control method.

An effective Model Predictive Control for grid connected Packed U Cells multilevel inverter

This paper presents a Model Predictive Control (MPC) for grid-tied Packed U Cells (PUC) multilevel inverter. The system under study consists of a single-phase 3-cell PUC inverter connected to the grid through filtering inductor. The proposed topology allows the generation of 7-level output voltage with reduction of passive and active components compared to the conventional multilevel inverters. The aim of the proposed MPC technique is to achieve grid-tied current injection, low Total Harmonic Distortion (THD) of the current, unity power factor, while balancing the capacitor voltages at maximum power point (MPP). The feasibility of this strategy is validated by simulation using Matlab/Simulink environment.

Model Predictive Control of Z-Source four-leg inverter for standalone Photovoltaic system with unbalanced load

In this paper, Z-Source (ZS) four-leg inverter equipped with a Model Predictive Control (MPC) strategy for standalone Photovoltaic (PV) system under various load conditions is presented. The inverter is aimed to supply simultaneously three-phase and single-phase AC loads with balanced voltage and constant frequency. In this topology, a simple LC (Z) network is used instead of the boost converter to obtain single stage power conversion system. The MPC scheme is developed to control output currents of the inverter and ZS network current and voltage. Hardware-In-The-Loop (HIL) tests have been performed to verify steady-state and transient performances of the proposed system under balance/unbalanced load and/or reference current conditions.

1-MW quasi-Z-source based multilevel PV energy conversion system

This paper proposes a 1 MW grid-tie photovoltaic system by connecting quasi-Z-Source (qZS) networks into Cascaded H-Bridge (CHB) inverter. The proposed Power Conditioning System (PCS) consists of a 3-phase 48-cell CHB inverter where each module is fed by a qZS network. The proposed topology allows boosting the input voltage to a higher level while solving the traditional imbalance problem of DC-link voltage in CHB inverters. A multilevel output voltage waveform is produced using an enhanced phase shifted pulse width modulation technique, which inserts shoot-through states into the conventional zero states to control the qZS-CHB module. The effective control schemes are proposed to achieve grid-tie current injection, low Total Harmonic Distortion (THD) current, unity power factor, and DC-link voltage balance for all qZS-CHB inverter modules. A 1 MW system is built in MATLAB/SIMULINK environment to verify the proposed multilevel PV energy conversion system and its control principles.

A unique active anti-islanding protection for a quasi-Z-Source based Power Conditioning System

This paper presents the performance assessment of three active Islanding Detection Methods (IDMs) for a quasi-Z-Source (qZS) based Power Conditioning System (PCS). The proposed PCS consists of a three-cell Cascaded H-Bridge (CHB) inverter where each module is fed by a qZS network. The proposed combined controller achieves grid-tie current injection, DC-link voltage balance for all qZS-CHB inverter modules, and anti-islanding protection. The anti-islanding protection performance is evaluated for three phase-shift based active Islanding Detection Methods (IDMs), namely Active Frequency Drift (AFD), Slide-Mode Frequency Shift (SMS), and Sandia Frequency Shift (SFS). Simulation results and real-time implementation using dS1103 board are carried out to validate the proposed design.

A Model Predictive Control technique for utility-scale grid connected battery systems using packed U cells multilevel inverter

Grid-connected energy storage systems have been implemented in ac power systems as uninterruptable power supplies (UPS). Batteries and bi-directional power converters provide electrical power when off-grid and recharge when grid-connected. In this paper, a packed U cells (PUC) seven-level inverter has been selected as the grid-interface due to the lower cost and fewer number of components compared to other bi-directional topologies. Additionally, the PUC has higher power quality when compared to the traditional H-bridge. Compared to the traditional PI controller, Model Predictive Control (MPC) is attracting more interest due to its good dynamic response and high accuracy of reference tracking. Through the minimization of a user-defined cost function, the proposed MPC technique can simultaneously achieve unity power factor, low total harmonics distortion of the grid-side current and balance the PUC capacitors voltages at the grid side, and control bi-directional power flow in the batteries-PUC system. The presented topology and proposed control technique are verified by simulating a 600 W reduced-scale prototype. The theoretical principles are validated by implementing the controller on the prototype using dSPACE 1007 platform.

Model predictive control of three-phase three-level Neutral-Point-Clamped qZS inverter

This paper presents a Model Predictive Control (MPC) of three-phase three-level Neutral-Point-Clamped (NPC) quasi-Impedance Source (qZS) inverter. This topology features a wide range of voltage gain that is suitable for applications in renewable energy based power systems where the output of the renewable energy sources varies widely with operating conditions such as wind speed, temperature, and solar irradiation. To eliminate the complexity of the modulation techniques and to overcome the weaknesses of the inner control loop performance, MPC technique is employed in control stage. To verify the dynamic and steady-state performances of the proposed MPC scheme under various load and reference currents. Simulation studies are performed with Matlab&Simulink software. Obtained results show that the proposed MPC strategy works properly at steady-state and transients.

Improvements of Object Grabbing Method by Using Color Images and Neural Networks Classification

The works presented in this paper allow the improvement of an object grabbing method developed for the ARPH project (Robotic Assistance for Disabled people). The basic tool of this project is a mobile robot with an embedded MANUS arm. The main aim of the method developed here is to make a manipulator arm equipped with a gripper and two kinds of sensors (camera and sonar) able to seizure and handle several objects in the human environment. Ultrasonic echo is recovered by a neural network classifier in order to recognize the object shape. At the same time, a little wireless HF camera takes a color image of the target, which allows to compute the coordinates of the object in the image then in the camera frame. The combination of these various information about the object with the camera geometrical model and the visual servoing principle allows to adopt an iterative strategy to approach the object under good conditions and to seize it at the end. Two kinds of object are used (spherical and cylindrical) with several colors and diameters

Finite control set model predictive control for grid-tied quasi-Z-source based multilevel inverter

In this paper, a finite control set Model Predictive Control (MPC) for grid-tie quasi-Z-Source (qZS) based multilevel inverter is proposed. The proposed Power Conditioning System (PCS) consists of a single-phase 2-cell Cascaded H-Bridge (CHB) inverter where each module is fed by a qZS network. The aim of the proposed control technique is to achieve grid-tie current injection, low Total Harmonic Distortion (THD) current, unity power factor, while balancing DC-link voltage for all qZS-CHB inverter modules. The feasibility of this strategy is validated by simulation using Matlab/Simulink environment.

Investigation into the effect of unbalanced supply voltage on detection of stator winding turn fault in PMSM

In recent years, permanent magnet synchronous motors (PMSM) are becoming popular in industrial applications. It is difficult to have perfectly balanced 3 phase ac supply voltages in power system. In fact, line supply voltages typically differ by a few volts or more, and hence unbalanced supply voltages can exist anywhere in a three-phase power distribution system. This paper presents a study of PMSM with stator winding turn fault and different load conditions under balanced and various unbalanced supply voltages. Three-phase PMSM is simulated for healthy and faulty conditions with stator turn fault at different loads using finite element analysis. Experimental laboratory results validate the analysis and demonstrate that the proposed approach can be used for both balanced and unbalanced supply voltage, as a reliable indicator for detecting different insulation failure percentages of stator winding at different load conditions.