I.M. de Alegria

Intelligent PV module for grid-connected PV systems

Most issues carried out about building integrated PV systems (BIPV) performance show average losses about 20-25% in electricity production. The causes are varied: mismatching losses, partial shadows, variations in I-V characteristics of PV modules due to manufacturing processes, differences in the orientations and inclinations of solar surfaces, temperature effects, etc. Most losses can be decreased by means of a suitable electronics. This paper shows the Intelligent PV Module concept, a low cost and high efficiency DC-DC converter with MPPT functions, control and power line communications (PLC). Several DC-DC configurations have been considered as well as MPPT methods and communication options. Matlab simulations describe how this Intelligent PV Module can make a significant improvement in the energy delivered, above all in facades and roofs, in which the number of modules and the external conditions make this device most suitable.

New Protection Circuit for High-Speed Switching and Start-Up of a Practical Matrix Converter

The matrix converter (MC) presents a promising topology that needs to overcome certain barriers (protection systems, durability, the development of converters for real applications, etc.) in order to gain a foothold in the market. Taking into consideration that the great majority of efforts are being oriented toward control algorithms and modulation, this paper focuses on MC hardware. In order to improve the switching speed of the MC and thus obtain signals with less harmonic distortion, several different insulated-gate bipolar transistor (IGBT) excitation circuits are being studied. Here, the appropriate topology is selected for the MC, and a recommended configuration is selected, which reduces the excursion range of the drivers, optimizes the switching speed of the IGBTs, and presents high immunity to common-mode voltages in the drivers. Inadequate driver control can lead to the destruction of the MC due to its low ride-through capability. Moreover, this converter is especially sensitive during start-up, as, at that moment, there are high overcurrents and overvoltages. With the aim of finding a solution for starting up the MC, a circuit is presented (separate from the control software), which ensures correct sequencing of supplies, thus avoiding a short circuit between input phases. Moreover, it detects overcurrent, connection/disconnection, and converter supply faults. Faults cause the circuit to protect the MC by switching off all the IGBT drivers without latency. All this operability is guaranteed even when the supply falls below the threshold specified by the manufacturers for the correct operation of the circuits. All these features are demonstrated with experimental results. Lastly, an analysis is made of the interaction that takes place during the start-up of the MC between the input filter, clamp circuit, and the converter. A variation of the clamp circuit and start-up strategy is presented, which minimizes the overcurrents that circulate through the converter. For all these reasons, it can be said that the techniques described in this paper substantially improve the MC start-up cycle, representing a step forward toward the development of reliable MCs for real applications.

Improvement of the Design Process of Matrix Converter Platforms Using the Switching State Matrix Averaging Simulation Method

The matrix converter (MC) is arousing considerable attention as an alternative for conventional ac/ac converters due to the advantages it offers. However, the control and modulation of this converter is complex. This, together with the fact that the MC usually operates at high modulation frequencies, makes the computational load of the platform to be simulated excessively high. All this makes the simulation time of models including the MC excessively long, even more so when both the transient and steady state of the system must be analyzed. This paper presents a new MC simulation technique called Switching State Matrix Averaging (SSMA). Although this is a fixed-step technique, a long simulation step can be used without forfeiting the accuracy of an ideal variable-step simulation. Likewise, the SSMA drastically speeds up the simulation, reducing the amount of required resources and the tuning time of the complex platforms in which the MC is used. A series of simulations has been performed in order to verify the proposed method. Moreover, a comparison between experimental and simulation results has been made, demonstrating the effectiveness of the proposed method.

Novel power error vector control for wind turbine with doubly fed induction generator

This paper proposes a new control method for a doubly fed induction generator. The control is suited for variable speed operation where stator active and reactive power control is required, especially in wind power generation with variable-speed wind turbine. The control is similar to DTC controllers but significantly simplifies the determination of the rotor voltage vector to be applied. The stator active and reactive power errors are directly translated into a reference vector that determines the rotor voltage vector without the need of subsynchronous or supersynchronous operation detection or rotor flux sector detection. This control method, called power error vector control, is less complex and obtains similar results to direct power control. This is supported by theoretical analysis, and verified by simulation results. A test rig is being built to experimentally test the algorithm.

Switching Frequency Behaviour for a Practical Matrix Converter

There are certain barriers that prevent the commercialisation of matrix converters. In the design of a real converter, the parameters that intervene in the modulation and commutation must be quantified with precision. This paper defines the commutation parameters that intervene in a vector modulation algorithm (SVM). On the other hand, it determines the commutations that occur when the double sided SVM technique is used both with semi-soft commutation, assessing the influence of the modulation frequency and technique on the design of the filter, driver consumption and their power sources. All this facilitates the dimensioning of the clock and the resolution of the variables to be used in the implementation of the algorithm in a FPGA. The paper concludes with a series of experimental results made with a Eupec module, allowing the selection of MC components

Reconfigurable microstepping control of stepper motors using FPGA embedded RAM

The control of stepper motors using microsteps in open-loop systems is a well known mechanical and electrical improvement. The most common stepper motor controllers have all the elements to perform a microstepping control requiring only the digital control for the current level and direction in each winding. In this design we propose a cost-effective solution to integrate this control into high speed FPGA based designs. The core uses an FPGA block RAM to save conventional logic resources. Also, if the system needs to change the microstepping control granularity while the system is running, a partial dynamic reconfiguration can be performed only by changing the content of the block RAM attached to the core.

Matrix Converter Double Sided Space Vector Modulation: a fast way to synthesize via S-Function

Matrix converters present several advantages in some power electronic applications, but their control is a complicated task. Matlab-Simulink is a very useful tool for the developing of control algorithms for power converters. In this paper, an exhaustive analysis of double sided space vector modulation for a matrix converter is presented. The synthesis of the complex space vector modulator is done using a single S-function, thus simplifying and accelerating the implementation and debugging of the controller of a matrix converter. The implementation procedure of the modulation strategy presents several important advantages, mainly, an acceleration of the time dedicated to the implementation of the modulator, a more reliable design, because the debugging is straightforward, and faster simulation times of the entire platform. The simulation method allows an easy way to synthesize the modulation algorithm in a FPGA, in this way, this paper contributes to the development of the IPMs. Finally, the results of applying the present space vector modulator for the control of a doubly fed induction generator (DFIG) are presented, showing that complex control of matrix converter is minimized

Improvement of the Matrix Converter Start-up Process

The behavior of the matrix converter (MC) is usually studied in the stationary condition, however, turning on the MC is a very crucial moment for this device as overcurrents and overvoltages may appear, damaging the converter. In order to improve its ride-through capability, this paper focuses on matrix converter (MC) power-up protection strategies. The input filter and clamp circuit protect the MC and improve its behavior in the stationary condition. Given the importance of the filter, the design criteria to be followed, its mathematical analysis and the procedure to select its constituent parts, are defined in great detail. Following this, an analysis is made of the interaction between the filter and the MC clamp circuit during the initial transitory of the MC. Finally, a new, improved solution for the clamp circuit is proposed. This optimizes the behavior of the MC during power-up. The start-up procedure that mitigates overcurrents around the input filter is also defined. Consequently, this paper represents a step forward in the work being done to increase the reliability of the MC.

Matrix converter: improvement on the start-up and the switching behavior

The matrix converter (MC) presents a promising topology that needs to overcome certain barriers (protection systems, durability, the development of converters for real applications, etc.) in order to gain a foothold in the market. Taking into consideration that the great majority of efforts are being oriented towards control algorithms and modulation, this article focuses on MC hardware. In order to improve the switching speed of the MC and thus obtain signals with less harmonic distortion, several different IGBT excitation circuits are being studied. Below, the appropriate topology is selected for the MC and a configuration is presented which reduces the excursion range of the drivers and optimizes the switching speed of the IGBTs. Inadequate driver control can lead to the destruction of the MC due to its low ride-through capability. Moreover, this converter is specially sensitive during start-up, as at that moment there are high overcurrents and overvoltages. With the aim of finding a solution for starting-up the MC, a circuit is presented (separate from the control software) which ensures a correct sequencing of supplies, thus avoiding a short-circuit between input phases. Moreover, it detects overcurrent, connection/disconnection, and converter supply faults. Faults cause the circuit to protect the MC by switching off all the IGBT drivers without latency. All this operability is guaranteed even when the supply falls below the threshold specified by the manufacturers for the correct operation of the circuits. All these features are demonstrated with experimental results. For all these reasons, it can be said that the techniques proposed in this article substantially improve the MC start-up cycle, representing a step forward towards the development of reliable matrix converters for real applications.

Flexible and Reduced Modulation and Simulation Method for a Matrix Converter-DFIM Modulator

This paper proposes a practical way to simplify large simulation platforms. A complete model with a Matrix Converter (MC) linked to a doubly fed induction machine (DFIM) is described, proving that this kind of converter is suitable for use in power regenerative rigs. Besides contributing to the study of DFIM with MC, a fast way to describe all kinds of Space Vector Modulation (SVM) algorithms is presented. The modulator algorithm is supported theoretically and described in an S-Function of Matlab-Simulink, which reduces the design time and provides a way to implement the control code in new integrated circuits, such as FPGAs; in this sense, this paper contributes to the development of IPMs (Integrated Power Module). Platform modules have been analyzed showing the links between the entire platform (MC and a DFIM). Finally, the entire design has been verified by simulation results.