T. J. Summers

Multigoal Heuristic Model Predictive Control Technique Applied to a Cascaded H-bridge StatCom

A multilevel H-bridge StatCom inherently contains redundancy in the available switching states. This paper develops a variation on the typical model predictive control scheme which is able to exploit this redundancy to simultaneously balance the H-bridge capacitor voltages, provide excellent current reference tracking, and minimize converter switching losses. The scheme consists of a dead-beat current controller that has been integrated with heuristic models of the voltage balancing and switching loss characteristics. The integration of a pulsewidth modulation scheme is also described. Simulation and experimental results are presented that confirm the correct operation of the control and modulation strategies. Comparison with traditional control and modulation schemes is provided in terms of the key performance indicators associated with multilevel H-bridge StatComs.

Dead-time issues in predictive current control

Current control in inverter-driven machine systems is the inner most component of the hierarchy of control loops. If the control of current in the machine is not fast and accurate then it is difficult, if not impossible, to build a high-performance drive system. Unfortunately, the implementation of current control in power electronic systems is not ideal. Practical effects can have a significant influence on its performance. This paper examines one of these effects, dead time, and considers the influence it has on the performance of predictive current controllers (PCCs). The paper presents analysis that shows that a PCC implicitly compensates for voltage loss due to dead time. Also, a modified PCC is introduced that reduces the zero-current-clamp problem caused by dead time. Simulation and experimental results are presented to verify the analysis and confirm the performance of the new algorithm.

Arctan Power–Frequency Droop for Improved Microgrid Stability

The term microgrid is usually reserved for a modest sized, local distributed generation network that will largely operate standalone (i.e., without a grid connection.) The most common power flow control method utilized in a standalone microgrid is a technique known as power-frequency droop. This paper introduces the concept of utilizing an arctan function for the power-frequency droop profile. The use of this arctan function improves the small signal stability of the two-inverter microgrid, provides natural frequency bounding, and is flexible in its application. SABER simulations are performed to obtain the operating points about which the system is linearized for the stability analysis. Experimental results obtained from a dSPACE-controlled, low-voltage, two-inverter hardware system are presented to verify the theoretical and simulation results.

Control and modulation scheme for a Cascaded H-Bridge multi-level converter in large scale photovoltaic systems

Multi-level Cascaded H-bridge (CHB) converters are ideal for implementing large scale photovoltaic systems. The improved quality of the voltage waveforms, high efficiency and ability to employ multiple Maximum Power Point Tracking (MPPT) algorithms are just some of the advantages. In this paper a three-phase CHB converter supplied with photovoltaic arrays is considered. A control and modulation structure based on Model Predictive Control (MPC) is described. The scheme inherently controls the DC link voltages while also providing the ability to modify any of those voltages to meet MPPT requirements. This avoids the cost and added complexity of extra DC/DC converters that are typically required to keep the DC link voltages uniform. Simulation and experimental results are presented that confirm the correct operation of the proposed approach.

Using a cascaded H-bridge STATCOM for rebalancing unbalanced voltages

This paper considers specific issues related to using the cascade (also known as the H-bridge) multilevel STATCOM with unbalanced voltages and currents at the STATCOM-AC line connection terminals. Under this condition the average power into the individual phase legs of the converter, in general, is not zero, although the total three phase power is zero. The resultant phase cluster power unbalance can result in increasing or decreasing capacitor voltages within a phase leg. This paper will present a technique for zeroing the phase cluster average power under unbalanced line conditions when the control strategy is to rebalance unbalanced voltages.

Impact of Practical Issues on the Harmonic Performance of Phase-Shifted Modulation Strategies for a Cascaded H-Bridge StatCom

Phase-shifted carrier (PSC) modulation has become an industry standard in its application to multilevel H-bridge static compensators (H-StatComs). The technique uses the cancellation of harmonics within each phase leg to significantly improve the harmonic performance relative to the switching frequency. This paper investigates subtle practical implementation issues which deteriorate the harmonic performance of this technique. The effects of nonuniform dc bus voltages and capacitor voltage balancing strategies are investigated. Simulation and experimental results are presented which show that the harmonic performance of the PSC technique deteriorates as the number of voltage levels produced by the H-StatCom increases.

Outline of the Control Design for a Cascaded H-Bridge STATCOM

The cascaded H-bridge converter (CHC) topology is ideal for implementation of direct connection static compensators (STATCOMs). However the topology requires significantly different control strategies as compared to conventional two level transformer connected STATCOMs and other multilevel STATCOMs. This paper outlines all aspects of a novel controller for a CHC STATCOM, from the capacitor voltage balancing, limiting strategies, through to the real and imaginary power loops. Simulation and experimental results are presented to demonstrate its viability and performance.

Phase-Shifted Carrier Modulation Techniques for Cascaded H-Bridge Multilevel Converters

Phase-shifted carrier modulation is an industry standard in its application to multilevel H-bridge converters. The major advantage of this scheme over level-shifted and space vector modulation schemes is its inherent ability to evenly distribute losses between semiconductor devices. However, until now, its implementation has necessitated a capacitor voltage balancing scheme that degrades converter harmonic performance. This paper develops two phase-shifted carrier modulation schemes that avoid harmonic degradation by exploiting an extra degree of freedom in the available switching states. Simulation and experimental results are presented that confirm the correct operation of the modulation strategies.

Comparison of modulation strategies for a cascaded H-bridge StatCom — Part 1: Theoretical background

Phase shifted carrier modulation has become an industry standard in its application to multi-level H-bridge Stat-Coms. The technique uses the cancellation of harmonics within each phase-leg to significantly improve the harmonic performance relative to the switching frequency. The purpose of this paper is to investigate whether the practical implementation of the technique deteriorates the harmonic cancellation. The effects of non-uniform DC bus voltage, transient capacitor voltages and phase shift error are investigated. A comparison with alternative modulation strategies is developed in terms of a defined set of performance indicators. Part 2 of this paper applies the developed analysis to a particular StatCom and investigates the effect of increasing the StatCom level number.

A laboratory system to produce a highly accurate and uniform magnetic field for sensor calibration

This paper addresses one particular issue that the authors came across when developing a DC Motor Duty Meter - a comprehensive condition monitoring tool for large DC motors based on flux density measurement inside the motor air gap. This particular issue crucial for the project was how to calibrate, with high accuracy, a large number of flux density sensors to be used to measure the flux density distribution along the surface of the motor poles. From available literature and IEEE standards, the recommended and most common way to generate a calibration magnetic field is based on Helmholtz coil principle, which in practice is limited to small flux density magnitudes. A method used by the authors and described in this paper utilizes a step test and simple calculations based on Faradays Law. The novelty of the described method is in particularities of the system, measurements and calculations, which make the method highly immune to typical sources of errors. The method can be used to produce a very accurate and uniform magnetic field with densities ranging from fractions of Tesla to above one Tesla to calibrate flux sensors in this flux range. Its accuracy is only limited by that of a current source, a current probe and a digital oscilloscope. Uniformity of the generated field is discussed and experimentally confirmed. Experimental results obtained within the DC Motor Duty Meter project are included to illustrate correctness of the sensor calibration. Conditions that guarantee the maximum efficiency of the method are discussed. The paper is concluded by a step by step calibration procedure that other researchers may find useful.