Estanis Oyarbide

Predictive Direct Power Control - A New Control Strategy for DC/AC Converters

This paper proposes the predictive direct power control (P-DPC), a new control approach where the well-known direct power control (DPC) is combined with a predictive selection of a voltage-vectors sequence, obtaining both high transient dynamic and constant switching frequency. Two different P-DPC versions are developed based on two different types of voltage sequences: a two-vector one and a three-vector one. Simulation results of the three-vector sequence P-DPC are compared to standard voltage oriented control (VOC) strategies. Thanks to its high transient capability and its constant switching behaviour the P-DPC could become an interesting alternative of standard VOC techniques for grid-connected converters

New vector control algorithm for brushless doubly-fed machines

A new vector control algorithm for the brushless doubly-fed machine (BDFM) has been developed. The goal of the BDFM control is to achieve the same or higher dynamic performance as the wound rotor machine, exploiting the well known induction motor vector control philosophy. For this purpose a unified reference-frame model is preferred. This paper exploits this new model leading to some interesting coupling properties and allowing a high performance dynamic response. All this work is tested and validated in an experimental machine prototype.

Stability Analysis of a Brushless Doubly-Fed Machine under Closed Loop Scalar Current Control

This paper presents a stability analysis of the brushless doubly-fed machine (BDFM) under closed loop scalar current control (CLSCC). The recently developed unified reference-frame dq model leads to an invariant small signal model reflecting the dynamic of the BDFM under CLSCC. Theoretical stability analysis shows a stable behavior along all the operation range. This result is confirmed by an experimental BDFM set-up. It is concluded that CLSCC becomes the simplest control alternative of any real BDFM application

Dynamic Model of Class-E Inverter With Multifrequency Averaged Analysis

This paper presents a novel dynamic model of a single-switch class-E inverter. Both the transient and steady-state behaviors of main variables are derived using a multifrequency averaged (MFA) modeling method. The analysis of the boundary conditions reveals that, under some approximations, the performance of the generalized averaged model is close to that of the “exact” or topological one. A detailed analysis of the errors and limitations of the model has been derived. Contrary to other class-E inverter models, based on the input-output behavioral black-box approach, the MFA modeling technique leads to a large-signal knowledge-based model. The advantage of this type of model is that it is applicable to a wide set of analysis and control design tools. The proposed model is verified by comparing the MFA, state-space, and experimental results, showing good agreement for the dynamic behavior of the inverter under supply voltage modulation and also for the frequency and phase modulation of the input control signal.

Ultracapacitor-based plug & play energy-recovery system for elevator retrofit

Vertical transportation is a promising field in terms of efficiency improvement opportunities. Several manufacturers are working on the development of new and efficient elevator lifts based on energy-recovery schemes. This paper is focused on the retrofit of already existing inverter-based non-efficient elevator actuators. This purpose is achieved through a) the connection of only two wires to the DC bus of the existing inverter, and b) a very simple but robust control based on several voltage hysteresis-bands. Simulation and experimental results show the validity of the approach.

Predictive direct power control of MV-grid-connected two-level and, three-level npc converters: experimental results

This paper deals with the Predictive Direct Power Control of MV-grid-connected two-level and three- level NPC converters. The well-known direct power control is combined with a predictive selection of a voltage-vectors sequence, obtaining both high transient dynamic and constant switching frequency on high-power converters. The initial Predictive Direct Power Control proposal explored only the basic formulation based on simple vector-sequences applied to two-level converters. Later on, the same approach was extended to the family of three-level NPC converters, including the predictive direct control of neural voltage. In the same way, an advanced hybrid Predictive Direct Power Control was presented, providing only simulated data. The work presented in this paper provides experimental results of the hybrid Predictive Direct Power Control applied to two-level and three-level NPC converters. These results show that the P-DPC approach is an attractive solution to MV grid-connected applications.

Investigation of Core Loss Effect on Steady-State Characteristics of Inverter Fed Brushless Doubly Fed Machines

Brushless doubly fed machine (BDFM) shows promising results for wind power and adjustable-speed drive applications. Due to poor rotor magnetic coupling and relatively high value of slip, core loss is an important factor which can affect the steady-state and dynamic performance of BDFM. Specifically, core loss plays a major role when evaluating the efficiency of the machine. By now, there is no comprehensive study on core loss effect on performance characteristics of BDFM in the literature. In this paper, based on the derived relations and applying transformation of parameters from one winding to another one, the steady-state equivalent circuit model of BDFM without core loss is modified in order to take the core loss into account. Experimental results from a BDFM prototype are compared with the predicted results given by the model and the accuracy and effectiveness of the new approach is validated.

Design and characterization of a 1 kW AC self - excited Switched Reluctance Generator

The Switched Reluctance Machine has certain advantages such as robustness and the absence of permanent magnets, which have encouraged its use in recent years. One of its applications is as a self - excited generator. This behavior is produced when a capacitor is connected in parallel to each single phase forming a resonant circuit. This application can be used in isolated sites with a wind turbine for energy generation. This paper presents the specific design of a 1 kW self - excited switched reluctance machine for wind power applications. In addition, presentation is made of the specifications, finite elements analysis, thermal profile and experimental results.

Multi-frequency model of a single switch ZVS class E inverter

This paper presents a novel dynamic model of a single switch resonant inverter. Both transient and steady state behavior of main variables are derived using a multi-frequency average modeling method. The analysis of the boundary conditions reveals that, under some approximations, the performance of the generalized averaged model is close to that of the “exact” or topological one. Contrary to other Class E inverter models, based on input-output behavioral black-box approach, the multi-frequency average modeling technique leads to a large-signal knowledge-based model. The main advantage of this type of model is that it is possible to make use of a wide set of analysis and control design tools like distortion analysis in power amplifiers, advanced control techniques in high frequency resonant inverters and dynamic-oriented novel design methodologies. Thanks to this model a predictive direct control of the transient trajectory could be achieved, leading to a smooth evolution from the initial steady-state operating point to the new desired state.

Inductance characterization of a SRM using finite element simulation data

FE simulation data provides information about magnetic properties corresponding to a range of foreseen current excitation and rotor positions. This information needs further processing to derive other electric parameters, among which, inductance plays an important role for simulation and modeling purposes. This paper describes this process looking for analytical expressions and evaluating the incidence of simplicity against accuracy. Although searching for analytical expressions is usually justified by time saving, in this case it is because analytical expressions are necessary to model the machine as a second order nonlinear system using analytical equations. This representation provides a better insight in the understanding of the behavior of autonomous self-excited AC switched reluctance generators.