Search

The paper aims to provide a guideline for rotor design of permanent magnet (PM) assisted synchronous reluctance machine (PMa-SynRM). Unlike interior PM machines, the PMa-SynRM is essentially an SynRM with dominant reluctance torque. The PMa-SynRM is often designed by directly embedding magnets into an SynRM rotor to improve the performance. However, how the PMs can be arranged in the rotor to enhance motor performance has not been fully discussed. This paper investigates the influence of several key factors concerning the PM arrangements, including number of PM layers, inserted positions, and PM sizes. The torque, flux linkages, power factor, and motor performance are comprehensively evaluated. Based on the analysis, a brief guideline is suggested for PMa-SynRM rotor design. Finite element analysis is used for evaluation and experiments are conducted to validate the analysis.

This paper outlines a method using finite-element analysis (FEA) to model the magnetization process of a small brushless surface-magnet three-phase motor with isotropic ferrite magnets and concentrated windings. This is aimed at allowing in situ magnetization using the three-phase machine winding. Radial magnetization is the nominal manufactured magnetization pattern; however, the in situ method leads to a variation in the magnetization pattern, and the performance is examined using open-circuit simulations. The properties of the magnet material (a ferrite 5 type) are examined in order to simulate the magnetization. Bulk magnetization Hs-sat levels for ferrite type 5 indicate that a 10-kOe field is required. In this paper, localized magnetization is examined using FEA, and it is measured in actual rotors by examination of the surface flux density when removed from the stator. Methods for deriving the approximate back electromotive force from both the FEA and the surface measurements are put forward; these show a reasonable correlation. The use of magnetization functions allows the magnetization pattern to be applied to the simulations. This paper uses a standard manufactured machine to carry these out. Open-circuit experiments are conducted to validate the presented method.

In large induction machines (such as large cage induction motor pumps in the petrochemical industry and cage- or wound-rotor induction generators in wind turbines), reliability and longevity are advantageous. This is particularly relevant to wind turbine generators which can be inaccessible. There will be some degree of tolerance and wear that will lead to low-level rotor eccentricity. For a pm-pole-pair machine, there will be pm ±1 pole-pair flux waves set up by the eccentricity which will generate unbalanced magnetic pull (UMP), as well as additional higher space harmonics. This paper addresses the use of stator damper windings to reduce the sideband flux waves and hence attenuate the UMP. Examples are put forward in terms of a ten-pole cage-rotor machine with static or dynamic rotor eccentricity and then extended to use a four-pole wound-rotor machine. A tested analytical model is developed to include these damper windings and the wound rotor; they are shown to reduce the UMP, particularly in a wound-rotor machine. The simulations here are in terms of a wound-rotor machine, but this can be extended for doubly fed induction generator operation.

This paper describes different arrangements for a dual-rotor, radial-flux, and permanent-magnet brushless dc motor for application to variable-speed air conditioners. In conventional air conditioners, two motors of appropriate ratings are usually used to drive the condenser and evaporator separately. Alternatively, a motor with two output shafts may be employed, and this is studied here. The motor has inner and outer rotors with a stator in between which is toroidally wound or axially wound with inner and outer slotted stator surfaces. The power sharing on the two rotors is designed to meet the requirement of the condenser and evaporator. Finite element analysis (FEA) is employed to verify the designs. A prototype is made and tested to evaluate the performance. Alternative windings are investigated to assess the possibilities of decoupling the rotors so that they run independently. In the final section, a new and novel arrangement is proposed, where one three-phase winding set and one two-phase winding set (both toroidal) are wound on the same stator to control two rotors of different pole numbers. The two winding sets can be bifilar or share the same set of phase windings. This design simplifies the winding (because it is toroidal) and reduces the copper loss or amount of copper required. The design is tested using FEA solutions, and the initial results indicate that this machine could operate successfully.

This paper studies the post-assembly magnetization of a fractional-slot, concentrated winding brushless rare-earth surface permanent-magnet machine with six rotor poles and nine stator slots. In the paper, the simulation method is discussed, and it is illustrated that it is possible to fully magnetize the machine after assembly using two pulses of the winding (“two-shot method”) with a rotation of the rotor between pulses. The MMF required for this process is predicted. The method is then validated experimentally using a test motor and a magnetizer. The machine can be magnetized using connection of either two phases or three phases, and both scenarios are investigated. The open-circuit back electromotive forces are obtained to illustrate the partial magnetization after one pulse and full magnetization after two pulses. Load measurements are also carried out. With the increasing use and size of rare-earth magnet machines, the magnetizing procedure is an important aspect of their manufacture; premagnetized assembly can be difficult.

This paper describes a small segmented oscillating water column (OWC) with three sections and the modelling techniques that can be used to simulate its performance. The turbine uses cascaded Savonius rotors and the unit is developed and tested for validation of the performance algorithms. It is shown that the systems can be easily described and developed. It would be suitable for a shoreline location such as a harbor wall, where waves are random and not orthogonal to the column. Conversion rates in the region of 20% are tabulated for the system with an output of 25-W peak. A full algorithm used for the system is put forward and solved using the Runge-Kutta-Nystrom (RKN) method.