Search

One of the attractive benefits of slotless machines is low losses at high speeds, which could be emphasized by a careful stator core loss assessment, potentially available already at the predesign stage. Unfortunately, mainstream iron loss estimation methods are typically implemented in the finite element analysis environment with a constant coefficients dummy model, leading to weak extrapolations with huge errors. In this article, an analytical method for the iron loss prediction in the stator core of slotless PM machines is derived. It is based on the extension of the 2-D field solution over the entire machine geometry. Then, the analytical solution is combined with variable coefficient loss models (VARCO) or constant coefficient loss models, which can be efficiently computed by vectorized postprocessing. VARCO loss models are shown to be preferred at a general level. Moreover, this article proposes a lookup-table-based solution as an alternative approach. The main contribution lies in the numerical link between the analytical field solution and the iron loss estimate, with the aid of a code implementation of the proposed methodology. First, the models are compared against a sufficiently dense dataset available from laminations manufacturer for validation purposes. Then, all the methods are compared for the slotless machine case. Finally, the models are applied to a real case study and validated experimentally. [ABSTRACT FROM AUTHOR]

This paper presents a novel design concept for Synchronous Reluctance (SynRel) machines aimed at reducing the torque ripple. Two general sizing approaches based on the homothetic scaling principle are defined and compared. An in depth analysis on the torque ripple, for a wide range of scaled geometries, evaluated by finite element, has been carried out at different operating conditions. A further analysis is performed on 4 scaled geometries that have been optimized starting from 4 random rotor geometries. It is shown that the main rotor geometrical variables converge to similar values for all scaled machines. The accuracy of the proposed model is then validated by comparing the FE simulated torque ripple waveforms with the experimental data carried out, for a range of operating conditions, on a machine prototype. The outcome of this work is a fast and accurate scaling technique for the preliminary design of SynRel machines with reduced torque ripple. [ABSTRACT FROM AUTHOR]

Large medium-voltage electric machine stators are usually equipped with form wound coils made of flat conductors (strands) and embedded in open (rectangular) slots. Air-gap magnetic flux lines can enter the slot and, sweeping the strands placed nearest the slot opening, induce eddy currents in them. Such eddy currents cause additional losses which can be much higher than usual skin-effect and proximity losses. In order to avoid dangerous overheating and hot spots, the additional losses in question need to be carefully predicted in the design stage. Time-stepping finite-element analysis (TSFEA) can be used for the purpose, which however implies a large computational burden and requires the machine geometry to be modeled in detail. This article proposes alternative methods based on time-harmonic finite-element analysis (THFEA) simulations performed on highly simplified machine models and with no need to take rotor motion into account. The proposed methods are shown to produce very accurate results, compared to TSFEA, but with very significant time and computational savings. [ABSTRACT FROM AUTHOR]

The rotor structure of synchronous reluctance motor (SynRm) is usually provided with redial ribs to improve structural ruggedness for high‐speed operation. While these radial ribs contribute to mechanical robustness, they also reduce the average torque and limit the flux‐weakening capability due to an increase in inductance along the q ‐axis. To address this issue, the rotor design is refined by adopting non‐magnetic radial ribs in this study. The introduction of titanium as non‐magnetic material for radial ribs prevents magnetic leakage flux from flowing along the q ‐axis, and thereby improves the average torque, saliency ratio, efficiency and power factor of the SynRm. Furthermore, along with the electromagnetic analysis, the mechanical integrity of the proposed rotor design is also investigated. Two prototypes rated 1 kW at 1500 rpm are fabricated using both the conventional and the proposed rotor design, while the stator and other parameters are maintained constant in the two cases. The conventional design uses magnetic radial ribs whereas the proposed design uses non‐magnetic radial ribs. Finally, a thorough comparative study, including inductance tests on the prototypes, is performed between the proposed and the conventional SynRm, confirming the strengths of the technology solution being set forth. [ABSTRACT FROM AUTHOR]

An accurate prediction of medium-voltage induction motor (IM) power factor in rated conditions is important in the design stage to verify the machine compliance with specifications according to international testing standards. Laboratory and industrial experiences suggest that significant errors in full-load power factor calculation can result for two-pole IMs in particular, due to the eddy currents induced in the solid-steel shaft at rated slip. Such eddy currents are responsible for rejecting the main flux into rotor laminated yokes causing an increase in their saturation and, hence, in the required magnetizing current with respect to no-load conditions. This article proposes a method to study the phenomenon through a combination of analytical and simplified Finite Element (FE) calculations as a computationally-efficient alternative to conventional FE simulations. The results of the proposed approach are experimentally assessed by comparison with measurements on a set of built and tested medium-voltage two-pole IMs of different sizes, showing very good accuracy and computational performance. [ABSTRACT FROM AUTHOR]

Fractional-slot concentrated windings (FSCWs) are an attractive option for the design of synchronous permanent-magnet machines. It is commonly assumed in the existing literature that a symmetrical three-phase FSCW is feasible only on a condition that the number of slots Z is an integer multiple of three times the maximum common divisor between Z and the number of pole pairs p. Slot-pole combinations satisfying this rule can be defined conventionally, the others unconventionally. In contrast to the common belief, this paper shows that, using a multi-layer arrangement, it is possible to synthetize a symmetrical FSCW having unconventional slot-pole combinations. A general design methodology for this purpose is presented and validated by finite element analysis. The pros and contras of FSCWs with unconventional slot-pole combinations are examined. Finally, the application of an unconventional FSCW to a shipboard surface permanent-magnet machine prototype is presented to illustrate the possible practical convenience of this kind of winding and tests on the prototype are reported for experimental validation. [ABSTRACT FROM AUTHOR]

Interior permanent magnet (IPM) machines with spoke-type design are very attractive for several applications, including vehicle traction, which requires a good flux weakening performance to achieve high speed. For this purpose, a high demagnetizing current is usually injected to reduce the machine flux at high speed, with detrimental effect in terms of efficiency. This problem can be mitigated by equipping the rotor with a mechanical flux weakening device, which activates during high-speed operation by centrifugal force. This paper addresses the analytical modeling of this special kind of IPM motor through a simplified magnetic equivalent circuit (MEC) approach, which incorporates saturation and slotting effects as well as the flux weakening device operation. It is shown how the MEC model can effectively describe the IPM machine behavior despite its simplicity. Results are successfully validated by comparison both with finite element analysis, which requires much more computational resources, and with measurements from laboratory prototype testing. [ABSTRACT FROM AUTHOR]

This paper presents a droop-based distributed control strategy for multithree-phase machines that provides augmented controllability during power sharing transients. The proposed strategy is able to mitigate the mutual interactions among different sets of windings without controlling any subspace variable, also offering a modular and redundant design. On the contrary, in a centralized configuration, subspaces would be controlled using the vector space decomposition, but fault tolerance and reliability levels required by the stricter regulations and policies expected in future transportation systems would not be satisfied. The proposed method is analytically compared against the state-of-the-art power sharing technique and equivalent models and control design procedures have been derived and considered in the comparison. Uncontrolled power sharing transients and their effects on mutual couplings among isolated sets of windings have been compared against the proposed regulated ones. Experimental results on a 22-kW nine-phase multithree-phase synchronous machine rig validate the design procedures showing good agreement with the expected performances. [ABSTRACT FROM AUTHOR]

Load-commutated inverters (LCIs), combined with wound-field synchronous machines (WFSMs), can be an excellent solution for high power drives, but their present technology suffers from important drawbacks related to low power factor, large torque pulsations, and poor starting performance. This paper presents a new LCI design intended to overcome the mentioned limitations. An SCR-based forced-commutation circuit is added to the common inverter topology to obtain a five-level waveform for the stator current. This leads to significantly reduced current harmonics and torque pulsations, in addition to bringing benefits in terms of lower additional losses. As a further advantage, the proposed design allows for a significant power factor enhancement. Finally, it enables the WFSM to be started with a much smoother torque compared to the traditional pulsed operating mode of conventional LCI drives. Simulation studies are conducted on a high-power drive scheme to show the aforementioned improvements. Also, a reduced-scale laboratory prototype of a WFSM drive system is tested to verify the feasibility of the proposed converter. [ABSTRACT FROM AUTHOR]

Direct-drive generators to be used in off-shore wind farms are very large low-speed electric machines, which pose remarkable design and manufacturing challenges. Demand forecasts for this kind of machines are urging manufacturers to work out design and technological solutions capable of facilitating series production at competitive costs. This paper presents the development of an interior permanent-magnet generator design and technology aimed at reducing series manufacturing costs while preserving good performance levels. The focus is on two of the most critical issues in the machine design and analysis, namely, the prediction and reduction of eddy-current losses in stator conductors and in permanent magnets. The proposed design solutions are validated through the construction and testing of a 780 kVA generator prototype conceived for easy scalability to higher power ratings (up to around 2.5 MVA) by core length increase. [ABSTRACT FROM AUTHOR]

Wound-field synchronous motors (WFSMs) fed by load-commutated inverters (LCIs) are widely used for high-power applications in many fields like ship propulsion, oil and gas industry, and pumped-storage hydropower generation. Several design architectures exist for LCI drives, depending on the number of LCIs and their dc-link connection as well as on the number of WFSM phase count. The prediction of LCI drive performance at steady state is important in the design stage, especially in regard to the prediction of the torque pulsations, which can give rise to serious mechanical resonance issues. This paper proposes an algebraic method to simulate the steady-state behavior of LCI drives in all their configurations of practical interest. Compared to conventional dynamic simulation approaches based on differential equation solution, the method is much more computationally efficient and requires a very limited knowledge of system parameters. Its accuracy is experimentally assessed by comparison against measurements taken on a real LCI drive arranged according to various possible schemes. Furthermore, the advantages of the proposed algebraic method over the dynamic simulations are highlighted by comparison against the simulation results on a high-power LCI-fed WFSM drive in MATLAB/Simulink environment. [ABSTRACT FROM AUTHOR]