With high efficiency and high flux-weakening capability, hybrid synchronous motor (HSM) is a competitive alternative as a cost-effective solution for traction applications. This paper presents a cascaded and systematic method to define specifications of target motor and design a high efficiency 172 kW, 18000 rpm HSM for a two-speed e-Powertrain system. Vehicle-level specifications and gear ratio combinations are taken into consideration to determine motor requirements. Four concept rotor typologies are investigated to obtain efficiency maps for comparisons in simulation models. Detailed loss analyses are done at both base speed and maximum speed for a selected HSM. 0.2 mm stator lamination and 4 segmented magnets in each magnet slot have been applied to further improve motor maximum efficiency. Performance tests are completed in the dyne bay to show that 97.5% maximum efficiency is achieved.
This paper deals with the comparative design and testing of synchronous reluctance machines with and without the addition of ferrite magnets for lifting applications. First, an efficient design process is proposed and showcased for both machine types. Given the application specifications of peak torque and maximum speed, the process permits quickly determining the cross-section of the motors, as well as the corresponding inverter current rating. Once the final design is defined, the optimal step-skewing angle is defined and applied in seamless computational time; for the ferrite-assisted machine, the demagnetization limit is assessed. After comparing the two designs in terms of key performance indexes and expected cost of production, the synchronous reluctance machine prototype was built and tested. The experimental tests include the measurement of the flux linkage maps and tests under load with related thermal measurements; the experimental findings are compared with the model results demonstrating an excellent match.
To predict the DC winding induced voltage of the wound field switched flux (WFSF) machine, the mathematical model of the WFSF machine considering harmonics in the DC winding self-inductance L-f, and harmonics in the mutual inductance M-fx (x=A, B, C) between x-phase armature winding and the DC winding is proposed in this paper. The modelling method decouples each current differential term in the d- and qaxis voltage equations and the DC winding terminal voltage. The control system simulation with the proposed mathematical model is built to verify the accuracy of the modelling method, and the capability to predict the DC winding induced voltage pulsation. Meanwhile, by adopting this mathematical model, finite element analysis can be avoided when conducting the motor control simulation, which can reduce the simulation time cost whilst ensuring the control precision.
In this paper, a fast evaluation of efficiency map and driving cycle efficiency of interior permanent magnet synchronous machines (IPMSMs) for electric vehicles (EVs) is proposed. Several techniques are applied to speed up the simulation process, including the iron loss and AC copper loss calculations via static and time-harmonic magnetic simulation, efficiency map buildup via surface-fit and driving cycle efficiency via finite element interpolation. Furthermore, an equivalent phase current advance angle technique is used to imitate the step-skewing simulation. As a result, compared to the commercial software, the calculation time consumption of the efficiency map and driving cycle efficiency can be reduced by 99.3% and 98.1%, respectively. It shows a great potential for the fast evaluation of the IPMSMs performance. Finally, this fast calculation method is combined with differential evolution algorithm to carry out a multi-mode and multi-objective optimization of an IPMSM. It will be of great beneficial for the determination of the optimal traction motors for EVs.
This paper expounds upon the implementation of a highly sophisticated deadbeat predictive current control (DPCC) that makes use of inductance identification to effectively regulate a low-inductance permanent magnet synchronous motor (PMSM). The principal objective of this groundbreaking control approach is to mitigate the existing current fluctuations in a manner that minimizes the computational burden, while simultaneously ensuring that additional inductance is not required. It is worth noting that the parameters inherent in the incremental prediction model display a curious pattern in which only the inductance mismatch makes a significant contribution to the prediction error, while the impact of resistance and permanent magnet (PM) flux linkage is comparatively negligible. As such, a meticulously designed inductance influence weighting factor has been devised to facilitate the assessment of the inductance error with a high degree of precision. Furthermore, the innovative DPCC with inductance identification methodology delivers substantially lower total harmonic distortion (THD) values when contrasted against the performance exhibited by the conventional FOC approach.
This paper presents an advanced torque modulation technique for motor control called dynamic motor drive that enables existing electric powertrains to achieve an extended driving range. Dynamic motor drive technology reduces low load losses by modulating torque, selectively operating the motor and power electronics intermittently at high loads, resulting in higher system efficiencies. Simulation results predict the reduction of energy consumption over the WLTP drive cycle of 2% for an electrically excited synchronous motor (EESM) and 0.4% for an interior permanent magnet (IPM) motor. This paper provides a comprehensive analysis of DMD technology, exploring its underlying principles, strategies for mitigating vibrations, and approaches to optimizing motor control for enhancing efficiency.
Induction motors usually present multiple spatial saliencies that contain useful information about the spatial magnetization and geometric state of the rotor. Each saliency modulates the transient leakage inductance producing periodical variations in the transient stator current response. A robust method to access the transient leakage inductance consists of exciting the motor with voltage pulses and computing a current slope on the measured resulting stator currents. Due to the parallel connection of the transient leakage inductances during active inverter switching, additional saliency harmonic products appear in the measured phase current responses, which have not been researched so far in literature. These saliency harmonics products can have relevant amplitudes depending on the motor construction and operating point, and possess deterministic frequencies that are related to the motor saliency frequencies. In this work, saliency harmonic products will be mathematically analysed for a star-connected induction motor for a voltage step excitation method. For a precise saliency separation, it is meaningful to compensate the saliency harmonic products. Experimental results will show the emergence of several saliency harmonic products at high loads, and will show an improvement in the rotor position estimation accuracy when such harmonic products are compensated.
This paper presents and assesses parameter estimation methods for multi-phase permanent magnet synchronous machines. More specifically, three different methods are considered, namely, a step-response method for identifying the stator resistance and absolute inductance values, a Fourier analysis-based method to obtain the harmonics in the permanent magnet flux-linkage, and a methodology to acquire the flux-linkage maps. To verify the effectiveness of the methods in question, experimental tests based on a six-phase PMSM drive are performed, and the results are compared with those acquired with finite-element-analysis simulations. Finally, to promote open science and make the findings of this work publicly available, all results are published within the open-source UltraZohm project.
As a cleaner and less expensive alternative to fossil fuel-based energy, electric energy is an inevitable choice for the future of the aviation industry. Electrical machines are one of the major components of electric aircraft, and their reliability is critical according to the standards and requirements of the aviation industry. Because of their unique properties, permanent magnet (PM) motors have become the first choice of electric aircraft manufacturers. In the present study, the reliability of PM motors and their drives is calculated using the Isograph Reliability Workbench software and the Markov chain method. Moreover, considering the failure rate, switching efficiency, power quality, and cost, the study results reveal that the dual extra switching leg three-phase PM drive configuration can currently be the optimal alternative for electric aircraft.
Permanent Magnet Synchronous Machines (PMSM), are frequently used as traction motors and powertrains. To achieve precise speed and torque control, accurate rotor speed and position information are essential, which requires the use of rotor angular position sensors. For the advantages of small size and low cost, the giant magnetoresistance (GMR) sensor is a novel rotor position detection solution. Unfortunately, the accuracy of GMR sensors highly depends on the installation and magnetic field. To investigate the mechanism between the angular position error and mechanical installation error, the installation mathematical model in 2D space is derived. The mathematical model transformed the rotating magnetic field in the sensor coordination system into a stationary field in the magnetic coordination system. Based on this model, a 3D finite element simulation model is built in Ansys/Maxwell on which the different installation errors are compared. The results show that the misaligning of the magnetic and the shaft will introduce the second-order harmonic in the angular position signal. The simulation results also validated that increasing the gap between the GMR sensor and the magnetic can reduce the error. The experimental results validated the correction of 3D modeling approach and position error mechanism.
Ensuring simplicity in design is crucial for achieving cost-effective mass production in small brushless permanent magnet motors with power ratings of up to 5 watts. In this regard, the present study proposes a novel three-phase axial flux permanent magnet motor architecture that employs a printed circuit board (PCB) in place of a conventional winding, thereby reducing the drive's complexity, cost, and component count. The use of a ferrite core and a high number of pole pairs further enhances the motor's efficiency and torque density. The paper provides a detailed account of the motor's construction and substantiates its feasibility through finite element studies conducted under both no-load and load scenarios. Additionally, critical design factors are elucidated. Overall, the findings highlight the potential of the proposed uncomplicated and economical motor topology for fan applications.
High frequency, slotless permanent magnet synchronous machines are proposed for promising aircraft electrified propulsion applications to meet carbon reduction targets. This paper presents a co-design optimization methodology in pursuit of high specific power and high efficiency. Combined analysis has been made on the motor-filter-inverter drive system performance and mass, with special emphasis on the prediction of carrier current harmonics and additional losses. The optimal design of the proposed topology can have 2-4 times higher specific power than state-of-the-art motor drive systems.
A non-intrusive method for the detection of the Rotor Inter-Turn Short Circuit and Static Eccentricity (SE) of a large hydrogenerator (74 MVA, 76 poles), by the means of real vibratory stator data, Finite Element Model (FEM) and Deep Learning technique, is presented in this paper. The Variational AutoEncoder is implemented for fault diagnosis: it is trained and validated on healthy vibratory signals. The signature of both faults is obtained from FEM, injected into another set of healthy real vibratory signals, and tested in the VAE. The classification in the latent space is investigated. The results prove the capability of the method in fault classification in a user friendly space, where each colored zone corresponds to a different case.
Because of the constant progress regarding calculation methods, materials and production technologies, new opportunities for alternative machine concepts are offered. The concept of axial flux machines is one of these alternative machine concepts and currently under investigation in many research projects. Axial flux machines are well suited for applications, in which high torque densities are required in combination with a short axial motor length. To reduce costs of powerful axial flux machines, mostly used rare earth magnets could be replaced with alternative solutions, e. g. excitation coils. There are several arrangements of rotor and stator in axial flux machines. This paper deals with the design process of a prototype machine in double stator, single rotor arrangement with aerostatic stabilization. The aerostatic stabilization primarily serves to center and stabilize the rotor disk between the two stators. One conceivable application for this machine would be mobile applications where high torque requirements have to be handled. In addition to the design of the axial flux machine, the evaluation of a centrifugal test and the assembly of the machine as well as the installation on a test bench are carried out. Furthermore, first measurement results are available, which are compared with a 3D-FEM simulation. The central objective of the research work is the development of a design and manufacturing methodology with regard to series production.
This paper presents a preliminary design and analysis of yokeless and segmented armature axial flux machines, intended to replace traditional radial flux machines in axle traction drive for battery electric vehicle applications. The focus is on torque density and the applicability of the machines in high-speed operations from an electromagnetic perspective. The approach taken in this paper is to benchmark a reference e-axle system without redesigning any other components, except for the machine itself. This allows for several design constraints to be directly adopted. To further narrow the design scope, different slot-pole combinations are compared based on their torque capability, material usage, torque density, and losses of each component, including magnet, copper, and iron losses at maximum speed. In addition, magnet segmentation is studied as a means to reduce magnet losses.
With electrification being an important route to achieve "net zero", ambitious roadmaps have been proposed to significantly improve powertrain performance in pure/hybrid electric vehicles (xEVs), at the heart of which there is the traction motor. Hairpin winding technology, as a key enabler of boosting the power density level of traction motors, is being extensively investigated in both industry and academia. Compared to random winding, hairpin winding inherently features the "design for manufacturing" characteristic to achieve a much higher slot fill factor and consistent end turn patterns for automated massive production. It is widely accepted that circulating current could be completely eliminated for hairpin windings based on the careful layout design of transposition. However, this is based on the assumption of pure uniform positioning of hairpin conductors, with manufacturing deviation neglected. This paper investigates the potential circulating current issues due to manufacturing-based misalignment of hairpin conductors in stator slots. It is interesting to find that considerable circulating current could be induced by tiny clearance reserved for the manufacturing process. The unbalanced current and additional power losses with reference to the gap between conductors and stator slots are quantitatively analysed based on the case study of a typical 150kW high-speed traction motor.
Magnetic gears perform the same function as mechanical gears but rely on noncontact operation to transfer power, producing many potential advantages over mechanical gears. This paper proposes two new topologies of transverse flux magnetic gear (TFMG). The new topologies are parametrically evaluated using 3D finite element analysis (FEA). The homopolar consequent pole TFMG offers simple assembly with only a single magnet but suffers from a low volumetric torque density (VTD) based on preliminary studies and a proof-of-concept prototype. The doubly magnetized consequent pole TFMG requires more magnets but could produce a higher VTD, but that VTD is still relatively low for a coaxial magnetic gear. A prototype of the homopolar consequent pole TFMG gear was built. Its measured slip torque agreed well with simulations, but it experienced relatively significant bearing losses. Although the VTD of this new topology is low, its simplicity with only a single permanent magnet could be advantageous for small-torque applications where manufacturing complexity can drive cost.
Electromagnetic interference (EMI) in pulse width modulated converters is exacerbated through carrier based modulation techniques that are associated with sharp spectral peaks around multiples of the switching frequency.As a mitigation measure, spread spectrum techniques are used, that distribute the carrier spectral energy over a certain frequency range to attenuate the EMI spectrum.This paper presents an alternative implementation of spread spectrum modulation through adaptive hysteresis bands, that regulate the moving average of switching frequency in a current controlled permanent magnet synchronous motor drive.Simulation results are presented through a comparison of input EMI and common mode voltage spectra for different modulation techniques.The proposed method demonstrates notable attenuation of spectral peaks when compared to carrier based modulation, while simultaneously overcoming large switching frequency deviations that are otherwise prevalent in classical hysteresis current control.
High power density electrical machines are mainly limited due to their cooling system. Oil spray cooled end windings are one possibility to significantly increase the power density. However, foundations for the estimation of the cooling capability are rather scarce in the design process. Different approaches to estimate the global heat transfer coefficients for spray cooled end windings exist in literature. However, local heat transfer coefficients, which are essential for the prediction of hot spots, can not be estimated. In this contribution, we present local heat transfer coefficient measurements for shaft spray cooled end windings. It is shown that the existence of a sump remarkably lowers the heat transfer coefficient, whereas it is almost constant over the remaining circumference. The rotational speed has a minor influence on the heat transfer. The latter is only increasing for very high speeds. A modeling approach for the Nusselt number is developed to apply it to different coolants. The achieved mean absolute percentage error between experimental and modeled Nusselt is 10.8%.
