Investigating the Effect of the Power Electronics on the Electrical Machine Performance – Zoltan Nadudvari, Development Engineer, Rolls-Royce Electrical Plc
The electrification of aircraft is playing a significant role on the path to reduce the aviation industry’s environmental footprint. The industry poses specific challenges for the development engineers due to the strict requirements and design boundary conditions. Developing an electrical drive train which meets the high safety standards, the desired efficiency and power to weight ratio requires a multidisciplinary design approach to cover all of the requirements and design aspects. In an agile environment different engineering teams often use different platforms / ecosystems (e.g. Matlab, Simcenter, Ansys) to deliver the results in time and face the challenge of real-time data exchange between the disciplines. The control method of the power electronics has a significant role in the electrical machine performance and the overall system efficiency. Our team developed a workflow and an autmated interface between Motor-CAD and MATLAB/Simscape to investigate the effect of different switching frequencies and modulation method on the electrical machine at the early phase of development.
An Automated Design Strategy for Multi-Physics and Multi-Target Optimization of E-Machines – Patricia Penabad, R&D Engineer eMotor, Mercedes-Benz AG
Efficiency is established nowadays as a key factor in the development of traction drives for electric vehicles. To increase the overall drive cycle efficiency of the electric drives allows a considerable reduction of the battery’s capacity and consequently decreases overall costs. The main challenge automotive engineers face is to develop the most cost-and energy-efficient electric drive system. For this purpose several system concepts need to be considered, which leads to different EM specifications in terms of available mounting space and performance, requiring to explore wide design space in short time. A novel metamodel-based optimisation workflow using Ansys optiSLang and Ansys Motor-CAD allows to optimize different EM topologies with focus on performance, efficiency and costs, taking into account different physical domains that interact with each other. By using this workflow is possible to rapidly change the optimization scenario and show the trade-off between several optimization objectives. The results enable data-driven decisions on EM topology (i.e. stator slots, winding turns, number of poles) and Inverter (i.e. current level) for the given system requirements. A chosen EM concept could be used for system-level evaluation and serve as a starting point for the next steps towards fine optimization.
Traction eMachine Drive Cycle Efficiency for an AWD BEV with Mixed eMachine Technology vs Dual PM eMachine – Matthew Crouch, eMachine Design Engineer, Jaguar Land Rover
This study compares the electromagnetic losses over a set of drive cycles for an AWD electric vehicle using different eMachine technologies. It highlights the efficiency advantages of using a dual permanent magnet machine for the front and rear axle versus a mixture of permanent magnet and rare-earth free technologies. This study shows how the intended driving style for the vehicle can influence the technology selection.
Skewing Strategies in Automotive Traction Machines – Leon Rodrigues, Technical Specialist Electrical Machines, Arrival
Torque ripple in interior Permanent Magnet (IPM) motors is often aggravated by the selection of the location and orientation of the magnets in the rotor lamination that are meant to benefit other targets like high reluctance torque or low magnet loss at high speed. If no design countermeasures are applied, the peak-to-peak amplitude of the torque ripple signal can reach 40% or even higher, of the rated machine torque. While this is undesirable in a wide range of applications, in traction motors is particularly detrimental since it can damage the transmission, excite vibrations modes, and create acoustic noise that can cross the admissible thresholds.
Rotor skewing is a common method to reduce torque ripple levels and reduce back-emf harmonics. Although this method is very effective at reducing the cogging torque, it can lead to uneven torque ripple reduction for different working points and the creation of a non-zero axial force. Due to mechanical construction of permanent magnet rotors skewing must be carried out in discrete steps. The number of steps, angle between consequent steps and the axial order of the rotor stacks have an impact on torque ripple, axial force and rotating direction asymmetries that require special consideration. This study analyses the skewing strategy for a typical traction motor.
First the optimum skewing angle is selected to reduce the torque ripple at a specific torque and speed operating point. The selection of the skew angle typically relates to the wavelength of the space harmonic that needs to be minimized. In motors with distributed windings, the most common approach is to cancel out the cogging harmonic by skewing by 1 slot pitch. When traction motors operate in the field weakening region, the dominant harmonic may not be the cogging harmonic and hence an alternative skewing angle may be more suitable. Motor-CAD is used to run a sweep of operating points for which the magnitude of the different torque ripple harmonics is extracted. The impact of varying the skewing angle is then explored which shows the effectiveness of skewing and the impact on average torque over the entire operating range.
The skewed sections can be arranged in a linear or V-skew arrangement which is a well-established method to reduce axial forces that exist in the linear skewing arrangement. The disadvantage of the V-skew is it requires a larger number of rotor sections which increases manufacturing cost and complexity. A compromise between the V-skew and the linear skew is a partial V skew where the orientation of the linear skew segments is re-arranged to mimic a V shape skew without increasing the number of segments. Ansys Maxwell is used to run 3D analysis for the different axial arrangements at several operating points to estimate the axial force reduction.
eMotor Development: Competences and Tools – Giuseppe Volpe, Lead e-Motor Design Engineer, Magna Powertrain GmbH & Co KG
Competencies without appropriate tools or tools without adequate competencies hampers the creation of outstanding products. Only by combining Magna Powertrain’s eMotor design expertise with Motor-CAD in the toolchain we have been able to achieve this target. Real-developed automotive platforms will be presented, with a special focus on the design approach and utilisation of Motor-CAD in our toolchain. A case study including test bench data and eMotor model calibration will be also shown.
Exporting a Thermal Model from Motor-CAD to Simulink/Simscape Including Rotational Speed and Flow Rate Dependencies – Angel Gonzalez Llacer, Simscape Electrical Development Mechatronics Modeling, MathWorks Inc
This presentation presents a newly developed method of exporting thermal models to Simulink. The exported model can be solved rapidly and also provide high accuracy as the variation of the thermal resistances due to variables such as rotation speed and flow rate can be accounted for. This new method is applied to a model for an IPMSM where the temperature of the magnet as well as the efficiency of the motor is predicted over a drive cycle.
Systematic Metamodel-Based Optimization Study of Synchronous Reluctance Machine Rotor Barrier Topologies – Branko Ban, E-machine Specialist, China Euro Vehicle Technology (CEVT)
Nowadays, due to the confidence in modeling tools and rapid product iteration, electric machine designers primarily rely on simulation. This approach reduces time and cost and is very useful when comparing different machine topologies. The prototype stage usually comes after the depletion of all simulation resources. When designing a synchronous reluctance machine, the initial step is the selection of rotor barrier type. Literature provides several topologies but does not clearly state which one yields the best performance. The goal of this paper is to differentiate the best variant for a 6-pole machine and the selected requirements, using a metamodel-based optimization approach. The novelty of the proposed strategy is in the systematic and fair comparison of different rotor topologies. The optimization process couples automated geometry generation (Matlab), electromagnetic finite element analysis (Motor-CAD), and metamodel optimization (OptiSlang). Seven rotor topologies with different complexities have been derived from circular, hyperbolic, and Zhukovsky barrier types (Circular concentric, Circular variable depth, Hyperbolic with fixed eccentricity, Hyperbolic with variable eccentricity, Original Zhukovsky, Modified Zhukovsky variable depth, and Modified Zhukovsky with equal barrier depth). A novel Modified Zhukovsky variable depth topology, which merges the best qualities of all considered variants has been developed. The same optimization strategy has been applied to all variants, and final results prove that barrier type substantially affects the final machine performance. The best results are achieved by Modified Zhukovsky variable depth topology. In relation to the worst (baseline) topology, the performance gain is 14.9% and power factor is increased from 0.61 to 0.67.
Impact of Manufacturing on E-Machine Performance and Modelling – Juliette Soulard, Associate Professor Electric Machines, University of Warwick
Design software tools and advanced FEA simulations have reduced drastically the development time of new electric machines. However, key performance is impacted by manufacturing processes. The presentation will highlight why predicting iron losses and thermal behaviour of windings are still challenging. Examples of research activities involving characterization of material properties and metrology on parts will illustrate how some calibration factors may be informed earlier than when a full motor assembly is tested on a dynamometer.
Custom Motor-CAD Thermal Lumped Parameter Network for Simulation Study of Magnetic Coil Divider – Branimir Mrak, Flanders Make
In electrical motor design and manufacturing apart from the active subcomponents of the electromagnetic circuit (stator and rotor iron, copper winding, permanent magnets), other subcomponents are needed with insulating and mechanical functionalities (coil dividers, wedges, liners, potting, shaft …). Typically these components are selected to be non-magnetic and electrically insulating in order to avoid flux leakage, dielectric breakdown or additional losses in the form of eddy currents induced in conductive surfaces. Subsequently the larger insulating mass in the motor interior makes cooling of heat generating windings more difficult i.e. is detrimental to the thermal performance of the motor.
In contrast to the current state of practice, we propose the use of coil divider that is part of the lamination stack to improve the thermal performance of the motor. The existing Motor-CAD thermal model does not represent the thermal network of the magnetic coil divider, especially when multiple cuboids are used in the simulation. We have used the MATLAB scripting plugin to quickly configure a custom-made thermal lumped parameter network, including a cuboidal representation of the magnetic coil divider. We applied the model to an existing PM motor for automotive applications and validated the lumped parameter model with 2D thermal FEA simulations. In addition, a comparative study has been made to compare the magnetic coil divider with more traditional polymer coil dividers.
The Ansys Motor Design & Simulation vision – Marius Rosu, Product Manager Electronics Business, Ansys Inc.
Electric motors are an important component of the entire vehicle market value and keeps on rising due to proliferation and sophistication. The key trends for electric vehicle traction motor like proliferation and multifunctional to higher power density and efficiency, less cooling, more electronics, robust mechanical design, are altogether becoming the pillars of Ansys vision and technology roadmap on design simulation and deployment. This talk will cover the Ansys solution and strategy that is required to enable simulation market with competitive motor design solution for designers to analysts scaling their design process through Cloud and HPC.
Three Options for NVH analysis with Motor-CAD, Maxwell and Twin Builder – Philipp Siehr, Business Development Manager, CADFEM GmbH
The NVH analysis of an electric motor is a challenging task, especially because of its multiphysical nature. In this talk, we discuss three methods to achieve NVH results with different complexity levels for all stages in the development cycle. Standalone Motor-CAD provides fast methods with adequate precision suited for the early design phase. Coupling Maxwell and Mechanical in the established harmonic workflow provides highly accurate results. And the system level approach with Twin Builder combines accuracy and speed, and allows to model transient machine behavior.
ConceptEV Launch – James Goss & John Reeve, Motor Design Ltd. & FluxSys Ltd
This presentation introduces new innovative software from Motor Design Ltd to support rapid and collaborative electrified powertrain systems engineering.
From Design to Detailed Analysis with Ansys – Anna Kvarnstrom & Akeel Auckloo, Ansys Inc.
Ansys Motor-CAD provides a comprehensive platform for any Motor Designer to size and analyse electric machines, for Electromagnetic, Thermal & Mechanical aspects. There are cases where a Motor Designer may take the design further into the Ansys workflow for additional analyses. In this talk, we will demonstrate how to automatically transfer the Motor-CAD model further into the Ansys platform for calibration, validating empirical assumptions and incorporating effects such as AC Loss calculation and water jacket thermal design optimization.
Download Agenda (0.5Mb)