How to simulate switched reluctance machines

By Nigel Atkinson

In this blog I’ll outline how to simulate the switched reluctance machine, or SRM.

An SRM is typically driven using trapezoidal waveforms and the switch of the phases is done in sequence with the rotor position. The torque is produced by the tendency of the rotor to align with the currently excited phase (or phases). The main advantage of the SRM is its simple structure without permanent magnets, making it robust and cheap. On the other hand, the low power density and the torque ripple produced by the SRM are two potential problems that need to be addressed when designing an SRM.

Machine characterisation starts by looking at the static performance of the machine: flux linkages in the aligned and un-aligned position can be used to accurately predict the torque produced by the SRM. The SRM works for most of the time in the saturated region, so an investigation of the flux paths and saturation ‘hot-spots’ for different rotor positions is also a required step in the machine evaluation.

Next, the dynamic machine characteristics are important to determine the behaviour of the machine with different switching strategies and at different rotor speeds. Based on the conduction time for each phase and the rotor speed, the effect of varying the phase advance angles on the performance of the machine can be investigated. Since the drive waveforms used for the SRM are not sinusoidal, the model drive circuit needs to be capable of using switching patterns based on the rotor position.

Opera’s multi-physics simulation functionality has been developed extensively for over 30 years, and has a proven track record in the design of electrical machines, power systems, particle accelerators, medical devices, and a wide variety of other applications. Opera offers a range of solvers, enabling accurate robust solution of multi-physics problems in low and high-frequency electromagnetics, thermal, and stress domains. It also has a fully integrated circuit solver and optimizer. Opera’s graphical user interface is based on a scripting language, augmented with Python, which allows automation of all the steps in the modelling process, and customization of the user environment.

Built upon the scripting facilities offered by Opera is the application-specific environment designed for Electrical machines. This template-driven application is designed to offer an intuitive interface for machine designers to build and post-process their virtual prototypes. At the same time it integrates extensive technical knowledge about machine design into its setup and post-processing stages, in order to offer the user the full capabilities of the Finite Element solvers in a format that is more accessible to them.

The Opera Machines Environment offers several benefits for modelling switched reluctance machines over other finite element software due to its built-in application knowledge. The first thing you need when defining an electrical machine model is geometry. Rather than having to create it from scratch, the Environment will allow you to choose from a series of standard pre-configured geometries. If yours is non-standard then customization facilities are available.

Materials and other properties are pre-configured, then the drives will be setup and attached to the windings. The user is offered a series of standard analyses to choose from such as torque vs angle, flux linkage, inductance, cogging torque and so on. It’s not just electromagnetics; the Machines Environment will also solve multi-physics problems such as thermal distributions and structural performance.

So let’s take a quick walk-through the steps in our fully automated dialog- driven design process. First, choose your geometrical parameters such as numbers of poles and critical feature dimensions.

Full customization facilities are available should you be building a very novel type of machine. Material properties will automatically be applied to the various components, but again you can change the contents of the standard library. Drive circuits are selected, or again you can specify your own. The drive will be configured automatically to match the type of analysis you wish to perform and the machine type selected.

An example – if we choose to perform a transient electromagnetic problem, with motion at a fixed speed, then the Environment will setup the appropriate circuits and switching parameters.  And in the post­-processing stage the Environment will automatically display results such as speed, current and torque versus time or angle. It will also calculate nominal and ripple torques and save them in files for easy reference.

Depending on the assessment that you wish to carry out the Machines Environment will create the optimal number of cases to solve. Since the Environment can be used at the conceptual design stage, the user also has the choice of running a single-element thick slice of the model, rather than the full 3D model. This will speed up the analysis time enormously and allow the user to home in on initial designs before verifying the results with the full 3D model.  An example of some different model setups:
If the user asks for a torque vs angle run, then Opera solves a series of cases at different rotor positions, whereas inductance calculations would involve runs for aligned and unaligned positions at different drive currents. And you can make use of a combination of sliced and full models.

Because the Machines Environment has captured the experience of our applications team it will use the most appropriate calculations for the machine type and solution scheme. For example, when calculating losses, most software, Opera included, uses Steinmentz calculations. But they are only accurate for quasi-sinusoidal drives. So, for SRMs, with their trapezoidal drives, we automatically switch to a more accurate method. The losses from these calculations can then be automatically passed to thermal and stress solutions.  Either for the slice model, or full the full 3D model.

So, a final example of the use of Opera and the Machines Environment for multi-physics problem is the coupling between electromagnetics and stress simulation.

Here, the effect that deformations of the magnetic flux paths, due to stress in the machine, have on the forces produced in the gap can be investigated. The coupling is iterative, looking for convergence in the deformations under load.

Thank you for reading, I hope you have found the information useful. You can find more information about machine design at the link below.