Search results

The losses incurred in ferromagnetic materials under PWM excitations must be predicted accurately to optimize the design of modern electrical machines. The purpose of this paper is to employ mathematical hysteresis models (i.e. classical Preisach model) to predict iron losses in electrical steels under PWM excitation without compromising the computational complexity of the model.

In this paper, a novel approach based on the dynamic inverse Preisach model is proposed to model the iron losses. The PWM magnetic flux density waveform is decomposed into its harmonic component using Fourier series and a weighted Everett function is computed based on these harmonic components. The Preisach model is applied for the given flux waveform and results are validated against the measurements.

The paper predicts the total iron loss by computing a weighted Everett function based on the harmonics present in PWM waveform. Moreover, it formulates the possibility of utilizing the classical Preisach model to predict iron losses under PWM excitation.

The approach is still limited in terms of its application at high frequencies. This work may eventually lead toward the accurate prediction of iron loss under PWM excitation in electromagnetic machine design.

The paper provides a simple approach applying the Preisach model for the prediction of iron losses under PWM excitation. The proposed approach does not require additional experimental data beyond B-H loops measured under sinusoidal excitation.

A novel approach is presented to incorporate the frequency dependence into a static inverse Preisach model. The approach extends the ability of the static Preisach model to compute total iron loss under PWM excitation using a weighted Everett function.

Manufacturing processes, such as laminations, may introduce uncertainties in the magnetic properties of materials used in electrical machines. This issue, together with magnetization errors, can cause serious deterioration in the performance of the machines. Hence, stochastic material models are required for the study of the influences of the material uncertainties. The purpose of this paper is to present a methodology to study the impact of magnetization pattern uncertainties in permanent magnet electric machines.

The impacts of material uncertainties on the performances of an interior permanent magnet (IPM) machine were analyzed using two different robustness metrics (worst-case analysis and statistical study). In addition, two different robust design formulations were applied to robust multi-objective machine design problems.

The computational analyses show that material uncertainties may result in deviations of the machine performances and cause nominal solutions to become non-robust.

In this paper, the authors present stochastic models for the quantification of uncertainties in both ferromagnetic and permanent magnet materials. A robust multi-objective evolutionary algorithm is demonstrated and successfully applied to the robust design optimization of an IPM machine considering manufacturing errors and operational condition changes.

This paper aims to propose a numerical methodology to reduce the number of computations required to optimally design the rotors of synchronous reluctance machines (SynRMs) with multiple barriers.

Two objectives, average torque and torque ripple, have been simulated for thousands of SynRM models using 2D finite element analysis. Different rotor topologies (i.e. number of flux barriers) were statistically analyzed to find their respective design correlation for high average torque solutions. From this information, optimal geometrical constraints were then found to restrict the design space of multiple-barrier rotors.

Statistical analysis of two considered SynRM case studies demonstrated a design similarity between the different number of flux barriers. Upon setting the optimal geometrical constraints, it was observed that the design space of multiple-barrier rotors reduced by more than 56 per cent for both models.

Using the proposed methodology, optimal geometrical constraints of a multiple-barrier SynRM rotor can be found to restrict its corresponding design space. This approach can handle the curse of dimensionality when the number of geometric parameters increases. Also, it can potentially reduce the number of initial samples required prior to a multi-objective optimization.

Evaluating the multiphysics performance of an electric motor can be a computationally intensive process, especially where several complex subsystems of the motor are coupled together. For example, evaluating acoustic noise requires the coupling of the electromagnetic, structural and acoustic models of the electric motor. Where skewed poles are considered in the design, the problem becomes a purely three-dimensional (3D) multiphysics problem, which could increase the computational burden astronomically. This study, therefore, aims to introduce surrogate models in the design process to reduce the computational cost associated with solving such 3D-coupled multiphysics problems.

The procedure involves using the finite element (FE) method to generate a database of several skewed rotor pole surface-mounted permanent magnet synchronous motors and their corresponding electromagnetic, structural and acoustic performances. Then, a surrogate model is fitted to the data to generate mapping functions that could be used in place of the time-consuming FE simulations.

It was established that the surrogate models showed promising results in predicting the multiphysics performance of skewed pole surface-mounted permanent magnet motors. As such, such models could be used to handle the skewing aspects, which has always been a major design challenge due to the scarcity of simulation tools with stepwise skewing capability.

The main contribution involves the use of surrogate models to replace FE simulations during the design cycle of skewed pole surface-mounted permanent magnet motors without compromising the integrity of the electromagnetic, structural, and acoustic results of the motor.

The purpose of this paper is to develop surrogate models, using deep learning (DL), that can facilitate the application of EM analysis software. In the current status quo, electrical systems can be found in an ever-increasing range of products that are part of everyone’s daily live. With the advances in technology, industries such as the automotive, communications and medical devices have been disrupted with new electrical and electronic systems. The innovation and development of such systems with increasing complexity over time has been supported by the increased use of electromagnetic (EM) analysis software. Such software enables engineers to virtually design, analyze and optimize EM systems without the need for building physical prototypes, thus helping to shorten the development cycles and consequently cut costs.

The industry standard for simulating EM problems is using either the finite difference method or the finite element method (FEM). Optimization of the design process using such methods requires significant computational resources and time. With the emergence of artificial intelligence, along with specialized tools for automatic differentiation, the use of DL has become computationally much more efficient and cheaper. These advances in machine learning have ushered in a new era in EM simulations where engineers can compute results much faster while maintaining a certain level of accuracy.

This paper proposed two different models that can compute the magnetic field distribution in EM systems. The first model is based on a recurrent neural network, which is trained through a data-driven supervised learning method. The second model is an extension to the first with the incorporation of additional physics-based information to the authors’ model. Such a DL model, which is constrained by the laws of physics, is known as a physics-informed neural network. The solutions when compared with the ground truth, computed using FEM, show promising accuracy for the authors’ DL models while reducing the computation time and resources required, as compared to previous implementations in the literature.

The paper proposes a neural network architecture and is trained with two different learning methodologies, namely, supervised and physics-based. The working of the network along with the different learning methodologies is validated over several EM problems with varying levels of complexity. Furthermore, a comparative study is performed regarding performance accuracy and computational cost to establish the efficacy of different architectures and learning methodologies.

This paper aims to presents a novel application of neural network (NN) pattern recognition to ore rock sorting using inductive electromagnetic (EM) sensors.

The impedance of a metallic rock can be measured with an inductive method based on Faraday’s law and eddy current theory. A virtual rock model is then created for the simulation of the EM measurements. An NN is trained to differentiate between waste and useful ore samples (containing high amount of minerals) based on the EM sensor signals produced by the rocks.

The NN solution showed high accuracy of rock classification and produced relatively robust results from signals with noise.

A pattern recognition NN was applied to classify low- and high-grade ore samples. It has the potential to determine the approximate amount of conductive materials inside ore rocks through multiple classes. This method can be used to improve the performance of EM-based ore sorting for mineral pre-concentration.

The design of electromagnetic systems for a variety of applications such as induction heating, electrical machines, actuators and transformers requires the solution of a multi-physics problem often involving thermal, structural and mechanical coupling to the electromagnetic system. This results in a complex analysis system embedded within an optimization process. The appearance of high-performance computing systems over the past few years has made coupled simulations feasible for the design engineer. When coupled with surrogate modelling techniques, it is possible to significantly reduce the wall clock time for generating a complete design while including the impact of the multi-physics performance on the device.

An architecture is proposed for linking multiple singe physics analysis tools through the material models and a controller which schedules the execution of the various software tools. The combination of tools is implemented on a series of computational nodes operating in parallel and creating a “super node” cluster within a collection of interconnected processors.

The proposed architecture and job scheduling system can allow a parallel exploration of the design space for a device.

The originality of the work derives from the organization of the parallel computing system into a series of “super nodes” and the creation of a materials database suitable for multi-physics interactions.

This paper aims to propose a data-driven approach to determine the design guidelines for low-frequency electromagnetic devices.

Two different devices, a core-type single-phase transformer and a motor-drive system, are used to show the usefulness and generalizability of the proposed approach. Using a finite element solver, a large database of design possibilities is created by varying design parameters, i.e. the geometrical and control parameters of the systems. Design rules are then extracted by performing a statistical analysis and exploring optimal and sub-optimal designs considering various targets such as efficiency, torque ripple and power factor.

It is demonstrated that the correlation of the design parameters influences the way the data-driven approach must be made. Also, guidelines for defining new design constraints, which can lead to a more efficient optimization routine, are introduced for both case studies.

Using the proposed approach, new design guidelines, which are generally not obtainable by the classical design methods, are introduced. Also, the proposed approach can potentially deal with different parameter–objective correlations, as well as different number of connected systems. This approach is applicable regardless of the device type.