A Comprehensive Framework for Iron Loss Modelling based on Experimental Characterisation of Electrical Steels

Iron losses play a pivotal role in the efficiency, thermal performance, and overall operational stability of electrical machines. Consequently, their accurate modelling is paramount for optimising design and performance. These losses, primarily categorised as hysteresis, eddy current, and excess losses, are influenced by various factors, including frequency, magnetic flux density, and material properties. Traditional analytical models often fall short in capturing the intricacies of these losses with high precision, requiring the adoption of defined methodologies that seamlessly integrate experimental data and theoretical principles. This thesis presents a comprehensive framework for iron loss modelling grounded in experimental characterisation of electrical steels using the Epstein frame. The preliminary analysis begins by estimating total iron loss from the area enclosed within B-H loops across various frequencies and flux densities. Hysteresis loss is then extracted by assuming linear scaling with frequency, using near-DC measurements as the baseline. A crossover frequency is identified, beyond which dynamic losses begin to dominate. However, further analysis reveals that eddy current losses deviate from the expected quadratic dependence on frequency and magnetic flux density, indicating the influence of inhomogeneity in the material, variations in thickness, skin-effect, coating on the laminations, etc. To address this, Bertotti’s classical loss separation model is extended with a correction factor and tuneable exponents for magnetic flux density and frequency terms. The frequency range is segmented into low, medium, and high bands to improve fitting accuracy, and nonlinear least squares fitting with data weighting is applied to extract model parameters. The model is validated using measured data for three grades of electrical steels with varying thicknesses, showing improved predictive accuracy and better insight into dominant loss mechanisms. These findings provide deeper insights into the behaviour of iron losses under varying conditions, facilitating the development in Scania’s electrification journey. The outcomes of this research contribute significantly to the field of electrical machine design by offering a systematic and validated approach for iron loss estimation. The refined modelling techniques presented in this study can aid in improving energy efficiency, reducing thermal stress, and enhancing the reliability of electrical machines, making them particularly valuable for industrial applications where precision and performance optimisation are critical.